liu.seSearch for publications in DiVA
Change search
Refine search result
3456789 251 - 300 of 467
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 251.
    Markus, Nenad
    et al.
    University of Zagreb.
    Pandzic, Igor
    University of Zagreb.
    Ahlberg, Jörgen
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion2017Conference paper (Refereed)
    Abstract [en]

    Current best local descriptors are learned on a large dataset of matching and non-matching keypoint pairs. However, data of this kind is not always available since detailed keypoint correspondences can be hard to establish. On the other hand, we can often obtain labels for pairs of keypoint bags. For example, keypoint bags extracted from two images of the same object under different views form a matching pair, and keypoint bags extracted from images of different objects form a non-matching pair. On average, matching pairs should contain more corresponding keypoints than non-matching pairs. We describe an end-to-end differentiable architecture that enables the learning of local keypoint descriptors from such weakly-labeled data.

  • 252.
    Markus, Nenad
    et al.
    University of Zagreb, Croatia.
    Pandzic, Igor S.
    University of Zagreb, Croatia.
    Ahlberg, Jörgen
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion2016In: 2016 23RD INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), IEEE COMPUTER SOC , 2016, p. 2380-2385Conference paper (Refereed)
    Abstract [en]

    Current best local descriptors are learned on a large dataset of matching and non-matching keypoint pairs. However, data of this kind is not always available since detailed keypoint correspondences can be hard to establish. On the other hand, we can often obtain labels for pairs of keypoint bags. For example, keypoint bags extracted from two images of the same object under different views form a matching pair, and keypoint bags extracted from images of different objects form a non-matching pair. On average, matching pairs should contain more corresponding keypoints than non-matching pairs. We describe an end-to-end differentiable architecture that enables the learning of local keypoint descriptors from such weakly-labeled data.

  • 253.
    Markus, Nenad
    et al.
    Univ Zagreb, Croatia.
    Pandzic, Igor S.
    Univ Zagreb, Croatia.
    Ahlberg, Jörgen
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion: Applications to Face Matching, Learning From Unlabeled Videos and 3D-Shape Retrieval2019In: IEEE Transactions on Image Processing, ISSN 1057-7149, E-ISSN 1941-0042, Vol. 28, no 1, p. 279-290Article in journal (Refereed)
    Abstract [en]

    Current best local descriptors are learned on a large data set of matching and non-matching keypoint pairs. However, data of this kind are not always available, since the detailed keypoint correspondences can be hard to establish. On the other hand, we can often obtain labels for pairs of keypoint bags. For example, keypoint bags extracted from two images of the same object under different views form a matching pair, and keypoint bags extracted from images of different objects form a non-matching pair. On average, matching pairs should contain more corresponding keypoints than non-matching pairs. We describe an end-to-end differentiable architecture that enables the learning of local keypoint descriptors from such weakly labeled data. In addition, we discuss how to improve the method by incorporating the procedure of mining hard negatives. We also show how our approach can be used to learn convolutional features from unlabeled video signals and 3D models.

  • 254.
    Markuš, Nenad
    et al.
    University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia.
    Fratarcangeli, Marco
    Chalmers University of Technology, Dept. of Applied Information Technology, Göteborg, Sweden.
    Pandžić, Igor
    University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia.
    Ahlberg, Jörgen
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Fast Rendering of Image Mosaics and ASCII Art2015In: Computer graphics forum (Print), ISSN 0167-7055, E-ISSN 1467-8659, Vol. 34, no 6, p. 251-261Article in journal (Refereed)
    Abstract [en]

    An image mosaic is an assembly of a large number of small images, usually called tiles, taken from a specific dictionary/codebook. When viewed as a whole, the appearance of a single large image emerges, i.e. each tile approximates a small block of pixels. ASCII art is a related (and older) graphic design technique for producing images from printable characters. Although automatic procedures for both of these visualization schemes have been studied in the past, some are computationally heavy and cannot offer real-time and interactive performance. We propose an algorithm able to reproduce the quality of existing non-photorealistic rendering techniques, in particular ASCII art and image mosaics, obtaining large performance speed-ups. The basic idea is to partition the input image into a rectangular grid and use a decision tree to assign a tile from a pre-determined codebook to each cell. Our implementation can process video streams from webcams in real time and it is suitable for modestly equipped devices. We evaluate our technique by generating the renderings of a variety of images and videos, with good results. The source code of our engine is publicly available.

  • 255.
    Markuš, Nenad
    et al.
    University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia.
    Frljak, Miroslav
    University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia.
    Pandžić, Igor
    University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia.
    Ahlberg, Jörgen
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Forchheimer, Robert
    Linköping University, Department of Electrical Engineering, Information Coding. Linköping University, Faculty of Science & Engineering.
    High-performance face tracking2012Conference paper (Refereed)
    Abstract [en]

    Face tracking is an extensively studied field. Nevertheless, it is still a challenge to make a robust and efficient face tracker, especially on mobile devices. This extended abstract briefly describes our implementation of a high-performance multi-platform face and facial feature tracking system. The main characteristics of our approach are that the tracker is fully automatic and works with the majority of faces without any manual initialization. It is robust, resistant to rapid changes in pose and facial expressions, does not suffer from drifting and is modestly computationally expensive. The tracker runs in real-time on mobile devices.

  • 256.
    Meneghetti, Giulia
    et al.
    Linköping University, Faculty of Science & Engineering. Linköping University, Department of Electrical Engineering, Computer Vision.
    Danelljan, Martin
    Linköping University, Faculty of Science & Engineering. Linköping University, Department of Electrical Engineering, Computer Vision.
    Felsberg, Michael
    Linköping University, Faculty of Science & Engineering. Linköping University, Department of Electrical Engineering, Computer Vision.
    Nordberg, Klas
    Linköping University, Faculty of Science & Engineering. Linköping University, Department of Electrical Engineering, Computer Vision.
    Image alignment for panorama stitching in sparsely structured environments2015In: Image Analysis. SCIA 2015. / [ed] Paulsen, Rasmus R., Pedersen, Kim S., Springer, 2015, p. 428-439Conference paper (Refereed)
    Abstract [en]

    Panorama stitching of sparsely structured scenes is an open research problem. In this setting, feature-based image alignment methods often fail due to shortage of distinct image features. Instead, direct image alignment methods, such as those based on phase correlation, can be applied. In this paper we investigate correlation-based image alignment techniques for panorama stitching of sparsely structured scenes. We propose a novel image alignment approach based on discriminative correlation filters (DCF), which has recently been successfully applied to visual tracking. Two versions of the proposed DCF-based approach are evaluated on two real and one synthetic panorama dataset of sparsely structured indoor environments. All three datasets consist of images taken on a tripod rotating 360 degrees around the vertical axis through the optical center. We show that the proposed DCF-based methods outperform phase correlation-based approaches on these datasets.

  • 257.
    Miandji, Ehsan
    et al.
    Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
    Unger, Jonas
    Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
    ON NONLOCAL IMAGE COMPLETION USING AN ENSEMBLE OF DICTIONARIES2016In: 2016 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), IEEE , 2016, p. 2519-2523Conference paper (Refereed)
    Abstract [en]

    In this paper we consider the problem of nonlocal image completion from random measurements and using an ensemble of dictionaries. Utilizing recent advances in the field of compressed sensing, we derive conditions under which one can uniquely recover an incomplete image with overwhelming probability. The theoretical results are complemented by numerical simulations using various ensembles of analytical and training-based dictionaries.

  • 258.
    Mirabedini, Azadeh
    et al.
    ARC Centre of Excellence for Electromaterials Science, Intelligent Polymer Research Institute, AIIM Facility, University of Wollongong, Fairy Meadow, Australia.
    Aziz, Shazed
    ARC Centre of Excellence for Electromaterials Science, Intelligent Polymer Research Institute, AIIM Facility, University of Wollongong, Fairy Meadow, Australia.
    Spinks, Geoffrey M
    ARC Centre of Excellence for Electromaterials Science, Intelligent Polymer Research Institute, AIIM Facility, University of Wollongong, Fairy Meadow, Australia.
    Foroughi, Javad
    ARC Centre of Excellence for Electromaterials Science, Intelligent Polymer Research Institute, AIIM Facility, University of Wollongong, Fairy Meadow, Australia.
    Wet-Spun Biofiber for Torsional Artificial Muscles.2017In: Soft Robotics, ISSN 2169-5172, Vol. 4, no 4, p. 421-430Article in journal (Refereed)
    Abstract [en]

    The demands for new types of artificial muscles continue to grow and novel approaches are being enabled by the advent of new materials and novel fabrication strategies. Self-powered actuators have attracted significant attention due to their ability to be driven by elements in the ambient environment such as moisture. In this study, we demonstrate the use of twisted and coiled wet-spun hygroscopic chitosan fibers to achieve a novel torsional artificial muscle. The coiled fibers exhibited significant torsional actuation where the free end of the coiled fiber rotated up to 1155 degrees per mm of coil length when hydrated. This value is 96%, 362%, and 2210% higher than twisted graphene fiber, carbon nanotube torsional actuators, and coiled nylon muscles, respectively. A model based on a single helix was used to evaluate the torsional actuation behavior of these coiled chitosan fibers.

  • 259.
    Molin, David
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Pedestrian Detection Using Convolutional Neural Networks2015Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Pedestrian detection is an important field with applications in active safety systems for cars as well as autonomous driving. Since autonomous driving and active safety are becoming technically feasible now the interest for these applications has dramatically increased.The aim of this thesis is to investigate convolutional neural networks (CNN) for pedestrian detection. The reason for this is that CNN have recently beensuccessfully applied to several different computer vision problems. The main applications of pedestrian detection are in real time systems. For this reason,this thesis investigates strategies for reducing the computational complexity offorward propagation for CNN.The approach used in this thesis for extracting pedestrians is to use a CNN tofind a probability map of where pedestrians are located. From this probabilitymap bounding boxes for pedestrians are generated. A method for handling scale invariance for the objects of interest has also been developed in this thesis. Experiments show that using this method givessignificantly better results for the problem of pedestrian detection.The accuracy which this thesis has managed to achieve is similar to the accuracy for some other works which use CNN.

  • 260.
    Molin, Joel
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
    Foreground Segmentation of Moving Objects2010Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Foreground segmentation is a common first step in tracking and surveillance applications.  The purpose of foreground segmentation is to provide later stages of image processing with an indication of where interesting data can be found.  This thesis is an investigation of how foreground segmentation can be performed in two contexts: as a pre-step to trajectory tracking and as a pre-step in indoor surveillance applications.

    Three methods are selected and detailed: a single Gaussian method, a Gaussian mixture model method, and a codebook method.  Experiments are then performed on typical input video using the methods.  It is concluded that the Gaussian mixture model produces the output which yields the best trajectories when used as input to the trajectory tracker.  An extension is proposed to the Gaussian mixture model which reduces shadow, improving the performance of foreground segmentation in the surveillance context.

  • 261.
    Moreno, Rodrigo
    et al.
    Linköping University, Department of Biomedical Engineering. Linköping University, Center for Medical Image Science and Visualization (CMIV). Linköping University, The Institute of Technology.
    Borga, Magnus
    Linköping University, Department of Biomedical Engineering, Medical Informatics. Linköping University, Center for Medical Image Science and Visualization (CMIV). Linköping University, The Institute of Technology.
    Smedby, Örjan
    Linköping University, Center for Medical Image Science and Visualization (CMIV). Linköping University, Department of Medical and Health Sciences, Radiology. Linköping University, The Institute of Technology. Östergötlands Läns Landsting, Center for Diagnostics, Department of Radiology in Linköping.
    Soft classification of trabeculae in trabecular bone2011In: Biomedical Imaging: From Nano to Macro, 2011, IEEE , 2011, p. 1641-1644Conference paper (Refereed)
    Abstract [en]

    Classification of trabecular bone aims at discriminating different types of trabeculae. This paper proposes a method to perform a soft classification from binary 3D images. In a first step, the local structure tensor is used to estimate a membership degree of every voxel to three different classes, plate-, rod- and junction-like trabeculae. In a second step, the global structure tensor of plate-like trabeculae is compared with the local orientation of rod-like trabeculae in order to discriminate aligned from non-aligned rods. Results show that soft classification can be used for estimating independent parameters of trabecular bone for every different class, by using the classification as a weighting function.

  • 262.
    Moreno, Rodrigo
    et al.
    Linköping University, Department of Medical and Health Sciences, Radiology. Linköping University, Faculty of Health Sciences. Linköping University, Center for Medical Image Science and Visualization, CMIV.
    Garcia, Miguel Angel
    Department of Informatics Engineering, Autonomous University of Madrid, Madrid, Spain.
    Puig, Domenec
    Intelligent Robotics and Computer Vision Group at the Department of Computer Science and Mathematics, Rovira i Virgili University, Tarragona, Spain.
    Julià, Carme
    Intelligent Robotics and Computer Vision Group at the Department of Computer Science and Mathematics, Rovira i Virgili University, Tarragona, Spain.
    Edge-Preserving Color Image Denoising Through Tensor Voting2011In: Computer Vision and Image Understanding, ISSN 1077-3142, E-ISSN 1090-235X, Vol. 115, no 11, p. 1536-1551Article in journal (Refereed)
    Abstract [en]

    This paper presents a new method for edge-preserving color image denoising based on the tensor voting framework, a robust perceptual grouping technique used to extract salient information from noisy data. The tensor voting framework is adapted to encode color information through tensors in order to propagate them in a neighborhood by using a specific voting process. This voting process is specifically designed for edge-preserving color image denoising by taking into account perceptual color differences, region uniformity and edginess according to a set of intuitive perceptual criteria. Perceptual color differences are estimated by means of an optimized version of the CIEDE2000 formula, while uniformity and edginess are estimated by means of saliency maps obtained from the tensor voting process. Measurements of removed noise, edge preservation and undesirable introduced artifacts, additionally to visual inspection, show that the proposed method has a better performance than the state-of-the-art image denoising algorithms for images contaminated with CCD camera noise.

  • 263.
    Moreno, Rodrigo
    et al.
    Linköping University, Center for Medical Image Science and Visualization (CMIV). Linköping University, Department of Medical and Health Sciences, Radiology. Linköping University, Faculty of Health Sciences.
    Pizarro, Luis
    Imperial College London, Department of Computing, UK.
    Burgeth, Bernhard
    Saarland University, Faculty of Mathematics and Computer Science, Germany.
    Weickert, Joachim
    Saarland University, Faculty of Mathematics and Computer Science, Germany .
    Garcia, Miguel Angel
    Autonomous University of Madrid, Department of Electronic and Communications Technology, Spain.
    Puig, Domenec
    Rovira i Virgili University, Department of Computer Science and Mathematics, Spain.
    Adaptation of Tensor Voting to Image Structure Estimation2012In: New Developments in the Visualization and Processing of Tensor Fields / [ed] David Laidlaw and Anna Vilanova, Springer Berlin/Heidelberg, 2012, p. 29-50Chapter in book (Other academic)
    Abstract [en]

    Bringing together key researchers in disciplines ranging from visualization and image processing to applications in structural mechanics, fluid dynamics, elastography, and numerical mathematics, the workshop that generated this edited volume was the third in the successful Dagstuhl series. Its aim, reflected in the quality and relevance of the papers presented, was to foster collaboration and fresh lines of inquiry in the analysis and visualization of tensor fields, which offer a concise model for numerous physical phenomena. Despite their utility, there remains a dearth of methods for studying all but the simplest ones, a shortage the workshops aim to address.Documenting the latest progress and open research questions in tensor field analysis, the chapters reflect the excitement and inspiration generated  by this latest Dagstuhl workshop, held in July 2009. The topics they address range from applications of the analysis of tensor fields to purer research into their mathematical and analytical properties. They show how cooperation and the sharing of ideas and data between those engaged in pure and applied research can open new vistas in the study of tensor fields

  • 264.
    Moulis, Armand
    Linköping University, Department of Electrical Engineering, Computer Vision.
    Automatic Detection and Classification of Permanent and Non-Permanent Skin Marks2017Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    When forensic examiners try to identify the perpetrator of a felony, they use individual facial marks when comparing the suspect with the perpetrator. Facial marks are often used for identification and they are nowadays found manually. To speed up this process, it is desired to detect interesting facial marks automatically. This master thesis describes a method to automatically detect and separate permanent and non-permanent marks. It uses a fast radial symmetry algorithm as a core element in the mark detector. After candidate skin mark extraction, the false detections are removed depending on their size, shape and number of hair pixels. The classification of the skin marks is done with a support vector machine and the different features are examined. The results show that the facial mark detector has a good recall while the precision is poor. The elimination methods of false detection were analysed as well as the different features for the classifier. One can conclude that the color of facial marks is more relevant than the structure when classifying them into permanent and non-permanent marks.

  • 265.
    Muhammad Anwer, Rao
    et al.
    Aalto University, Finland.
    Khan, Fahad
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    van de Weijer, Joost
    University of Autonoma Barcelona, Spain.
    Laaksonen, Jorma
    Aalto University, Finland.
    Combining Holistic and Part-based Deep Representations for Computational Painting Categorization2016In: ICMR16: PROCEEDINGS OF THE 2016 ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA RETRIEVAL, ASSOC COMPUTING MACHINERY , 2016, p. 339-342Conference paper (Refereed)
    Abstract [en]

    Automatic analysis of visual art, such as paintings, is a challenging inter-disciplinary research problem. Conventional approaches only rely on global scene characteristics by encoding holistic information for computational painting categorization. We argue that such approaches are sub-optimal and that discriminative common visual structures provide complementary information for painting classification. We present an approach that encodes both the global scene layout and discriminative latent common structures for computational painting categorization. The region of interests are automatically extracted, without any manual part labeling, by training class-specific deformable part-based models. Both holistic and region-of-interests are then described using multi-scale dense convolutional features. These features are pooled separately using Fisher vector encoding and concatenated afterwards in a single image representation. Experiments are performed on a challenging dataset with 91 different painters and 13 diverse painting styles. Our approach outperforms the standard method, which only employs the global scene characteristics. Furthermore, our method achieves state-of-the-art results outperforming a recent multi-scale deep features based approach [11] by 6.4% and 3.8% respectively on artist and style classification.

  • 266.
    Muthumanickam, Prithiviraj
    et al.
    Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
    Vrotsou, Katerina
    Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
    Cooper, Matthew
    Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
    Johansson, Jimmy
    Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
    Shape Grammar Extraction for Efficient Query-by-Sketch Pattern Matching in Long Time Series2016Conference paper (Refereed)
    Abstract [en]

    Long time-series, involving thousands or even millions of time steps, are common in many application domains but remain very difficult to explore interactively. Often the analytical task in such data is to identify specific patterns, but this is a very complex and computationally difficult problem and so focusing the search in order to only identify interesting patterns is a common solution. We propose an efficient method for exploring user-sketched patterns, incorporating the domain expert’s knowledge, in time series data through a shape grammar based approach. The shape grammar is extracted from the time series by considering the data as a combination of basic elementary shapes positioned across different am- plitudes. We represent these basic shapes using a ratio value, perform binning on ratio values and apply a symbolic approximation. Our proposed method for pattern matching is amplitude-, scale- and translation-invariant and, since the pattern search and pattern con- straint relaxation happen at the symbolic level, is very efficient permitting its use in a real-time/online system. We demonstrate the effectiveness of our method in a case study on stock market data although it is applicable to any numeric time series data.

  • 267.
    Nawaz, Tahir
    et al.
    Computational Vision Group, Department of Computer Science, University of Reading.
    Berg, Amanda
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering. Termisk Systemteknik AB, Linköping, Sweden.
    Ferryman, James
    Computational Vision Group, Department of Computer Science, University of Reading.
    Ahlberg, Jörgen
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering. Termisk Systemteknik AB, Linköping, Sweden.
    Felsberg, Michael
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Effective evaluation of privacy protection techniques in visible and thermal imagery2017In: Journal of Electronic Imaging (JEI), ISSN 1017-9909, E-ISSN 1560-229X, Vol. 26, no 5, article id 051408Article in journal (Refereed)
    Abstract [en]

    Privacy protection may be defined as replacing the original content in an image region with a new (less intrusive) content having modified target appearance information to make it less recognizable by applying a privacy protection technique. Indeed the development of privacy protection techniques needs also to be complemented with an established objective evaluation method to facilitate their assessment and comparison. Generally, existing evaluation methods rely on the use of subjective judgements or assume a specific target type in image data and use target detection and recognition accuracies to assess privacy protection. This work proposes a new annotation-free evaluation method that is neither subjective nor assumes a specific target type. It assesses two key aspects of privacy protection: protection and utility. Protection is quantified as an appearance similarity and utility is measured as a structural similarity between original and privacy-protected image regions. We performed an extensive experimentation using six challenging datasets (having 12 video sequences) including a new dataset (having six sequences) that contains visible and thermal imagery. The new dataset, called TST-Priv, is made available online below for community. We demonstrate effectiveness of the proposed method by evaluating six image-based privacy protection techniques, and also show comparisons of the proposed method over existing methods.

  • 268.
    Ng, Theam Foo
    et al.
    The University of New South Wales, ADFA, Canberra, Australia.
    Pham, Tuan D
    The University of New South Wales, ADFA, Canberra, Australia.
    Sun, Changming
    CSIRO Mathematics, Informatics and Statistics, Locked Bag 17, N. Ryde, Australia.
    Automated feature weighting in fuzzy declustering-based vector quantization2010Conference paper (Refereed)
    Abstract [en]

    Feature weighting plays an important role in improving the performance of clustering technique. We propose an automated feature weighting in fuzzy declustering-based vector quantization (FDVQ), namely AFDVQ algorithm, for enhancing effectiveness and efficiency in classification. The proposed AFDVQ imposes weights on the modified fuzzy c-means (FCM) so that it can automatically calculate feature weights based on their degrees of importance rather than treating them equally. Moreover, the extension of FDVQ and AFDVQ algorithms based on generalized improved fuzzy partitions (GIFP), known as GIFP-FDVQ and GIFP-AFDVQ respectively, are proposed. The experimental results on real data (original and noisy data) and modified data (biased and noisy-biased data) have demonstrated that the proposed algorithms outperformed standard algorithms in classifying clusters especially for biased data.

  • 269.
    Ng, Theam Foo
    et al.
    School of Engineering and Information Technology, The University of New South Wales, ADFA, Canberra, ACT 2600, Australia .
    Pham, Tuan D
    School of Engineering and Information Technology, The University of New South Wales, ADFA, Canberra, ACT 2600, Australia.
    Zhou, Xiaobo
    Center for Biotechnology and Informatics, The Methodist Hospital Research Institute, Weill Cornell Medical College, Houston, TX 77030, USA .
    Justification of Fuzzy Declustering Vector Quantization Modeling in Classification of Genotype-Image Phenotypes2010Conference paper (Refereed)
    Abstract [en]

    With the fast development of multi‐dimensional data compression and pattern classification techniques, vector quantization (VQ) has become a system that allows large reduction of data storage and computational effort. One of the most recent VQ techniques that handle the poor estimation of vector centroids due to biased data from undersampling is to use fuzzy declustering‐based vector quantization (FDVQ) technique. Therefore, in this paper, we are motivated to propose a justification of FDVQ based hidden Markov model (HMM) for investigating its effectiveness and efficiency in classification of genotype‐image phenotypes. The performance evaluation and comparison of the recognition accuracy between a proposed FDVQ based HMM (FDVQ‐HMM) and a well‐known LBG (Linde, Buzo, Gray) vector quantization based HMM (LBG‐HMM) will be carried out. The experimental results show that the performances of both FDVQ‐HMM and LBG‐HMM are almost similar. Finally, we have justified the competitiveness of FDVQ‐HMM in classification of cellular phenotype image database by using hypotheses t‐test. As a result, we have validated that the FDVQ algorithm is a robust and an efficient classification technique in the application of RNAi genome‐wide screening image data.

  • 270.
    Niemi, Mikael
    Linköping University, Department of Biomedical Engineering. Linköping University, The Institute of Technology.
    Machine Learning for Rapid Image Classification2013Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    In this thesis project techniques for training a rapid image classifier that can recognize an object of a predefined type has been studied. Classifiers have been trained with the AdaBoost algorithm, with and without the use of Viola-Jones cascades. The use of Weight trimming in the classifier training has been evaluated and resulted in a significant speed up of the training, as well as improving the performance of the trained classifier. Different preprocessings of the images have also been tested, but resulted for the most part in worse performance for the classifiers when used individually. Several rectangle shaped Haar-like features including novel versions have been evaluated and the magnitude versions proved to be best at separating the image classes.

  • 271.
    Nordberg, Klas
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Introduction to Representations and Estimation in Geometry2018Other (Other academic)
    Abstract [en]

    This book contains material for an introductory course on homogeneous representations for geometry in 2 and 3 dimensions, camera projections, representations of 3D rotations, epipolar geometry, and estimation of various type of geometric objects. Based on these results, a set of applications are presented.  It also contains a toolbox of general results that are useful for the presented material.  The book is intended for undergraduate studies at advanced level in master programs, or in PhD-courses at introductory level.

  • 272.
    Nordberg, Klas
    et al.
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
    Viksten, Fredrik
    Linköping University, Department of Electrical Engineering, Information Coding. Linköping University, The Institute of Technology.
    A local geometry based descriptor for 3D data: Addendum on rank and segment extraction2010Report (Other academic)
    Abstract [en]

    This document is an addendum to the main text in A local geometry-based descriptor for 3D data applied to object pose estimation by Fredrik Viksten and Klas Nordberg. This addendum gives proofs for propositions stated in the main document. This addendum also details how to extract information from the fourth order tensor refered to as S22 in the main document.

  • 273.
    Nordgaard, Anders
    et al.
    Linköping University, Department of Computer and Information Science, Statistics. Linköping University, Faculty of Arts and Sciences.
    Höglund, Tobias
    Statens Kriminaltekniska Laboratorium.
    Assessment of Approximate Likelihood Ratios from Continuous Distributions: A Case Study of Digital Camera Identification2011In: Journal of Forensic Sciences, ISSN 0022-1198, E-ISSN 1556-4029, Vol. 56, no 2, p. 390-402Article in journal (Refereed)
    Abstract [en]

    A reported likelihood ratio for the value of evidence is very often a point estimate based on various types of reference data. When presented in court, such frequentist likelihood ratio gets a higher scientific value if it is accompanied by an error bound. This becomes particularly important when the magnitude of the likelihood ratio is modest and thus is giving less support for the forwarded proposition. Here, we investigate methods for error bound estimation for the specific case of digital camera identification. The underlying probability distributions are continuous and previously proposed models for those are used, but the derived methodology is otherwise general. Both asymptotic and resampling distributions are applied in combination with different types of point estimators. The results show that resampling is preferable for assessment based on asymptotic distributions. Further, assessment of parametric estimators is superior to evaluation of kernel estimators when background data are limited.

  • 274.
    Nordgaard, Anders
    et al.
    Linköping University, Department of Computer and Information Science, Statistics. Linköping University, Faculty of Arts and Sciences.
    Höglund, Tobias
    Swedish National Laboratory of Forensic Sciences.
    The use of likelihood ratios in digital camera identification2008In: The Seventh International Conference on Forensic Inference and Statistics, Lausanne, CH, 2008Conference paper (Other academic)
  • 275.
    Norström, Christer
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
    Underwater 3-D imaging with laser triangulation2006Independent thesis Basic level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    The objective of this master thesis was to study the performance of an active triangulation system for 3-D imaging in underwater applications. Structured light from a 20 mW laser and a conventional video camera was used to collect data for generation of 3-D images. Different techniques to locate the laser line and transform it into spatial coordinates were developed and evaluated. A field- and a laboratory trial were performed.

    From the trials we can conclude that the distance resolution is much higher than the lateral- and longitudinal resolution. The lateral resolution can be improved either by using a high frame rate camera or simply by using a low scanning speed. It is possible to obtain a range resolution of less than a millimeter. The maximum range of vision was 5 meters under water measured on a white target and 3 meters for a black target in clear sea water. These results are however dependent on environmental and system parameters such as laser power, laser beam divergence and water turbidity. A higher laser power would for example increase the maximum range.

  • 276.
    Nyberg, Adam
    Linköping University, Department of Electrical Engineering, Computer Vision.
    Transforming Thermal Images to Visible Spectrum Images Using Deep Learning2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Thermal spectrum cameras are gaining interest in many applications due to their long wavelength which allows them to operate under low light and harsh weather conditions. One disadvantage of thermal cameras is their limited visual interpretability for humans, which limits the scope of their applications. In this thesis, we try to address this problem by investigating the possibility of transforming thermal infrared (TIR) images to perceptually realistic visible spectrum (VIS) images by using Convolutional Neural Networks (CNNs). Existing state-of-the-art colorization CNNs fail to provide the desired output as they were trained to map grayscale VIS images to color VIS images. Instead, we utilize an auto-encoder architecture to perform cross-spectral transformation between TIR and VIS images. This architecture was shown to quantitatively perform very well on the problem while producing perceptually realistic images. We show that the quantitative differences are insignificant when training this architecture using different color spaces, while there exist clear qualitative differences depending on the choice of color space. Finally, we found that a CNN trained from daytime examples generalizes well on tests from night time. 

  • 277.
    Nyberg, Adam
    Linköping University, Department of Electrical Engineering. Linköping University, Faculty of Science & Engineering.
    Transforming Thermal Images to Visible Spectrum Images using Deep Learning2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Thermal spectrum cameras are gaining interest in many applications due to their long wavelength which allows them to operate under low light and harsh weather conditions. One disadvantage of thermal cameras is their limited visual interpretability for humans, which limits the scope of their applications. In this thesis, we try to address this problem by investigating the possibility of transforming thermal infrared (TIR) images to perceptually realistic visible spectrum (VIS) images by using Convolutional Neural Networks (CNNs). Existing state-of-the-art colorization CNNs fail to provide the desired output as they were trained to map grayscale VIS images to color VIS images. Instead, we utilize an auto-encoder architecture to perform cross-spectral transformation between TIR and VIS images. This architecture was shown to quantitatively perform very well on the problem while producing perceptually realistic images. We show that the quantitative differences are insignificant when training this architecture using different color spaces, while there exist clear qualitative differences depending on the choice of color space. Finally, we found that a CNN trained from day time examples generalizes well on tests from night time.

  • 278.
    Nyqvist, Hanna E.
    Linköping University, Department of Electrical Engineering, Automatic Control. Linköping University, Faculty of Science & Engineering.
    On Pose Estimation in Room-Scaled Environments2016Licentiate thesis, comprehensive summary (Other academic)
    Abstract [en]

    Pose (position and orientation) tracking in room-scaled environments is an enabling technique for many applications. Today, virtual reality (vr) and augmented reality (ar) are two examples of such applications, receiving high interest both from the public and the research community. Accurate pose tracking of the vr or ar equipment, often a camera or a headset, or of different body parts is crucial to trick the human brain and make the virtual experience realistic. Pose tracking in room-scaled environments is also needed for reference tracking and metrology. This thesis focuses on an application to metrology. In this application, photometric models of a photo studio are needed to perform realistic scene reconstruction and image synthesis. Pose tracking of a dedicated sensor enables creation of these photometric models. The demands on the tracking system used in this application is high. It must be able to provide sub-centimeter and sub-degree accuracy and at same time be easy to move and install in new photo studios.

    The focus of this thesis is to investigate and develop methods for a pose tracking system that satisfies the requirements of the intended metrology application. The Bayesian filtering framework is suggested because of its firm theoretical foundation in informatics and because it enables straightforward fusion of measurements from several sensors. Sensor fusion is in this thesis seen as a way to exploit complementary characteristics of different sensors to increase tracking accuracy and robustness. Four different types of measurements are considered; inertialmeasurements, images from a camera, range (time-of-flight) measurements from ultra wide band (uwb) radio signals, and range and velocity measurements from echoes of transmitted acoustic signals.

    A simulation study and a study of the Cramér-Rao lower filtering bound (crlb) show that an inertial-camera system has the potential to reach the required tracking accuracy. It is however assumed that known fiducial markers, that can be detected and recognized in images, are deployed in the environment. The study shows that many markers are required. This makes the solution more of a stationary solution and the mobility requirement is not fulfilled. A simultaneous localization and mapping (slam) solution, where naturally occurring features are used instead of known markers, are suggested solve this problem. Evaluation using real data shows that the provided inertial-camera slam filter suffers from drift but that support from uwb range measurements eliminates this drift. The slam solution is then only dependent on knowing the position of very few stationary uwb transmitters compared to a large number of known fiducial markers. As a last step, to increase the accuracy of the slam filter, it is investigated if and how range measurements can be complemented with velocity measurement obtained as a result of the Doppler effect. Especially, focus is put on analyzing the correlation between the range and velocity measurements and the implications this correlation has for filtering. The investigation is done in a theoretical study of reflected known signals (compare with radar and sonar) where the crlb is used as an analyzing tool. The theory is validated on real data from acoustic echoes in an indoor environment.

    List of papers
    1. A High-Performance Tracking System based on Camera and IMU
    Open this publication in new window or tab >>A High-Performance Tracking System based on Camera and IMU
    2013 (English)In: 16th International Conference on Information Fusion (FUSION), 2013, IEEE , 2013, p. 2065-2072Conference paper, Published paper (Refereed)
    Abstract [en]

    We consider an indoor tracking system consisting of an inertial measurement unit (IMU) and a camera that detects markers in the environment. There are many camera based tracking systems described in literature and available commercially, and a few of them also has support from IMU. These are based on the best-effort principle, where the performance varies depending on the situation. In contrast to this, we start with a specification of the system performance, and the design isbased on an information theoretic approach, where specific user scenarios are defined. Precise models for the camera and IMU are derived for a fusion filter, and the theoretical Cramér-Rao lower bound and the Kalman filter performance are evaluated. In this study, we focus on examining the camera quality versus the marker density needed to get at least a one mm and one degree accuracy in tracking performance.

    Place, publisher, year, edition, pages
    IEEE, 2013
    Keywords
    Tracking, IMU, Vision, Indoor, Landmarks, Cameras, Lenses, Earth, Accuracy, Optical sensors, Noise, Optical imaging
    National Category
    Control Engineering
    Identifiers
    urn:nbn:se:liu:diva-96751 (URN)000341370000274 ()9786058631113 (ISBN)9781479902842 (ISBN)9786058631113 (ISBN)
    Conference
    2013 16th International Conference on Information Fusion, Istanbul, Turkey, July 9-12, 2013
    Projects
    VPS
    Funder
    Swedish Foundation for Strategic Research
    Available from: 2013-08-26 Created: 2013-08-26 Last updated: 2016-11-22Bibliographically approved
    2. Pose Estimation Using Monocular Vision and Inertial Sensors Aided with Ultra Wide Band
    Open this publication in new window or tab >>Pose Estimation Using Monocular Vision and Inertial Sensors Aided with Ultra Wide Band
    2015 (English)In: International Conference on Indoor Positioning and Indoor Navigation (IPIN), 2015, IEEE , 2015Conference paper, Published paper (Refereed)
    Abstract [en]

    This paper presents a method for global pose estimation using inertial sensors, monocular vision, and ultra wide band (UWB) sensors. It is demonstrated that the complementary characteristics of these sensors can be exploited to provide improved global pose estimates, without requiring the introduction of any visible infrastructure, such as fiducial markers. Instead, natural landmarks are jointly estimated with the pose of the platform using a simultaneous localization and mapping framework, supported by a small number of easy-to-hide UWB beacons with known positions. The method is evaluated with data from a controlled indoor experiment with high precision ground truth. The results show the benefit of the suggested sensor combination and suggest directions for further work.

    Place, publisher, year, edition, pages
    IEEE, 2015
    Keywords
    inertial sensor (IMU), ultra wide band (UWB), monocular camera, simultaneous localization and mapping (SLAM)
    National Category
    Signal Processing Control Engineering
    Identifiers
    urn:nbn:se:liu:diva-122140 (URN)10.1109/IPIN.2015.7346940 (DOI)000379160900049 ()9781467384025 (ISBN)9781467384018 (ISBN)
    Conference
    Sixth International Conference on Indoor Positioning and Indoor Navigation, Banff, October 13-16, 2015
    Projects
    Virtual Photo Studio (VPS)
    Funder
    Swedish Foundation for Strategic Research , IIS11-0081Swedish Research CouncilSecurity Link
    Available from: 2015-10-20 Created: 2015-10-20 Last updated: 2016-11-22Bibliographically approved
    3. On Joint Range and Velocity Estimation in Detection and Ranging Sensors
    Open this publication in new window or tab >>On Joint Range and Velocity Estimation in Detection and Ranging Sensors
    2016 (English)In: Proceedings of 19th International Conference on Information Fusion (FUSION), Institute of Electrical and Electronics Engineers (IEEE), 2016, p. 1674-1681Conference paper, Published paper (Refereed)
    Abstract [en]

    Radar and sonar provide information of both range and radial velocity to unknown objects. This is accomplished by emitting a signal waveform and computing the round trip time and Doppler shift. Estimation of the round trip time and Doppler shift is usually done separately without considering the couplings between these two object related quantities. The purpose of this contribution is to first model the amplitude, time shift and time scale of the returned signal in terms of the object related states range and velocity, and analyse the Cramér-Rao lower bound (CRLB) for the joint range and velocity estimation problem. One of the conclusions is that there is negative correlation between range and velocity. The maximum likelihood (ML) cost function also confirms this strong negative correlation. For target tracking applications, the use of the correct covariance matrix for the measurement vector gives a significant gain in information, compared to using the variance of range and velocity assuming independence. In other words, the knowledge of the correlation tells the tracking filter that a too large range measurement comes most likely with a too small velocity measurement, and vice versa. Experiments with sound pulses reflected in a wall indoors confirm the main conclusion of negative correlation.

    Place, publisher, year, edition, pages
    Institute of Electrical and Electronics Engineers (IEEE), 2016
    Keywords
    Cramér-Rao lower bound (CRLB), time scale, Doppler shift, time shift, time delay, tracking
    National Category
    Control Engineering Signal Processing
    Identifiers
    urn:nbn:se:liu:diva-130476 (URN)000391273400222 ()9780996452748 (ISBN)9781509020126 (ISBN)
    Conference
    19th International Conference on Information Fusion, Heidelberg, Germany, July 5-8, 2016
    Projects
    Virtual Photo Studio (VPS)Scalable Kalman Filters
    Funder
    Swedish Foundation for Strategic Research , IIS11-0081Swedish Research Council
    Available from: 2016-08-09 Created: 2016-08-09 Last updated: 2017-02-03Bibliographically approved
  • 279.
    Nyström, Daniel
    Linköping University, Department of Science and Technology, Digital Media. Linköping University, The Institute of Technology.
    Colorimetric and Multispectral Image Acquisition2006Licentiate thesis, monograph (Other academic)
    Abstract [en]

    The trichromatic principle of representing color has for a long time been dominating in color imaging. The reason is the trichromatic nature of human color vision, but as the characteristics of typical color imaging devices are different from those of human eyes, there is a need to go beyond the trichromatic approach. The interest for multi-channel imaging, i.e. increasing the number of color channels, has made it an active research topic with a substantial potential of application.

    To achieve consistent color imaging, one needs to map the imaging-device data to the device-independent colorimetric representations CIEXYZ or CIELAB, the key concept of color management. As the color coordinates depend not only on the reflective spectrum of the object but also on the spectral properties of the illuminant, the colorimetric representation suffers from metamerism, i.e. objects of the same color under a specific illumination may appear different when they are illuminated by other light sources. Furthermore, when the sensitivities of the imaging device differ from the CIE color matching functions, two spectra that appear different for human observers may result in identical device response. On contrary, in multispectral imaging, color is represented by the object’s physical characteristics namely the spectrum which is illuminant independent. With multispectral imaging, different spectra are readily distinguishable, no matter they are metameric or not. The spectrum can then be transformed to any color space and be rendered under any illumination.

    The focus of the thesis is high quality image-acquisition in colorimetric and multispectral formats. The image acquisition system used is an experimental system with great flexibility in illumination and image acquisition setup. Besides the conventional trichromatic RGB filters, the system also provides the possibility of acquiring multi-channel images, using 7 narrowband filters. A thorough calibration and characterization of all the components involved in the image acquisition system is carried out. The spectral sensitivity of the CCD camera, which can not be derived by direct measurements, is estimated using least squares regression, optimizing the camera response to measured spectral reflectance of carefully selected color samples.

    To derive mappings to colorimetric and multispectral representations, two conceptually different approaches are used. In the model-based approach, the physical model describing the image acquisition process is inverted, to reconstruct spectral reflectance from the recorded device response. In the empirical approach, the characteristics of the individual components are ignored, and the functions are derived by relating the device response for a set of test colors to the corresponding colorimetric and spectral measurements, using linear and polynomial least squares regression.

    The results indicate that for trichromatic imaging, accurate colorimetric mappings can be derived by the empirical approach, using polynomial regression to CIEXYZ and CIELAB. Because of the media-dependency, the characterization functions should be derived for each combination of media and colorants. However, accurate spectral data reconstruction requires for multi-channel imaging, using the model-based approach. Moreover, the model-based approach is general, since it is based on the spectral characteristics of the image acquisition system, rather than the characteristics of a set of color samples.

  • 280.
    Nyström, Daniel
    Linköping University, Department of Science and Technology, Digital Media. Linköping University, The Institute of Technology.
    High Resolution Analysis of Halftone Prints: A Colorimetric and Multispectral Study2009Doctoral thesis, monograph (Other academic)
    Abstract [en]

    To reproduce color images in print, the continuous tone image is first transformed into a binary halftone image, producing various colors by discrete dots with varying area coverage. In halftone prints on paper, physical and optical dot gains generally occur, making the print look darker than expected, and making the modeling of halftone color reproduction a challenge. Most available models are based on macroscopic color measurements, averaging the reflectance over an area that is large in relation to the halftone dots. The aim of this study is to go beyond the macroscopic approach, and study halftone color reproduction on a micro-scale level, using high resolution images of halftone prints. An experimental imaging system, combining the accuracy of color measurement instruments with a high spatial resolution, opens up new possibilities to study and analyze halftone color prints.

    The experimental image acquisition offers a great flexibility in the image acquisition setup. Besides trichromatic RGB filters, the system is also equipped with a set of 7 narrowband filters, for multi-channel images. A thorough calibration and characterization of all the components in the imaging system is described. The spectral sensitivity of the CCD camera, which can not be derived by direct measurements, is estimated using least squares regression. To reconstruct spectral reflectance and colorimetric values from the device response, two conceptually different approaches are used. In the model-based characterization, the physical model describing the image acquisition process is inverted, to reconstruct spectral reflectance from the recorded device response. In the empirical characterization, the characteristics of the individual components are ignored, and the functions are derived by relating the device response for a set of test colors to the corresponding colorimetric and spectral measurements, using linear and polynomial least squares regression techniques.

    Micro-scale images, referring to images whose resolution is high in relation to the resolution of the halftone, allow for measurements of the individual halftone dots, as well as the paper between them. To capture the characteristics of large populations of halftone dots, reflectance histograms are computed as well as 3D histograms in CIEXYZ color space. The micro-scale measurements reveal that the reflectance for the halftone dots, as well as the paper between the dots, is not constant, but varies with the dot area coverage. By incorporating the varying micro-reflectance in an expanded Murray-Davies model, the nonlinearity caused by optical dot gain can be accounted for without applying the nonphysical exponentiation of the reflectance values, as in the commonly used Yule-Nielsen model.

    Due to their different intrinsic nature, physical and optical dot gains need to be treated separately when modeling the outcome of halftone prints. However, in measurements of reflection colors, physical and optical dot gains always co-exist, making the separation a difficult task. Different methods to separate the physical and optical dot gain are evaluated, using spectral reflectance measurements, transmission scans and micro-scale images. Further, the relation between the physical dot gain and the halftone dot size is investigated, demonstrated with FM halftones of various print resolutions. The physical dot gain exhibits a clear correlation with the dot size and the dot gain increase is proportional to the increase in print resolution. The experimental observations are followed by discussions and a theoretical explanation.

  • 281.
    Ochs, Matthias
    et al.
    Goethe Univ, Germany.
    Bradler, Henry
    Goethe Univ, Germany.
    Mester, Rudolf
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering. Goethe Univ, Germany.
    Learning Rank Reduced Interpolation with Principal Component Analysis2017In: 2017 28TH IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV 2017), IEEE , 2017, p. 1126-1133Conference paper (Refereed)
    Abstract [en]

    Most iterative optimization algorithms for motion, depth estimation or scene reconstruction, both sparse and dense, rely on a coarse and reliable dense initialization to bootstrap their optimization procedure. This makes techniques important that allow to obtain a dense but still approximative representation of a desired 2D structure (e.g., depth maps, optical flow, disparity maps) from a very sparse measurement of this structure. The method presented here exploits the complete information given by the principal component analysis (PCA), the principal basis and its prior distribution. The method is able to determine a dense reconstruction even if only a very sparse measurement is available. When facing such situations, typically the number of principal components is further reduced which results in a loss of expressiveness of the basis. We overcome this problem and inject prior knowledge in a maximum a posteriori (MAP) approach. We test our approach on the KITTI and the Virtual KITTI dataset and focus on the interpolation of depth maps for driving scenes. The evaluation of the results shows good agreement to the ground truth and is clearly superior to the results of an interpolation by the nearest neighbor method which disregards statistical information.

  • 282.
    Ogniewski, Jens
    et al.
    Linköping University, Department of Electrical Engineering, Information Coding. Linköping University, Faculty of Science & Engineering.
    Forssén, Per-Erik
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Pushing the Limits for View Prediction in Video Coding2017In: 12th International Conference on Computer Vision Theory and Applications (VISAPP’17), Scitepress Digital Library , 2017Conference paper (Refereed)
    Abstract [en]

    The ability to direct visual attention is a fundamental skill for seeing robots. Attention comes in two flavours: the gaze direction (overt attention) and attention to a specific part of the current field of view (covert attention), of which the latter is the focus of the present study. Specifically, we study the effects of attentional masking within pre-trained deep neural networks for the purpose of handling ambiguous scenes containing multiple objects. We investigate several variants of attentional masking on partially pre-trained deep neural networks and evaluate the effects on classification performance and sensitivity to attention mask errors in multi-object scenes. We find that a combined scheme consisting of multi-level masking and blending provides the best trade-off between classification accuracy and insensitivity to masking errors. This proposed approach is denoted multilayer continuous-valued convolutional feature masking (MC-CFM). For reasonably accurate masks it can suppress the influence of distracting objects and reach comparable classification performance to unmasked recognition in cases without distractors.

  • 283.
    Ogniewski, Jens
    et al.
    Linköping University, Department of Electrical Engineering, Information Coding. Linköping University, Faculty of Science & Engineering.
    Forssén, Per-Erik
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Pushing the Limits for View Prediction in Video Coding2017In: PROCEEDINGS OF THE 12TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (VISIGRAPP 2017), VOL 4, SCITEPRESS , 2017, p. 68-76Conference paper (Refereed)
    Abstract [en]

    More and more devices have depth sensors, making RGB+D video (colour+depth video) increasingly common. RGB+D video allows the use of depth image based rendering (DIBR) to render a given scene from different viewpoints, thus making it a useful asset in view prediction for 3D and free-viewpoint video coding. In this paper we evaluate a multitude of algorithms for scattered data interpolation, in order to optimize the performance of DIBR for video coding. This also includes novel contributions like a Kriging refinement step, an edge suppression step to suppress artifacts, and a scale-adaptive kernel. Our evaluation uses the depth extension of the Sintel datasets. Using ground-truth sequences is crucial for such an optimization, as it ensures that all errors and artifacts are caused by the prediction itself rather than noisy or erroneous data. We also present a comparison with the commonly used mesh-based projection.

  • 284.
    Ogniewski, Jens
    et al.
    Linköping University, Department of Electrical Engineering, Information Coding. Linköping University, Faculty of Science & Engineering.
    Forssén, Per-Erik
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    What is the best depth-map compression for Depth Image Based Rendering?2017In: Computer Analysis of Images and Patterns: 17th International Conference, CAIP 2017, Ystad, Sweden, August 22-24, 2017, Proceedings, Part II / [ed] Michael Felsberg, Anders Heyden and Norbert Krüger, Springer, 2017, Vol. 10425, p. 403-415Conference paper (Refereed)
    Abstract [en]

    Many of the latest smart phones and tablets come with integrated depth sensors, that make depth-maps freely available, thus enabling new forms of applications like rendering from different view points. However, efficient compression exploiting the characteristics of depth-maps as well as the requirements of these new applications is still an open issue. In this paper, we evaluate different depth-map compression algorithms, with a focus on tree-based methods and view projection as application.

    The contributions of this paper are the following: 1. extensions of existing geometric compression trees, 2. a comparison of a number of different trees, 3. a comparison of them to a state-of-the-art video coder, 4. an evaluation using ground-truth data that considers both depth-maps and predicted frames with arbitrary camera translation and rotation.

    Despite our best efforts, and contrary to earlier results, current video depth-map compression outperforms tree-based methods in most cases. The reason for this is likely that previous evaluations focused on low-quality, low-resolution depth maps, while high-resolution depth (as needed in the DIBR setting) has been ignored up until now. We also demonstrate that PSNR on depth-maps is not always a good measure of their utility.

  • 285.
    Olausson, Erik
    Linköping University, Department of Science and Technology.
    Face Recognition for Mobile Phone Applications2008Independent thesis Advanced level (degree of Magister), 20 points / 30 hpStudent thesis
    Abstract [en]

    Applying face recognition directly on a mobile phone is a challenging proposal due to the unrestrained nature of input images and limitations in memory and processor capabilities.

    A robust, fully automatic recognition system for this environment is still a far way off. However, results show that using local feature extraction and a warping scheme to reduce pose variation problems, it is possible to capitalize on high error tolerance and reach reasonable recognition rates, especially for a semi-automatic classification system where the user has the final say.

    With a gallery of 85 individuals and only one gallery image per individual available the system is able to recognize close to 60 % of the faces in a very challenging test set, while the correct individual is in the top four guesses 73% of the time.

    Adding extra reference images boosts performance to nearly 75% correct recognition and 83.5% in the top four guesses. This suggests a strategy where extra reference images are added one by one after correct classification, mimicking an online learning strategy.

  • 286.
    Olgemar, Markus
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
    Camera Based Navigation: Matching between Sensor reference and Video image2008Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    an Internal Navigational System and a Global Navigational Satellite System (GNSS). In navigational warfare the GNSS can be jammed, therefore are a third navigational system is needed. The system that has been tried in this thesis is camera based navigation. Through a video camera and a sensor reference the position is determined. This thesis will process the matching between the sensor reference and the video image.

    Two methods have been implemented: normalized cross correlation and position determination through a homography. Normalized cross correlation creates a correlation matrix. The other method uses point correspondences between the images to determine a homography between the images. And through the homography obtain a position. The more point correspondences the better the position determination will be.

    The results have been quite good. The methods have got the right position when the Euler angles of the UAV have been known. Normalized cross correlation has been the best method of the tested methods.

  • 287.
    Olofsson, Anders
    Linköping University, Department of Electrical Engineering, Information Coding.
    Modern Stereo Correspondence Algorithms: Investigation and Evaluation2010Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Many different approaches have been taken towards solving the stereo correspondence problem and great progress has been made within the field during the last decade. This is mainly thanks to newly evolved global optimization techniques and better ways to compute pixel dissimilarity between views. The most successful algorithms are based on approaches that explicitly model smoothness assumptions made about the physical world, with image segmentation and plane fitting being two frequently used techniques.

    Within the project, a survey of state of the art stereo algorithms was conducted and the theory behind them is explained. Techniques found interesting were implemented for experimental trials and an algorithm aiming to achieve state of the art performance was implemented and evaluated. For several cases, state of the art performance was reached.

    To keep down the computational complexity, an algorithm relying on local winner-take-all optimization, image segmentation and plane fitting was compared against minimizing a global energy function formulated on pixel level. Experiments show that the local approach in several cases can match the global approach, but that problems sometimes arise – especially when large areas that lack texture are present. Such problematic areas are better handled by the explicit modeling of smoothness in global energy minimization.

    Lastly, disparity estimation for image sequences was explored and some ideas on how to use temporal information were implemented and tried. The ideas mainly relied on motion detection to determine parts that are static in a sequence of frames. Stereo correspondence for sequences is a rather new research field, and there is still a lot of work to be made.

  • 288.
    Olsson, Martin
    Linköping University, Department of Science and Technology.
    Obstacle detection using stereo vision for unmanned ground vehicles2009Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    In recent years, the market for automatized surveillance and use of unmanned ground vehicles (UGVs) has increased considerably. In order for unmanned vehicles to operate autonomously, high level algorithms of artificial intelligence need to be developed and accompanied by some way to make the robots perceive and interpret the environment. The purpose of this work is to investigate methods for real-time obstacle detection using stereo vision and implement these on an existing UGV platform. To reach real-time processing speeds, the algorithms presented in this work are designed for parallel processing architectures and implemented using programmable graphics hardware. The reader will be introduced to the basics of stereo vision and given an overview of the most common real-time stereo algorithms in literature along with possible applications. A novel wide-baseline real-time depth estimation algorithm is presented. The depth estimation is used together with a simple obstacle detection algorithm, producing an occupancy map of the environment allowing for evasion of obstacles and path planning. In addition, a complete system design for autonomous navigation in multi-UGV systems is proposed.

  • 289. Olwal, Alex
    et al.
    Henrysson, Anders
    Linköping University, Department of Science and Technology, Visual Information Technology and Applications (VITA). Linköping University, The Institute of Technology.
    LUMAR: A Hybrid Spatial Display System for 2D and 3D Handheld Augmented Reality2007In: 17th International Conference on Artificial Reality and Telexistence (ICAT 2007), Esbjerg, Denmark, 2007, Los Alamitos, CA, USA: IEEE Computer Society Press , 2007, p. 63-70Conference paper (Other academic)
    Abstract [en]

    LUMAR is a hybrid system for spatial displays, allowing cell phones to be tracked in 2D and 3D through combined egocentric and exocentric techniques based on the Light-Sense and UMAR frameworks. LUMAR differs from most other spatial display systems based on mobile phones with its three-layered information space. The hybrid spatial display system consists of printed matter that is augmented with context-sensitive, dynamic 2D media when the device is on the surface, and with overlaid 3D visualizations when it is held in mid-air.

  • 290.
    Othberg, Fredrik
    Linköping University, Department of Science and Technology.
    Standardized Volume Rendering Protocols for Magnetic Resonance Imaging using Maximum-Likelihood Modeling2006Independent thesis Advanced level (degree of Magister), 20 points / 30 hpStudent thesis
    Abstract [en]

    Volume rendering (VRT) has been used with great success in studies of patients using computed tomography (CT), much because of the possibility of standardizing the rendering protocols. When using magnetic resonance imaging (MRI), this procedure is considerably more difficult, since the signal from a given tissue can vary dramatically, even for the same patient. This thesis work focuses on how to improve the presentation of MRI data by using VRT protocols including standardized transfer functions. The study is limited to exclusively examining data from patients with suspected renal artery stenosis. A total number of 11 patients are examined.

    A statistical approach is used to standardize the volume rendering protocols. The histogram of the image volume is modeled as the sum of two gamma distributions, corresponding to vessel and background voxels. Parameters describing the gamma distributions are estimated with a Maximum-likelihood technique, so that expectation (E1 and E2) and standard deviation of the two voxel distributions can be calculated from the histogram. These values are used to generate the transfer function.

    Different combinations of the values from the expectation and standard deviation were studied in a material of 11 MR angiography datasets, and the visual result was graded by a radiologist. By comparing the grades, it turned out that using only the expectation of the background distribution (E1) and vessel distribution (E2) gave the best result. The opacity is then defined with a value of 0 up to a signal threshold of E1, then increasing linearly up to 50 % at a second threshold E2, and after that a constant opacity of 50 %. The brightness curve follows the opacity curve to E2, after which it continues to increase linearly up to 100%.

    A graphical user interface was created to facilitate the user-control of the volumes and transfer functions. The result from the statistical calculations is displayed in the interface and is used to view and manipulate the transfer function directly in the volume histogram.

    A transfer function generated with the Maximum-likelihood VRT method (ML-VRT) gave a better visual result in 10 of the 11 cases than when using a transfer function not adapting to signal intensity variations.

  • 291.
    Ovrén, Hannes
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Continuous Models for Cameras and Inertial Sensors2018Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    Using images to reconstruct the world in three dimensions is a classical computer vision task. Some examples of applications where this is useful are autonomous mapping and navigation, urban planning, and special effects in movies. One common approach to 3D reconstruction is ”structure from motion” where a scene is imaged multiple times from different positions, e.g. by moving the camera. However, in a twist of irony, many structure from motion methods work best when the camera is stationary while the image is captured. This is because the motion of the camera can cause distortions in the image that lead to worse image measurements, and thus a worse reconstruction. One such distortion common to all cameras is motion blur, while another is connected to the use of an electronic rolling shutter. Instead of capturing all pixels of the image at once, a camera with a rolling shutter captures the image row by row. If the camera is moving while the image is captured the rolling shutter causes non-rigid distortions in the image that, unless handled, can severely impact the reconstruction quality.

    This thesis studies methods to robustly perform 3D reconstruction in the case of a moving camera. To do so, the proposed methods make use of an inertial measurement unit (IMU). The IMU measures the angular velocities and linear accelerations of the camera, and these can be used to estimate the trajectory of the camera over time. Knowledge of the camera motion can then be used to correct for the distortions caused by the rolling shutter. Another benefit of an IMU is that it can provide measurements also in situations when a camera can not, e.g. because of excessive motion blur, or absence of scene structure.

    To use a camera together with an IMU, the camera-IMU system must be jointly calibrated. The relationship between their respective coordinate frames need to be established, and their timings need to be synchronized. This thesis shows how to automatically perform this calibration and synchronization, without requiring e.g. calibration objects or special motion patterns.

    In standard structure from motion, the camera trajectory is modeled as discrete poses, with one pose per image. Switching instead to a formulation with a continuous-time camera trajectory provides a natural way to handle rolling shutter distortions, and also to incorporate inertial measurements. To model the continuous-time trajectory, many authors have used splines. The ability for a spline-based trajectory to model the real motion depends on the density of its spline knots. Choosing a too smooth spline results in approximation errors. This thesis proposes a method to estimate the spline approximation error, and use it to better balance camera and IMU measurements, when used in a sensor fusion framework. Also proposed is a way to automatically decide how dense the spline needs to be to achieve a good reconstruction.

    Another approach to reconstruct a 3D scene is to use a camera that directly measures depth. Some depth cameras, like the well-known Microsoft Kinect, are susceptible to the same rolling shutter effects as normal cameras. This thesis quantifies the effect of the rolling shutter distortion on 3D reconstruction, depending on the amount of motion. It is also shown that a better 3D model is obtained if the depth images are corrected using inertial measurements.

    List of papers
    1. Improving RGB-D Scene Reconstruction using Rolling Shutter Rectification
    Open this publication in new window or tab >>Improving RGB-D Scene Reconstruction using Rolling Shutter Rectification
    2015 (English)In: New Development in Robot Vision / [ed] Yu Sun, Aman Behal & Chi-Kit Ronald Chung, Springer Berlin/Heidelberg, 2015, p. 55-71Chapter in book (Refereed)
    Abstract [en]

    Scene reconstruction, i.e. the process of creating a 3D representation (mesh) of some real world scene, has recently become easier with the advent of cheap RGB-D sensors (e.g. the Microsoft Kinect).

    Many such sensors use rolling shutter cameras, which produce geometrically distorted images when they are moving. To mitigate these rolling shutter distortions we propose a method that uses an attached gyroscope to rectify the depth scans.We also present a simple scheme to calibrate the relative pose and time synchronization between the gyro and a rolling shutter RGB-D sensor.

    For scene reconstruction we use the Kinect Fusion algorithm to produce meshes. We create meshes from both raw and rectified depth scans, and these are then compared to a ground truth mesh. The types of motion we investigate are: pan, tilt and wobble (shaking) motions.

    As our method relies on gyroscope readings, the amount of computations required is negligible compared to the cost of running Kinect Fusion.

    This chapter is an extension of a paper at the IEEE Workshop on Robot Vision [10]. Compared to that paper, we have improved the rectification to also correct for lens distortion, and use a coarse-to-fine search to find the time shift more quicky.We have extended our experiments to also investigate the effects of lens distortion, and to use more accurate ground truth. The experiments demonstrate that correction of rolling shutter effects yields a larger improvement of the 3D model than correction for lens distortion.

    Place, publisher, year, edition, pages
    Springer Berlin/Heidelberg, 2015
    Series
    Cognitive Systems Monographs, ISSN 1867-4925 ; 23
    National Category
    Computer Vision and Robotics (Autonomous Systems)
    Identifiers
    urn:nbn:se:liu:diva-114344 (URN)10.1007/978-3-662-43859-6_4 (DOI)978-3-662-43858-9 (ISBN)978-3-662-43859-6 (ISBN)
    Projects
    Learnable Camera Motion Models
    Available from: 2015-02-19 Created: 2015-02-19 Last updated: 2018-06-19Bibliographically approved
    2. Gyroscope-based video stabilisation with auto-calibration
    Open this publication in new window or tab >>Gyroscope-based video stabilisation with auto-calibration
    2015 (English)In: 2015 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2015, p. 2090-2097Conference paper, Published paper (Refereed)
    Abstract [en]

    We propose a technique for joint calibration of a wide-angle rolling shutter camera (e.g. a GoPro) and an externally mounted gyroscope. The calibrated parameters are time scaling and offset, relative pose between gyroscope and camera, and gyroscope bias. The parameters are found using non-linear least squares minimisation using the symmetric transfer error as cost function. The primary contribution is methods for robust initialisation of the relative pose and time offset, which are essential for convergence. We also introduce a robust error norm to handle outliers. This results in a technique that works with general video content and does not require any specific setup or calibration patterns. We apply our method to stabilisation of videos recorded by a rolling shutter camera, with a rigidly attached gyroscope. After recording, the gyroscope and camera are jointly calibrated using the recorded video itself. The recorded video can then be stabilised using the calibrated parameters. We evaluate the technique on video sequences with varying difficulty and motion frequency content. The experiments demonstrate that our method can be used to produce high quality stabilised videos even under difficult conditions, and that the proposed initialisation is shown to end up within the basin of attraction. We also show that a residual based on the symmetric transfer error is more accurate than residuals based on the recently proposed epipolar plane normal coplanarity constraint.

    Series
    IEEE International Conference on Robotics and Automation ICRA, ISSN 1050-4729
    Keywords
    Calibration, Cameras, Cost function, Gyroscopes, Robustness, Synchronization
    National Category
    Electrical Engineering, Electronic Engineering, Information Engineering Signal Processing
    Identifiers
    urn:nbn:se:liu:diva-120182 (URN)10.1109/ICRA.2015.7139474 (DOI)000370974902014 ()978-1-4799-6922-7; 978-1-4799-6923-4 (ISBN)
    Conference
    2015 IEEE International Conference on Robotics and Automation (ICRA), Seattle, WA, USA, 26-30 May, 2015
    Projects
    LCMMVPS
    Funder
    Swedish Research Council, 2014-5928Swedish Foundation for Strategic Research , IIS11-0081
    Available from: 2015-07-13 Created: 2015-07-13 Last updated: 2018-06-19Bibliographically approved
    3. Spline Error Weighting for Robust Visual-Inertial Fusion
    Open this publication in new window or tab >>Spline Error Weighting for Robust Visual-Inertial Fusion
    2018 (English)In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, p. 321-329Conference paper, Published paper (Refereed)
    Abstract [en]

    In this paper we derive and test a probability-based weighting that can balance residuals of different types in spline fitting. In contrast to previous formulations, the proposed spline error weighting scheme also incorporates a prediction of the approximation error of the spline fit. We demonstrate the effectiveness of the prediction in a synthetic experiment, and apply it to visual-inertial fusion on rolling shutter cameras. This results in a method that can estimate 3D structure with metric scale on generic first-person videos. We also propose a quality measure for spline fitting, that can be used to automatically select the knot spacing. Experiments verify that the obtained trajectory quality corresponds well with the requested quality. Finally, by linearly scaling the weights, we show that the proposed spline error weighting minimizes the estimation errors on real sequences, in terms of scale and end-point errors.

    Series
    Computer Vision and Pattern Recognition, ISSN 1063-6919
    National Category
    Electrical Engineering, Electronic Engineering, Information Engineering
    Identifiers
    urn:nbn:se:liu:diva-149495 (URN)10.1109/CVPR.2018.00041 (DOI)000457843600034 ()978-1-5386-6420-9 (ISBN)978-1-5386-6421-6 (ISBN)
    Conference
    The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 18-22, 2018, Salt Lake City, USA
    Funder
    Swedish Research Council, 2014-5928Swedish Research Council, 2014-6227
    Available from: 2018-07-03 Created: 2018-07-03 Last updated: 2019-02-26Bibliographically approved
  • 292.
    Ovrén, Hannes
    et al.
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
    Forssén, Per-Erik
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
    Törnqvist, David
    Linköping University, Department of Electrical Engineering, Automatic Control. Linköping University, The Institute of Technology.
    Improving RGB-D Scene Reconstruction using Rolling Shutter Rectification2015In: New Development in Robot Vision / [ed] Yu Sun, Aman Behal & Chi-Kit Ronald Chung, Springer Berlin/Heidelberg, 2015, p. 55-71Chapter in book (Refereed)
    Abstract [en]

    Scene reconstruction, i.e. the process of creating a 3D representation (mesh) of some real world scene, has recently become easier with the advent of cheap RGB-D sensors (e.g. the Microsoft Kinect).

    Many such sensors use rolling shutter cameras, which produce geometrically distorted images when they are moving. To mitigate these rolling shutter distortions we propose a method that uses an attached gyroscope to rectify the depth scans.We also present a simple scheme to calibrate the relative pose and time synchronization between the gyro and a rolling shutter RGB-D sensor.

    For scene reconstruction we use the Kinect Fusion algorithm to produce meshes. We create meshes from both raw and rectified depth scans, and these are then compared to a ground truth mesh. The types of motion we investigate are: pan, tilt and wobble (shaking) motions.

    As our method relies on gyroscope readings, the amount of computations required is negligible compared to the cost of running Kinect Fusion.

    This chapter is an extension of a paper at the IEEE Workshop on Robot Vision [10]. Compared to that paper, we have improved the rectification to also correct for lens distortion, and use a coarse-to-fine search to find the time shift more quicky.We have extended our experiments to also investigate the effects of lens distortion, and to use more accurate ground truth. The experiments demonstrate that correction of rolling shutter effects yields a larger improvement of the 3D model than correction for lens distortion.

  • 293.
    Park, Sung
    et al.
    University of California, Davis, USA.
    Yu, Hongfeng
    University of California, Davis, USA.
    Hotz, Ingrid
    University of California, Davis, USA.
    Linsen, Lars
    Ernst-Moritz-Arndt-Universität Greifswald Greifswald, Germany.
    Hamann, Bernd
    University of California, Davis, USA.
    Structure-accentuating Dense Flow Visualization2006Conference paper (Refereed)
    Abstract [en]

    Vector field visualization approaches can broadly be categorized into approaches that directly visualize local orintegrated flow and approaches that analyze the topological structure and visualize extracted features. Our goal was to come up with a method that falls into the first category, yet brings out structural information. We have developed a dense flow visualization method that shows the overall flow behavior while accentuating structural information without performing a topological analysis. Our method is based on a geometry-based flow integration step and a texture-based visual exploration step. The flow integration step generates a density field, which is written into a texture. The density field is generated by tracing particles under the influence of the underlying vector field.When using a quasi-random seeding strategy for initialization, the resulting density is high in attracting regions and low in repelling regions. Density is measured by the number of particles per region accumulated over time. We generate one density field using forward and one using backward propagation. The density fields are explored using texture-based rendering techniques. We generate the two output images separately and blend the results, which allows us to distinguish between inflow and outflow regions. We obtained dense flow visualizations that display the overall flow behavior, emphasize critical and separating regions, and indicate flow direction in the neighborhood of these regions. We analyzed the results of our method for isolated first-order singularities and real data sets.

  • 294.
    Pedrosa, Joao
    et al.
    Katholieke University of Leuven, Belgium.
    Heyde, Brecht
    Katholieke University of Leuven, Belgium.
    Heeren, Laurens
    Katholieke University of Leuven, Belgium.
    Engvall, Jan
    Linköping University, Department of Medical and Health Sciences, Division of Cardiovascular Medicine. Linköping University, Faculty of Medicine and Health Sciences. Region Östergötland, Heart and Medicine Center, Department of Clinical Physiology in Linköping. Linköping University, Center for Medical Image Science and Visualization (CMIV).
    Zamorano, Jose
    Ramon and Cajal University Hospital, Spain.
    Papachristidis, Alexandros
    Kings Coll Hospital London, England.
    Edvardsen, Thor
    Oslo University Hospital, Norway; Oslo University Hospital, Norway.
    Claus, Piet
    Katholieke University of Leuven, Belgium.
    Dhooge, Jan
    Katholieke University of Leuven, Belgium.
    Automatic Short Axis Orientation of the Left Ventricle in 3D Ultrasound Recordings2016In: MEDICAL IMAGING 2016: ULTRASONIC IMAGING AND TOMOGRAPHY, SPIE-INT SOC OPTICAL ENGINEERING , 2016, Vol. 9790, no 97900EConference paper (Refereed)
    Abstract [en]

    The recent advent of three-dimensional echocardiography has led to an increased interest from the scientific community in left ventricle segmentation frameworks for cardiac volume and function assessment. An automatic orientation of the segmented left ventricular mesh is an important step to obtain a point-to-point correspondence between the mesh and the cardiac anatomy. Furthermore, this would allow for an automatic division of the left ventricle into the standard 17 segments and, thus, fully automatic per-segment analysis, e.g. regional strain assessment. In this work, a method for fully automatic short axis orientation of the segmented left ventricle is presented. The proposed framework aims at detecting the inferior right ventricular insertion point. 211 three-dimensional echocardiographic images were used to validate this framework by comparison to manual annotation of the inferior right ventricular insertion point. A mean unsigned error of 8, 05 degrees +/- 18, 50 degrees was found, whereas the mean signed error was 1, 09 degrees. Large deviations between the manual and automatic annotations (amp;gt; 30 degrees) only occurred in 3, 79% of cases. The average computation time was 666ms in a non-optimized MATLAB environment, which potentiates real-time application. In conclusion, a successful automatic real-time method for orientation of the segmented left ventricle is proposed.

  • 295.
    Pedrosa, Joao
    et al.
    Katholieke University of Leuven, Belgium.
    Queiros, Sandro
    Katholieke University of Leuven, Belgium; University of Minho, Portugal; University of Minho, Portugal.
    Bernard, Olivier
    University of Lyon 1, France.
    Engvall, Jan
    Linköping University, Department of Medical and Health Sciences, Division of Cardiovascular Medicine. Linköping University, Faculty of Medicine and Health Sciences. Region Östergötland, Heart and Medicine Center, Department of Clinical Physiology in Linköping.
    Edvardsen, Thor
    University of Oslo, Norway; Oslo University Hospital, Norway.
    Nagel, Eike
    University Hospital Frankfurt Main, Germany.
    Dhooge, Jan
    Katholieke University of Leuven, Belgium.
    Fast and Fully Automatic Left Ventricular Segmentation and Tracking in Echocardiography Using Shape-Based B-Spline Explicit Active Surfaces2017In: IEEE Transactions on Medical Imaging, ISSN 0278-0062, E-ISSN 1558-254X, Vol. 36, no 11, p. 2287-2296Article in journal (Refereed)
    Abstract [en]

    Cardiac volume/function assessment remains a critical step in daily cardiology, and 3-D ultrasound plays an increasingly important role. Fully automatic left ventricular segmentation is, however, a challenging task due to the artifacts and low contrast-to-noise ratio of ultrasound imaging. In this paper, a fast and fully automatic framework for the full-cycle endocardial left ventricle segmentation is proposed. This approach couples the advantages of the B-spline explicit active surfaces framework, a purely image information approach, to those of statistical shape models to give prior information about the expected shape for an accurate segmentation. The segmentation is propagated throughout the heart cycle using a localized anatomical affine optical flow. It is shown that this approach not only outperforms other state-of-the-art methods in terms of distance metrics with a mean average distances of 1.81 +/- 0.59 and 1.98 +/- 0.66 mm at end-diastole and end-systole, respectively, but is computationally efficient (in average 11 s per 4-D image) and fully automatic.

  • 296.
    Persson, Mikael
    et al.
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Piccini, Tommaso
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering.
    Felsberg, Michael
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering. Linköping University, Center for Medical Image Science and Visualization (CMIV).
    Mester, Rudolf
    Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering. Frankfurt University, Germany.
    Robust Stereo Visual Odometry from Monocular Techniques2015In: 2015 IEEE Intelligent Vehicles Symposium (IV), Institute of Electrical and Electronics Engineers (IEEE), 2015, p. 686-691Conference paper (Refereed)
    Abstract [en]

    Visual odometry is one of the most active topics in computer vision. The automotive industry is particularly interested in this field due to the appeal of achieving a high degree of accuracy with inexpensive sensors such as cameras. The best results on this task are currently achieved by systems based on a calibrated stereo camera rig, whereas monocular systems are generally lagging behind in terms of performance. We hypothesise that this is due to stereo visual odometry being an inherently easier problem, rather than than due to higher quality of the state of the art stereo based algorithms. Under this hypothesis, techniques developed for monocular visual odometry systems would be, in general, more refined and robust since they have to deal with an intrinsically more difficult problem. In this work we present a novel stereo visual odometry system for automotive applications based on advanced monocular techniques. We show that the generalization of these techniques to the stereo case result in a significant improvement of the robustness and accuracy of stereo based visual odometry. We support our claims by the system results on the well known KITTI benchmark, achieving the top rank for visual only systems∗ .

  • 297.
    Pettersson, Erik
    Linköping University, Department of Electrical Engineering.
    Signal- och bildbehandling på moderna grafikprocessorer2005Independent thesis Basic level (professional degree), 20 points / 30 hpStudent thesis
    Abstract [en]

    The modern graphical processing unit, GPU, is an extremely powerful unit, potentially many times more powerful than a modern microprocessor. Due to its increasing programmability it has recently become possible to use it in computation intensive applications outside its normal usage. This work investigates the possibilities and limitations of general purpose programming on GPUs. The work mainly concentrates on signal and image processing although much of the principles are applicable to other areas as well.

    A framework for image processing on GPUs is implemented and a few computer vision algorithms are implemented and evaluated, among them stereo vision and optical flow. The results show that some applications can gain a substantial speedup when implemented correctly in the GPU but others can be inefficent or extremly hard to implement.

  • 298.
    Pettersson, Johan
    Linköping University, Department of Electrical Engineering.
    Real-time Object Recognition on a GPU2007Independent thesis Advanced level (degree of Master (One Year)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Shape-Based matching (SBM) is a known method for 2D object recognition that is rather robust against illumination variations, noise, clutter and partial occlusion.

    The objects to be recognized can be translated, rotated and scaled.

    The translation of an object is determined by evaluating a similarity measure for all possible positions (similar to cross correlation).

    The similarity measure is based on dot products between normalized gradient directions in edges.

    Rotation and scale is determined by evaluating all possible combinations, spanning a huge search space.

    A resolution pyramid is used to form a heuristic for the search that then gains real-time performance.

    For SBM, a model consisting of normalized edge gradient directions, are constructed for all possible combinations of rotation and scale.

    We have avoided this by using (bilinear) interpolation in the search gradient map, which greatly reduces the amount of storage required.

    SBM is highly parallelizable by nature and with our suggested improvements it becomes much suited for running on a GPU.

    This have been implemented and tested, and the results clearly outperform those of our reference CPU implementation (with magnitudes of hundreds).

    It is also very scalable and easily benefits from future devices without effort.

    An extensive evaluation material and tools for evaluating object recognition algorithms have been developed and the implementation is evaluated and compared to two commercial 2D object recognition solutions.

    The results show that the method is very powerful when dealing with the distortions listed above and competes well with its opponents.

  • 299.
    Pham, Tuan
    Linköping University, Department of Biomedical Engineering, Division of Biomedical Engineering. Linköping University, Faculty of Science & Engineering.
    COMPLEMENTARY FEATURES FOR RADIOMIC ANALYSIS OF MALIGNANT AND BENIGN MEDIASTINAL LYMPH NODES2017In: 2017 24TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), IEEE , 2017, p. 3849-3853Conference paper (Refereed)
    Abstract [en]

    The importance of developing effective strategies for investigating mediastinal lymph-node metastases in non-small cell lung cancers is increasingly emphasized. It is because the precise detection of this metastatic disease is critical for optimal surgical intervention and treatment for patients with lung cancer. Existing medical image analysis is of limited power for mediastinal lymph-node staging on computed tomography (CT). Motivated by the radiomics hypothesis, this paper explored deep-learning, texture features and their combinations to ascertain subtle difference between malignant and benign mediastinal lymph nodes on CT. The radiomics-based results are found to be promising for differentiating malignant from benign mediastinal lymph nodes of patients with lung cancer.

  • 300.
    Pham, Tuan
    Linköping University, Department of Biomedical Engineering, Division of Biomedical Engineering. Linköping University, Faculty of Science & Engineering.
    Scaling of Texture in Training Autoencoders for Classification of Histological Images of Colorectal Cancer2017In: ADVANCES IN NEURAL NETWORKS, PT II, SPRINGER INTERNATIONAL PUBLISHING AG , 2017, Vol. 10262, p. 524-532Conference paper (Refereed)
    Abstract [en]

    Autoencoding in deep learning has been known as a useful tool for extracting image features in multiple layers, which are subsequently configured for classification by deep neural networks. A practical burden for the implementation of autoencoders is the time required for training a large number of artificial neurons. This paper shows the effects of scaling of texture in the histology of colorectal cancer, which can result in significant training time reduction being approximately to an exponential function, with improved classification rates.

3456789 251 - 300 of 467
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf