Hierarchical Scale Analysis of Images Using Phase Description
1989 (English)Licentiate thesis, monograph (Other academic)
Scale analysis and description has over the last years become one of the major research field!<! in image processing. There are two main reasons for this. A single filter has a particular limited pass band which may or may not be tuned to the different sized objects to describe. Second, size or scale is a descriptive feature in its own right. All of this requires the integration of measurements from different scales.
The thesis describes a new algorithm which detects in what scale an event appears and also in what scale it disappears. In this way the scale space is subdivided into a number of intervals. Within each scale interval a consistency check is performed to get the certainty of the detection. It will be shown that using a three-dimensional phase representation of image data, it is possible to do both the subdivision and the consistency check in a simple manner. The scale levels between different events are detected when a certain dot product becomes negative and the consistency will be a vector summation between these scales. The specific levels where a split of scale space occurs will, of course, be contextually dependent and there will also be different numbers of levels in different parts of the images. Finally an application of size description of this information will be described.
Place, publisher, year, edition, pages
Linköping, Sweden: Linköping University, Department of Electrical Engineering , 1989. , 40 p.
Linköping Studies in Science and Technology. Thesis, ISSN 0280-7971 ; 168
Engineering and Technology
IdentifiersURN: urn:nbn:se:liu:diva-53382Local ID: LIU-TEK-LIC-1989:08ISBN: 91-7870-444-8OAI: oai:DiVA.org:liu-53382DiVA: diva2:288312