Cyclone track forecasting based on satellite images using artificial neural networks
2009 (English)In: ISPRS journal of photogrammetry and remote sensing (Print), ISSN 0924-2716, Vol. 64, no 6, 513-521 p.Article in journal (Refereed) Published
Many places around the world are exposed to tropical cyclones and associated storm surges. In spite of massive efforts, a great number of people die each year as a result of cyclone events. To mitigate this damage, improved forecasting techniques must be developed. The technique presented here uses artificial neural networks to interpret NOAA-AVHRR satellite images. A multi-layer neural network, resembling the human visual system, was trained to forecast the movement of cyclones based on satellite images. The trained network produced correct directional forecast for 98% of test images, thus showing a good generalization capability. The results indicate that multi-layer neural networks could be further developed into an effective tool for cyclone track forecasting using various types of remote sensing data. Future work includes extension of the present network to handle a wide range of cyclones and to take into account supplementary information, such as wind speeds, water temperature, humidity, and air pressure.
Place, publisher, year, edition, pages
2009. Vol. 64, no 6, 513-521 p.
Cyclones; Tracking; Artificial neural networks; Hazards; Meteorology
Engineering and Technology
IdentifiersURN: urn:nbn:se:liu:diva-52892DOI: 10.1016/j.isprsjprs.2009.03.002OAI: oai:DiVA.org:liu-52892DiVA: diva2:285765