Multiresolution image segmentation [Elektronische Ressource] / von Mohammed Abdel-Megeed M. Salem

Multiresolution Image SegmentationDISSERTATIONzur Erlangung des akademischen Gradesdoctor rerum naturalium(Dr. rer. nat.)im Fach Informatikeingereicht an derMathematisch-Naturwissenschaftlichen Fakultät IIHumboldt-Universität zu BerlinvonHerr M.Sc. Mohammed Abdel-Megeed M. Salemgeboren am 23.02.1976 in Kairo, ÄgyptenPräsident der Humboldt-Universität zu Berlin:Prof. Dr. Dr. h.c. Christoph MarkschiesDekan der Mathematisch-Naturwissenschaftlichen Fakultät II:Prof. Dr. Wolfgang CoyGutachter:1. Prof. Dr.-Ing. Beate Meffert2. Prof. Ulrich Rückert3. Prof. Dr. Mohamed F. Tolbaeingereicht am: 26. August 2008Tag der Verteidigung: 21. November 2008„Gedruckt mit Unterstützungdes Deutschen Akademischen Austauschdiensts“AbstractMore and more computer vision systems take part in the automation of var-ious applications. The main task of such systems is to automate the processof visual recognition and to extract relevant information from the images orimage sequences acquired or produced by such applications. One essentialand critical component in almost every computer vision system is image seg-mentation. The quality of the segmentation determines to a great extent thequality of the final results of the vision system.New algorithms for image and video segmentation based on the multire-solution analysis and the wavelet transform are proposed. The concept ofmultiresolution is explained as existing independently of the wavelet trans-form.