103-0287-00L  Image Interpretation

SemesterAutumn Semester 2014
LecturersK. Schindler
Periodicityyearly recurring course
Language of instructionEnglish


AbstractIntroduction to interactive, semi-automatic and automatic methods for image interpretation; methodological aspects of computer-assisted remote sensing, including semantic image classification and segmentation; detection and extraction of individual objects; estimation of physical parameters.
ObjectiveUnderstanding the tasks, problems, and applications of image interpretation; basic introduction of computational methods for image-based classification and parameter estimation (clustering, classification, regression), with focus on remote sensing.
ContentImage (and point-cloud) interpretation tasks: semantic classification (e.g. land-cover mapping), physical parameter estimation (e.g. forest biomass), object extraction (e.g. roads, buildings), visual driver assistance;
Image coding and features; probabilistic inference, generative and discriminative models; clustering and segmentation; continuous parameter estimation, regression; classification and labeling; atmospheric influences in satellite remote sensing;
LiteratureJ. A. Richards: Remote Sensing Digital Image Analysis - An Introduction
C. Bishop: Pattern Recognition and Machine Learning
Prerequisites / Noticebasics of probability theory and statistics; basics of image processing; elementary programming skills (Matlab);