Segmentation of High-resolution Remotely Sensed Data - Concepts, Applications and Problems

Segmentation algorithms have already been recognized as a valuable and complementary approach that similar to human operators perform a region-based rather than a point-based evaluation of high-resolution and multi-source remotely sensed data. Goal of this paper is to summarize the state-of-the-art of respective segmentation methods by describing the underlying concepts which are rather complex in the case of processing remotely sensed data, demonstrating various applications (automatical object recognition, signalbased fusion as support for visual interpretation, and estimation of the terrain surface from Digital Surface Models), and identifying yet existing problems and further research and development needs.