Segmentation of Digital Images Using a Priori Information About the Expected Image Contents

摘要

A concept of segmenting a given image is presented, which allows the use of different segmentation algorithms, using an a priori description of the expected image to guide the segmentation and interpretation process. This looks first for locations of the image where more than two regions meet. Small portions around these locations are then segmented by regional algorithms. Complete segmentation is performed only for small windows where more than two regions are expected to meet each other. In the rest of the image the boundaries can be found by line detection algorithms if start and endpoint of the line are known as edges in already segmented windows and the two adjacent regions of the line have known properties. Only pixels of interest for the expected result are inspected. Segmentation of subwindows and searching for different boundaries can be done in parallel.