Straight Line Edge Extraction in Noisy Images
Keywords:
Filtering, Segmentation, Objects ExtractionAbstract
An advanced edge-based method of feature detection and extraction is developed for object description in digital images. It is useful for comparison of different images of the same scene in an imagery, for description and recognition of categories, for automatic buildings extraction, and for localization of mutual regions in image matching. The method includes directional filtering and search for straight edge segments in every direction and scale, taking into account the edge gradient signs. The line segments are ordered in relation to their orientation and average slopes in the processed region. These segments are used to build an object descriptor. A hierarchical set of feature descriptors is developed, taking into consideration the proposed straight line segment detector. Comparative performance is evaluated on the noisy model and in a real aerial and satellite imagery.