17 August 1994 Linear feature extraction by least squares template matching constrained by internal shape forces
Author Affiliations +
Proceedings Volume 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision; (1994) https://doi.org/10.1117/12.182860
Event: Spatial Information from Digital Photogrammetry and Computer Vision: ISPRS Commission III Symposium, 1994, Munich, Germany
Abstract
This paper deals with monoscopic object extraction from digital imagery by least squares template matching. We present a globally enforced least squares template matching method, constrained by internal shape forces, for automatic precise geometric identification and registration of object outlines. This is a highly localized operation, and as such, it can be affected by gaps in the edge representation, occlusions, radiometric noise, and other artifacts. To bypass this problem, the method is extended to an object-wise global level by being performed simultaneously for the complete outline of a single object. The template matching least squares solutions for different edge segments along the outline of an object are tied together by using a geometric coherence condition which expresses the a priori acceptable shape behavior of this object. Finally, we discuss the use of the presented technique within a semi-automated monoplotting strategy for GIS object extraction.
© (1994) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Armin Gruen, Peggy Agouris, "Linear feature extraction by least squares template matching constrained by internal shape forces", Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182860; https://doi.org/10.1117/12.182860
PROCEEDINGS
8 PAGES


SHARE
Back to Top