1 August 1990 An algorithm for point positioning on digital images and in object space
Author Affiliations +
Proceedings Volume 1395, Close-Range Photogrammetry Meets Machine Vision; 139548 (1990) https://doi.org/10.1117/12.2294395
Event: Close-Range Photogrammetry Meets Machine Vision, 1990, Zurich, Switzerland
Abstract
Photogrammetric collinearity conditions are used directly for image coordinates where corresponding gray values are then differenced for least squares image matching. In this way, geometric and radiometric parameters related to forming of stereoscopic images are unified in a functional model. Matching between multiple images and si- multaneous object space positioning for points given as grid nodes has been found practicable. In terms of spatial intersection, we discover that hierarchic approaches to improving approximations can be well applied to reconstruction of a digital surface model, with image displacements being in the order of 12 pixels. At last, some ideas about further developments of our matching with .positioning algorithm are remarked. Key words: generation of digital surface models, at once matching and positioning, 2-photo versus 4-photo cases
© (1990) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
J. Wu, J. Wu, } "An algorithm for point positioning on digital images and in object space", Proc. SPIE 1395, Close-Range Photogrammetry Meets Machine Vision, 139548 (1 August 1990); doi: 10.1117/12.2294395; https://doi.org/10.1117/12.2294395
PROCEEDINGS
7 PAGES


SHARE
Back to Top