1 August 1990 Noise and object elimination from automatic correlation data by a finite element algorithm
Author Affiliations +
Proceedings Volume 1395, Close-Range Photogrammetry Meets Machine Vision; 13951W (1990) https://doi.org/10.1117/12.2294311
Event: Close-Range Photogrammetry Meets Machine Vision, 1990, Zurich, Switzerland
Abstract
The article presents the principal aspects of noise and object elimination from automatic correlation data by applying an algorithm based on the finite element theory. The algorithm developed is based on the establishment of a 3-D surface of finite elements fitted to the coordinates from the automatic correlation data by means of a least square adjustment. Three different approximations are discussed for noise and object elimination: a classical elimination by thresholding the residual errors; elimination by applying the Baarda error detection theory and a deterministic elimination using vectorized contours from an image processing.
© (1990) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Irineu da Silva, "Noise and object elimination from automatic correlation data by a finite element algorithm", Proc. SPIE 1395, Close-Range Photogrammetry Meets Machine Vision, 13951W (1 August 1990); doi: 10.1117/12.2294311; https://doi.org/10.1117/12.2294311
PROCEEDINGS
8 PAGES


SHARE
Back to Top