Premium
Experiments with symmetrized intensity‐based dynamic programming algorithms for reconstructing digital terrain model
Author(s) -
Gimel'Farb G. L.,
Krot V. M.,
Grigorenko M. V.
Publication year - 1992
Publication title -
international journal of imaging systems and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.359
H-Index - 47
eISSN - 1098-1098
pISSN - 0899-9457
DOI - 10.1002/ima.1850040104
Subject(s) - epipolar geometry , computer science , terrain , computer vision , artificial intelligence , algorithm , dynamic programming , computer stereo vision , probabilistic logic , range (aeronautics) , high dynamic range , ideal (ethics) , planar , stereopsis , dynamic range , computer graphics (images) , image (mathematics) , geography , materials science , cartography , composite material , philosophy , epistemology
We describe dynamic programming algorithms that are derived on the basis of symmatrized probabilistic models of the terrain and stereo pair and reconstruct the digital terrain model from the ideal horizontal stereo pair as a bunch of planar epipolar profiles. The algorithms show good performance under nonuniform intensity distortions over both images in the stereo pair and can be used both for the long‐range and close‐range computer binocular stereo vision. Experimental results for some real stereo images of the Earth's surface are presented.