Title :
Feature Point Correspondence of Stereo Images by Monogenic Phase
Author_Institution :
Saskatchewan Univ., Saskatoon
Abstract :
The concept of the monogenic phase was applied to feature point matching for stereo pair images. This paper evaluated the process of feature point matching from finding feature point by the Harris corner detector, forming a correlation matrix in terms of the monogenic signal elements {p(x),q1(x).q2(x)}, then to the stage of the RANSAC inlier selection based on the fundamental matrix of epipolar geometry. The use of the monogenic phase has lead to nearly perfect matching confirmed by visual inspections.
Keywords :
computational geometry; correlation methods; edge detection; feature extraction; image matching; matrix algebra; stereo image processing; Harris corner detector; RANSAC inlier selection; correlation matrix; epipolar geometry; monogenic signal phase; stereo image feature point matching; visual inspection; Convolution; Detectors; Frequency domain analysis; Geometry; Image analysis; Inspection; Kernel; Signal processing; Signal representations; Stereo vision;
Conference_Titel :
Communications, Computers and Signal Processing, 2007. PacRim 2007. IEEE Pacific Rim Conference on
Conference_Location :
Victoria, BC
Print_ISBN :
978-1-4244-1189-4
Electronic_ISBN :
1-4244-1190-4
DOI :
10.1109/PACRIM.2007.4313227