Premium
Multiple close‐range image matching based on a self‐adaptive triangle constraint
Author(s) -
Zhu Qing,
Zhang Yunsheng,
Wu Bo,
Zhang Yeting
Publication year - 2010
Publication title -
the photogrammetric record
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.638
H-Index - 51
eISSN - 1477-9730
pISSN - 0031-868X
DOI - 10.1111/j.1477-9730.2010.00603.x
Subject(s) - computer vision , artificial intelligence , matching (statistics) , constraint (computer aided design) , computer science , range (aeronautics) , triangulation , image (mathematics) , orientation (vector space) , consistency (knowledge bases) , photogrammetry , smoothness , affine transformation , mathematics , pattern recognition (psychology) , mathematical analysis , statistics , materials science , geometry , composite material , pure mathematics
Reliable image matching is an essential and difficult task in digital photogrammetry and computer vision. Possible problems from geometric distortions, illumination changes, scale changes and difficult texture conditions will result in matching ambiguity, especially for close‐range image matching. This paper presents a multiple close‐range image matching method for surface reconstruction based on a self‐adaptive triangle constraint. This method features two aspects. First, the triangles constructed from the previously matched interest points provide strong geometric constraints for the subsequent point matching combined with gradient orientation and disparity constraints. The dynamic update of the triangulation adapts automatically to the changes of image textures. Secondly, a consistency check in object space is performed to remove possible mismatches. Using three sets of actual triple overlapped close‐range images for the experiment, the results revealed that the proposed method provides improved matching reliability.