Feature Matching Based on Corner and Edge Constraints
Haichao Li
DOI: https://doi.org/10.1117/2.1200705.0541
2007-01-01
SPIE Newsroom
Abstract:The principle underpinning stereo vision—comparing like but disparate images—is essential for calculating distances in applications that include 3D recognition, aircraft navigation, and motion analysis. A critical component of such calculations is matching image features, in particular corners and edges, which simplifies search space by extracting essential information from complex data. But feature matching is difficult to do quickly and accurately. One problem is that of detecting edges and their connectivity. A second problem is a lack of geometric constraints, which makes image edges hard to define in the first place. Much work has been carried out in this area over the last several years.1–4 Han and Park,1 for example, proposed a contourmatching algorithm based on contour endpoint constraints and distance measures, and epipolar constraints, which reduce the search space from the whole image to a single line. The disadvantage of this algorithm is that it is sensitive to noise disturbance. Yuan2 used a robust cooperative strategy to match contours with epipolar geometry, but the computational demands of this approach are high. Park and Han3 proposed a technique that can estimate contour motion; however, it is unreliable if the contour arc length varies or the contour has occluding parts. Corners are important 2D local image features.4 With the aid of other known methods,5 corners can be matched with 100% accuracy. The epipolar constraint, already mentioned,1–6 and the so-called continuity constraint are two such techniques that can be applied. Yet the problem of speed remains. Here, we propose a new, fast edge-matching algorithm based on corner and edge constraints that limits the search region to only a few pixels and thus improves matching time. The algorithm consists, first, in finding a set of corners7 and edges.8 Second, matched corners are established using the method reported by Zhang and colleagues for recovering unknown epipolar geometry.5 The next challenge is how to use Figure 1. Configuration map of edge matching near matched corners.