A Practical Rough Image Curve Matching Method Based on Accumulation of Evidence
Zhao Xun · Chinese Journal of Computers · 2005
In this paper, an evidence accumulation based image curve matching method is pro posed. The method can efficiently and robustly find the rough location of a usua lly short extracted image curve to a long reference curve. The main characterist ics of the proposed method are: Firstly, experiments show that the distance betw een two curve points is more reliable than the curve length itself to be used as a matching invariant in the presence of noise, hence in the work, the distance from the start point to the end point of the extracted image curve is primarily used for the selection of matching candidates in the reference curve. Secondly, the evidence accumulation concept is introduced in the matching algorithm, which not only significantly eliminates the incorrect matching of control points, but also substantially decreases the proportion of curve segments to be verified fi nally, a time consuming process. As a result, the computational efficiency and r obustness of the proposed method is largely increased; Thirdly, the statistic Ga ussian model is introduced in the classical Hausdorff distance calculation, whic h makes it more fit for the matching of curves with significant partial deformat ions. Extensive experiments with simulated data and images of satellite and digi tal camera show that the proposed curve matching method is practical, efficient and robust.