Image matching method based on robust feature matching of advanced neighborhood topology consistency

• Main Authors: CHEN RIQING, LIU YIZHANG, YANG CHANGCAI, DAI LUANYUAN, WEI LIFANG
• Format: Patent
• Language: Chinese; English
• Published: 27.11.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention relates to an image matching method based on robust feature matching of advanced neighborhood topology consistency. The method is based on the Bayesian principle. Establishing a guide matching strategy model between the matching images A and B; and establishing a soft assignment expression between the vector length ratio and the included angle in the neighborhood of the images A andB, carrying out smoothness judgment according to neighborhood topology weighting consistency, and constructing a mathematical model for seeking the minimum cost according to the length ratio and the included angle of the vector selected in the two images and a multi-scale K neighborhood method. The image matching method is good in precision, recall and F value, so that the image matching method has good application prospect and can be applied to the fields of three-dimensional reconstruction, quick realization of splicing of unmanned aerial vehicle remote sensing images and the like. 本发明涉及一种基于先进邻域拓扑一致性的鲁棒特征匹配的图像匹配方法。该