摘要
成像设备、所用光谱和拍摄时间等因素的差异给红外与可见光图像匹配带来了较大的困难。考虑到边缘直线段在异源图像中的稳定性,提出一种基于线段上下文的红外与可见光图像匹配方法。首先,采用LSD(line segment detector)算法检测出图像中的直线段,接着按照几何约束规则挑选出关键直线段,并计算它们的交点,将交点与Harris角点一起组成图像特征点;通过计算特征点四象限邻域内线段的得分,得到每条线段对特征点的贡献,在此基础上采用圆形阵列的方式,构建基于线段上下文的特征描述子;最后运用双向匹配策略和RANSAC算法实现红外与可见光图像的匹配。实验结果表明,所提方法能够对灰度差异较大的红外与可见光图像实现精确匹配,并且在鲁棒性和时间效率方面都要优于主流异源图像匹配算法。
Infrared and visible images are usually captured by different sensors, in different spectra and/or at different times, which makes them difficult to match. Considering the stability of the edges, a context-based infrared and visible images matching method are presented. Firstly, LSD （line segment detector） algorithm is used to detect the line segments, then picking out key line segments according to the rules of geometric constraint, and calculating their intersections. The intersections and the Harris corners together constitute the image feature points. By calculating the score of segments in the four-quadrant neighborhood of feature points, the contribution of each segment to the feature points is computed. The context-based feature descriptors are constructed on the basis on the circular array pattern. Finally, bi-directional matching strategy and RANSAC algorithm are used to match the infrared and visible images. Experimental results show that, the proposed algorithm can achieve exact match results in infrared and visible images with great gray difference. In terms of robustness and time efficiency it is better than the classical heterologous image matching algorithms.
出处
《科学技术与工程》
北大核心
2015年第12期210-214,227,共6页
Science Technology and Engineering
基金
国家自然科学基金项目(51005229)资助
