多通道置信传播算法的核线影像密集匹配

发布时间：2018-06-22 07:31

  本文选题：稠密立体匹配 + 贝叶斯模型　； 参考：《测绘科学》2017年06期

【摘要】：针对航空核线影像的稠密匹配问题,该文提出了一种基于控制点与RGB三色置信传播模型的多通道置信传播算法。该算法利用点特征匹配算子提取影像的部分同名点作为区域控制点,随后将RGB 3个颜色分量都作为置信传播模型的独立基础变量,同时增加邻域像素点对的颜色差异的贡献。两组核线影像匹配实验结果表明,所提算法在使用较少参数的情况下,降低了同名点存在颜色差异造成的干扰,利用置信传播原理提高了同名点的匹配正确率,能有效进行航空核线影像的逐像素密集匹配。
[Abstract]:A multi - channel confidence propagation algorithm based on control point and RGB three - color belief propagation model is presented in this paper . The algorithm uses the point feature matching operator to extract part of the same point as the regional control point , then the RGB 3 color components are used as the independent foundation variables of the confidence propagation model , and the color difference of the neighborhood pixel points is increased . The results show that the proposed algorithm can reduce the interference caused by the color difference in the point of the same name , and improve the matching accuracy of the same point by using the confidence propagation principle , and can effectively carry out the pixel - by - pixel intensive matching of the aerial nuclear image .
，

本文编号：2052091