基于主曲率增强距离变换的形状相似性度量方法

摘要

The conventional shape similarity measurements face the problems of robustness and accuracy in noise interference,partial information occlusion and missing situation.A method of shape similarity measurement based on principal curvature enhancement distance transformation is proposed.The distance transformation is carried out to extend the range of the shape contour,improving the robustness of the similarity measure.The distance map is enhanced by the principal curvature of the shape contour,improving the response of rich information contours in the transformation map,and enhancing the accuracy of measurement results.Simulation experiments show that the robustness and accuracy of the method in the situation of noise interference and partial contour occlusion or absence are verified.Application experiments of road vector and GPS signal and optical remote sensing image show that the method is effective in practical application.Compared with traditional shape similarity measure methods,the method based on principal curvature enhancement distance transformation is suitable for the non-rigid target shape similarity measure,and the robustness and accuracy are better.%针对常规形状相似性度量方法在噪声干扰、部分信息遮挡和缺失情况下鲁棒性和准确性不足的问题,提出了一种基于主曲率增强距离变换的形状相似性度量方法.通过距离变换扩展形状轮廓的作用范围,提升相似性度量结果的鲁棒性;基于主曲率增强距离图,提升信息量丰富轮廓在变换图中的响应,提升度量结果的准确性.仿真对比实验验证了方法在噪声干扰以及部分轮廓遮挡或缺失情况下的鲁棒性和准确性,道路矢量与GPS信号和光学遥感图像的应用试验表明方法在实际应用中的有效性.与传统的形状相似性度量方法相比,基于主曲率增强距离变换的方法适用于非刚性目标形状相似性度量,而且鲁棒性和准确度较高.