论文部分内容阅读
针对可见光与SAR图像灰度差异大,共有特征提取难的问题,提出了一种基于k-均值聚类分割和形态学处理的轮廓特征配准方法。利用k-均值聚类算法对两类图像进行分割,得到图像分割区域;通过形态学处理,有效减少SAR图像斑点噪声影响,准确提取两类图像的封闭轮廓;采用轮廓不变矩理论,引入矩变量距离均值、方差约束机制和一致性检查的匹配策略,获取最佳匹配对,实现了两类图像的配准。通过实验,三组图像的配准精度分别达到0.3450、0.2163和0.1810,结果表明该法可行且能达到亚像素的配准精度。
Aiming at the problem of large difference of gray level between visible light and SAR image and the difficulty of extracting common features, a contour feature registration method based on k-means clustering segmentation and morphological processing is proposed. The k-means clustering algorithm is used to segment the two types of images to obtain the image segmentation region. Morphological processing can effectively reduce the speckle noise of SAR images and accurately extract the closed contour of the two types of images. By using the moment invariant moment theory, Variable distance mean, variance constraint mechanism and consistent check matching strategy, to obtain the best matching pairs, to achieve the two types of image registration. The experiments show that the registration accuracy of the three images reaches 0.3450, 0.2163 and 0.1810, respectively. The results show that this method is feasible and can achieve sub-pixel registration accuracy.