DSSRM级联分割的SAR图像变化检测

发布时间：2018-02-04 22:09

本文关键词： 合成孔径雷达图像变化检测动态排序统计区域合并　出处：《遥感学报》2017年04期 　论文类型：期刊论文

【摘要】：SRM(Statistical Region Merging)分割算法具有快速、稳定和抗噪强的优点,基于此,本文提出一种基于DSSRM(Dynamic Sorting Statistical Region Merging)级联分割的SAR图像变化检测方法。首先,针对SRM算法基于单特征静态排序导致的过分割问题,提出一种动态排序模式的DSSRM算法以减少差异图像分割错误,该算法建立基于合并区域的多特征马氏距离排序准则,在每次合并之后更新区域邻接矩阵并重新排序;然后,基于互信息最小化准则构造多通道差异数据集以提高算法对区域合并的约束能力;最后,提出一种级联分割变化检测框架,第1级利用SRM算法将差异图像映射到超像素空间,第2级采用DSSRM算法对超像素进行动态合并获得收敛的分割结果,第3级采用简化SRM方法进行三次合并获得最终的变化检测图。实验结果表明,该方法可以获得比SRM方法和目前流行方法更好的检测性能。
[Abstract]:SRM(Statistical Region merging) segmentation algorithm has the advantages of fast, stable and strong noise resistance, based on this. In this paper, a new method based on DSSRM(Dynamic Sorting Statistical Region Merging is proposed. Cascade segmentation of SAR image change detection method. First. Aiming at the over-segmentation problem caused by SRM algorithm based on single feature static sorting, a dynamic sorting DSSRM algorithm is proposed to reduce the difference image segmentation errors. The algorithm establishes the multi-feature Markov distance ranking criterion based on the merged region, updates the region adjacency matrix after each merge and reorders. Then, the multi-channel differential data set is constructed based on the mutual information minimization criterion to improve the constraint ability of the algorithm to the region merging. Finally, a cascade segmentation change detection framework is proposed. The first level uses SRM algorithm to map the difference image to the super-pixel space. In the second stage, the DSSRM algorithm is used to dynamically merge the super-pixels to obtain the convergent segmentation results. In the third stage, the simplified SRM method is used to combine three times to obtain the final change detection diagram. The experimental results show that. This method can obtain better detection performance than SRM method and current popular method.
【作者单位】：西安电子科技大学电子工程学院;
【基金】：国家自然科学基金(编号:61571347,61201293) 中央高校基本科研业务费专项资金(编号:JB150214)~~
【分类号】：TN957.52
【正文快照】： 引用格式:张建龙,王斌.2017.DSSRM级联分割的SAR图像变化检测.遥感学报,21(4):614 621Zhang J L and Wang B.2017.SAR image change detection method of DSSRM based on cascade segmentation.随着近几年遥感技术的迅猛发展,遥感数据数量日益增长并广泛应用于环境监测、大气

【参考文献】