利用改进Fisher分类器进行遥感图像变化检测  

Change detection in multitemporal remote sensing images based on local mean dynamic Fisher discriminant analysis

作　　者：辛芳芳[1] 焦李成[1] 王凌霞[1] 王桂婷[1] 

机构地区：[1]西安电子科技大学智能感知与图像理解教育部重点实验室,陕西西安710071

出　　处：《西安电子科技大学学报》2012年第5期12-17,29,共7页Journal of Xidian University

基　　金：国家自然科学基金资助项目(61173092;61072106;60972148;60971128;60970066;61003198;61001206;61077009和61050110144);高等学校学科创新引智计划(111计划)资助项目(B07048);教育部'长江学者和创新团队发展计划'资助项目(IRT1170)

摘　　要：将多时相遥感图像变化检测问题看成一个分类问题,利用改进的动态Fisher分类器通过二维联合直方图检测变化区域.考虑图像邻域关系,提出基于局部均值的动态Fisher分类器,在引入图像空间关系的同时,根据当前检测结果动态调整训练参数,解决了由于初始训练数据选取不同而造成的不稳定性.该算法不需要假设分布模型,不受差异算子的影响,且将原有的像素级检测提升为上下文相关检测.实验结果表明,该算法提高了检测精度,检测结果稳定。This paper proposes a novel change detection technique, which treats the detection problem as a classifier problem and uses the improved dynamic Fisher classifier to identify the changes in the joint intensity histogram. By considering the relationship between the pixel and its neighborhood, local mean dynamic Fisher discriminant analysis （LMDFDA） is proposed to introduce the neighborhood＇ s information. Meanwhile, the parameters of the classifier are adjusted according to the current detection result, which avoids the influences of initial conditions. The proposed method is distribution free, context-sensitive and not affected by comparison operators. Experiments show that the proposed algorithm is effective and feasible for real multi-temporal remote sensing images.

关 键 词：变化检测 非参数估计 动态Fisher分类器 均值漂移 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]