利用改进Fisher分类器进行遥感图像变化检测

doi:10.3969/j.issn.1001-2400.2012.05.003

J4 ›› 2012, Vol. 39 ›› Issue (5): 12-17+29.doi: 10.3969/j.issn.1001-2400.2012.05.003

利用改进Fisher分类器进行遥感图像变化检测

辛芳芳;焦李成;王凌霞;王桂婷

(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)

收稿日期:2011-06-28 出版日期:2012-10-20 发布日期:2012-12-13
通讯作者: 辛芳芳
作者简介:辛芳芳(1982-)，女，西安电子科技大学博士研究生，E-mail: xf9258@163.com．
基金资助:
国家自然科学基金项目;高等学校学科创新引智计划;国家部委科技项目;中央高校基本科研业务费专项资金

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

XIN Fangfang;JIAO Licheng;WANG Lingxia;WANG Guiting

(Ministry of Education Key Lab. of Intelligent Perception and Image Understanding, Xidian Univ., Xi'an 710071， China)

Received:2011-06-28 Online:2012-10-20 Published:2012-12-13
Contact: XIN Fangfang

摘要/Abstract

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．

关键词: 变化检测, 非参数估计, 动态Fisher分类器, 均值漂移

Abstract:

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.

Key words: change detection, distribution free, dynamic Fisher classifier, mean shift

辛芳芳;焦李成;王凌霞;王桂婷. 利用改进Fisher分类器进行遥感图像变化检测[J]. J4, 2012, 39(5): 12-17+29.

XIN Fangfang;JIAO Licheng;WANG Lingxia;WANG Guiting. Change detection in multitemporal remote sensing images based on local mean dynamic Fisher discriminant analysis[J]. J4, 2012, 39(5): 12-17+29.

参考文献

［1］ Bovolo F, Camps-Valls G, Bruzzone L. A Support Vector Domain Method for Change Detection in Multitemporal Images［J］. Pattern Recognition Letters, 2010, 31(10): 1148-1154.
［2］ Bazi Y, Bruzzone L, Melgani F. An Unsupervised Approach Based on the Generalized Gaussian Model to Automatic Change Detection in Multitemporal SAR Images［J］. IEEE Trans on Geosci Remote Sens, 2005, 43(4): 874-887.
［3］ Bovolo F, Bruzzone L. A Theoretical Framework for Unsupervised Change Detection Based on Change Vector Analysis in Polar Domain［J］. IEEE Trans on Geosci Remote Sens, 2007, 45(3): 778-789.
［4］ Kita Y. A Study of Change Detection from Satellite Images Using Joint Intensity Histogram［C］//19th International Conference on Digital Object Identifier. Tampa: Pattem Recodnition, 2008: 1-4.
［5］ Sugiyama M. Dimensionality Reduction of Multimodal Labeled Data by Local Fisher Discriminant Analysis［J］. Machine Learning Research, 2007(8): 1027-1061.
［6］魏莱, 王守觉, 徐菲菲. 近邻边界Fisher判别分析［J］. 电子与信息学报, 2009, 31(3): 509-513.
Wei Lai, Wang Shoujue, Xu Feifei. Neighborhood Margin Fisher Discriminant Analysis［J］. Journal of Electronics ＆ Information Technology, 2009, 31(3): 509-513.
［7］辛芳芳, 焦李成, 王桂婷, 等. 基于小波域Fisher分类器的SAR图像变化检测［J］. 红外与毫米波学报, 2011, 30(2): 173-178.
Xin Fangfang, Jiao Licheng, Wang Guiting. SAR Images Change Detection Based on Wavelet Domain Fisher Classifier［J］. Infrared Millim Waves, 2011, 30(2): 173-178.
［8］ Comaniciu D, Meer P. Mean Shift: A Robust Approach toward Feature Space Analysis［J］. IEEE Trans on Pattern Anal Mach Intell, 2002, 24(5): 603-619.
［9］王桂婷, 王幼亮, 焦李成. 自适应空间邻域分析和瑞利-高斯分布的多时相遥感影像变化检测［J］. 遥感学报, 2009, 13(4): 639-646.
Wang Guiting, Wang Youliang, Jiao Licheng. Adaptive Spatial Neighborhood Analysis and Rayleigh-Gaussian Distribution Fitting for Change Detection in Multi-temporal Remote Sensing Images［J］. Journal of Remote Sensing, 2009, 13(4): 639-646.
［10］ Francesca B, Lorenzo B. A Detail-Preserving Scalse-Driven Approach to Change Detection in Multitemporal SAR Images［J］. IEEE Trans on Geosci Remote Sens, 2005, 43(12): 2963-2972.

[1]	刘志强,张铭津,池源,李云松. 一种深度学习的硬件木马检测算法[J]. 西安电子科技大学学报, 2019, 46(6): 37-45.
[2]	王志浩,张金松,孙光才,李军,邢孟道. 一种联合多特征量的SAR人造微弱痕迹检测方法[J]. 西安电子科技大学学报, 2019, 46(6): 131-139.
[3]	刘逸;慕彩红;刘敬. 结合邻域信息粒子群聚类用于SAR图像变化检测[J]. J4, 2015, 42(1): 187-193.
[4]	王立霞;付小宁. 靶场钻地弹爆炸位置的探测[J]. J4, 2014, 41(5): 124-128+172.
[5]	颜学颖;焦李成;王凌霞. 利用Brushlet变换进行SAR图像变化检测[J]. J4, 2013, 40(6): 67-73.
[6]	刘逸;寇卫东;慕彩红. 结合多阈值法的模糊聚类用于SAR图像变化检测[J]. J4, 2013, 40(6): 13-18.
[7]	王凌霞;焦李成;颜学颖;辛芳芳. 利用免疫克隆进行小波域遥感图像变化检测[J]. J4, 2013, 40(4): 108-113.
[8]	辛芳芳;焦李成;王桂婷;万红林. 利用小波域HMC模型进行遥感图像变化检测[J]. J4, 2012, 39(3): 43-49.
[9]	张齐;曹琦;王晓薇;毕笃彦. 融合灰色预测和HOGI特征的红外目标跟踪方法[J]. J4, 2010, 37(4): 751-757.
[10]	王法松1;3;李宏伟1;李睿2. 非参数GKNN估计的高效独立成分分析算法 [J]. J4, 2008, 35(4): 764-768.

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

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 10

Metrics

本文评价

推荐阅读 0