多波段信息融合的遥感影像变化检测

¹ 武汉大学印刷与包装系,湖北武汉,430079;² 民政部减灾和应急工程重点实验室,北京,100124;³ 武汉大学测绘遥感信息工程国家重点实验室,湖北武汉,430079

基金项目: 国家自然科学基金青年基金资助项目(61201341)；国家博士后基金资助项目(2012M511269)；民政部减灾和应急工程重点实验室开放基金资助项目(LDRERE20120205)；长江水利委员会长江科学院开放研究基金资助项目(CKWV2012325/KY)

详细信息

Remote Sensing Image Change Detection Based onMulti-band Information Fusion

¹School of Printing and Packaging,Wuhan University,Wuhan 430079,China;²Key Laboratory of Disaster Reduction and Emergency Response Engineering of the Ministry of Civil Affairs,Beijing 100124,China;³State Key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,Wuhan 430079,China

Funds: 国家自然科学基金青年基金资助项目(61201341)；国家博士后基金资助项目(2012M511269)；民政部减灾和应急工程重点实验室开放基金资助项目(LDRERE20120205)；长江水利委员会长江科学院开放研究基金资助项目(CKWV2012325/KY)

摘要: 传统遥感影像变化检测方法大多基于单波段信息,难以完整检测影像的变化信息。针对该问题,提出了一种基于马尔可夫随机场模型(MRF)的多波段遥感影像变化检测方法,利用MRF模型融合所有波段的变化信息。在求解MRF模型参数的过程中,引入MoLC与EM混合模型进行迭代计算。实验结果表明,本文方法的检测精度优于现有的变化检测方法,并且稳定性良好。
- 遥感影像 /
- 变化检测 /
- MRF模型 /
- 能量函数 /
- 信息融合 /
- 迭代运算
Abstract: Traditional remote sensing image change detection methods have difficulty in detecting com-plete change information based on a single-band.To solve this problem,the paper proposes a multi-band remote sensing image change detection method using a Markov Random Field,fusing all bandchange information.In the process of solving the MRF model parameters,the MoLC(method of Log-Cumulants)and EM(expectation-maximization)hybrid model is introduced for iterative calculation.Experimental results show that the detection accuracy of the proposed method is superior to the cur-rent change detection methods,and is stable.
- remote sensing image /
- change detection /
- MRF model /
- energy function /
- information fusion /
- iterative calculation