摘要
针对同一地区不同时期的全色高分辨遥感影像,提出一种基于影像融合的变化检测算法。首先采用基于匹配点的三角网校正方法对两景影像进行几何校正,然后选用基于迭代多元变化检测(IR-MAD)的相对辐射校正方法进行辐射一致性处理,接着对经几何一致性处理、辐射一致性处理后两张影像进行显著性融合,采用Mean-Shift分割算法对融合影像进行分割,选用方向梯度直方图(HOG)特征获取影像纹理强度图,最后通过比较分割块纹理变化获得变化检测结果。以Toposys激光雷达系统搭载相机拍摄的全色影像对该算法进行了检验,并使用单一时期影像为分割对象进行对比实验结果。实验结果表明,以融合影像为分割对象的结果远优于以单一时期影像为分割对象的变化检测结果,极大地减少了误检和漏检,在城市、郊区等地区人工地物变化监测中有一定的应用价值。
A change detection algorithm based on image fusion was proposed for high resolution panchromatic images of the same area in different period. Firstly,Triangulated Irregular Network( TIN) constructed using matching points was used to rectify the two images,Secondly,relative radiometric rectification was applied using Iteration-Multivariate alteration detection( IR-MAD). Thirdly,fusing the geometric and radiometric rectified images,and then the Mean-Shift segmentation method was utilized tosegment the fusing image. Histogram of Oriented Gradient( HOG) feature was chose to get the texture images. Finally,comparing the texture of the segmented blocks to get the change result. The change detection algorithm was verified with panchromatic images acquired by camera on Topo Sys Laser Scanner-System,and the proposed algorithm was compared with change detection algorithm that segmentation was applied on image of one period. Results show that the proposed algorithm superior to segmentation applied on one period image both in false rate and omission rate,and can satisfy the requirement of man-made object change detection in urban and suburban area.
引文
[1] Bruzzone L,Prieto,Diego Fernàndez. Automatic analysis of the difference image for unsupervised change detection[J]. IEEE Transactions on Geoscience&Remote Sensing,2000,38(3):1 171-1 182.
[2] Baudouin Desclée,Bogaert P,Defourny P. Forest change detection by statistical object-based method[J]. Remote Sensing of Environment,2006,102(1-2):1-11.
[3] Gamanya R,Maeyer P D,Dapper M D. Object-oriented change detection for the city of Harare,Zimbabwe[J]. Expert Systems with Applications,2009,36(1):571-588.
[4] Xian G,Homer C. Updating the 2001 National Land Cover Database Impervious Surface Products to 2006 using Landsat Imagery Change Detection Methods[J]. Remote Sensing of Environment,2010,114(8):1 676-1 686.
[5] Miller O,Pikaz A,Averbuch A. Objects based change detection in a pair of gray-level images[J]. Pattern Recognition,2005,38(11):1 976-1 992.
[6] Hussain M,Chen D,Cheng A,et al. Change detection from remotely sensed images:From pixel-based to objectbased approaches[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2013,80(Complete):91-106.
[7]孙晓霞,张继贤,燕琴,等.遥感影像变化检测方法综述及展望[J].遥感信息,2011(1):119-123.
[8]佃袁勇,方圣辉,姚崇怀.多尺度分割的高分辨率遥感影像变化检测[J].遥感学报,2016,20(1):129-137.
[9]杜培军,柳思聪.融合多特征的遥感影像变化检测[J].遥感学报,2012,16(4):663-677.