XU Jun,YANG Hang,LI Jiansong,et al.The building change information detection of high resolution imagecombined with MBI mul-features and GIS buffer[J].Journal of Yanbian University,2017,43(01):84-88.
MBI多特征结合GIS缓冲区的高分影像建筑物变化信息检测
- Title:
- The building change information detection of high resolution image combined with MBI mul-features and GIS buffer
- Keywords:
- MBI; mul-feature; buffer; high resolution image; building; change detection
- 分类号:
- P237
- 文献标志码:
- A
- 摘要:
- 为准确地从高分辨率遥感影像中提取建筑物及其变化信息,基于对建筑物先进行提取再变化检测的思想,在采用改进的分水岭算法进行影像分割基础上,结合形态学建筑物指数(MBI)的多特征支持向量机建筑物提取方法,将GIS缓冲区引入建筑物变化检测.针对配准误差、提取误差和建筑物位移现象引起的误差,构建引入缓冲区的建筑物变化检测流程.以0.2 m分辨率航空遥感影像为例,对建筑物变化进行了检测试验,结果表明本文的建筑物变化检测流程能够有效提高建筑物变化检测的准确性.
- Abstract:
- In order to accurately extract the building and change information from high-resolution remote sensing images, the idea of first extracted building and then detected change was adopted. Based on the image segmentation with improved watershed algorithm, combinated the SVM method with MBI and multi-feature, the GIS buffer was introduced to change detection. According to the detection error caused by registration error, extraction error and building displacement phenomenon, the building change detection process that introduces GIS buffer was constructed. With 0.2 m resolution aerial remote sensing images case, the building changes detection was tested, the test results showed that building change detection process can effectively improve the accuracy of the building change detection.
参考文献/References:
[1] 段静辉.高分辨率卫星图像中建筑物提取及变化检测研究[D].北京:中国科学院自动化研究所,2007.
[2] 谭衢霖.高分辨率多光谱影像城区建筑物提取研究[J].测绘学报,2010,39(6):618-623.
[3] 季顺平,袁修孝.一种基于阴影检测的建筑物变化检测方法[J].遥感学报,2007,11(3):323-329.
[4] 袁修孝,宋妍.一种运用纹理和光谱特征消除投影差影响的建筑物变化检测方法[J].武汉大学学报(信息科学版),2007,32(6):489-493.
[5] Tian J, Chaabouni-Chouayakh H, Reinartz P. 3D building change detection from high resolution spaceborne steroimagery[C]//International Workshop on Multi-platform/multi-platform/multi-sensor remote sensing & mapping, Piscataway, NJ:IEEE, 2011:1-7.
[6] 邵振峰,陶峰.一种基于多特征融合的面向对象的建筑物变化检测方法:中国,CN201510187939.9[P].2015-07-01.
[7] Beumier C, Idrissa M. Building change detection by histogram classification[C]// International conference on signal-image technology & internet-based systems, Piscataway, NJ:IEEE, 2011:409-415.
[8] 吕钊,康志忠.利用地面激光扫描数据进行建筑物变化检测[J].武汉大学学报(信息科学版),2011,36(11):1284-1289.
[9] 汪承义,孔赟珑,陈静波.一种多时相激光雷达数据建筑物变化检测方法[C]//第十三届中国体视学与图像分析学术会议论文集.太原:中国体视学学会,2013:319-324.
[10] 巩翼龙,闫利.结合机载LIDAR数据和航空可见光影像多层次规则分类建筑物变化检测[J].光谱学与光谱分析,2015,35(5):1325-1330.
[11] 彭代峰,张永军,熊小东.结合LIDAR点云和航空影像的建筑物三维变化检测[J].武汉大学学报(信息科学版),2015,40(4):462-468.
[12] Huang X, Zhang L. A multidirectional and multiscale morphological index for automatic building extraction from multispectral geoeye-1 imagery[J]. Photogrammetric Engineering & Remote Sensing, 2011,77(7):721-732.
[13] 胡荣明,黄小兵,黄远程.增强形态学建筑物指数应用于高分辨率遥感影像中建筑物提取[J].测绘学报,2014,43(5):514-520.
[14] 任晓娟,肖双九,彭小朋.基于改进分水岭变换的遥感图像建筑物提取[J].计算机应用与软件,2011,28(12):249-252.
[15] 张峰.基于SVM的多源遥感影像面向对象建筑物提取方法[J].国土资源遥感,2008(2):27-29.
[16] 秦昆.GIS空间分析理论与方法[M].武汉:武汉大学出版社,2010:125-131.
备注/Memo
收稿日期: 2016-09-19 基金项目: 广西自然科学基金资助项目(2013GXNSFAA019285)
*通信作者: 韦金丽(1967—),女,高级工程师,研究方向为遥感和地理信息系统.