面向灾害应急土地覆被分类的样本自动选择方法研究
|
摘要
通过对自动化样本选择方法进行研究,实现了局部区域内面向对象的土地覆被自动分类。首先通过模糊聚类获得影像中的候选对象样本,分别提取影像特征和先验知识中的地类特征,通过预设阈值完成样本初步筛选,然后根据先验知识进行半监督距离度量学习,完成样本的自动选择,并为最终的监督分类提供度量依据。应用舟曲泥石流灾区影像进行了实验,结果表明,本文方法与基于人工选择样本的分类结果精度非常接近,同时在多次实验中表现出较高的稳定性,相对人工方法更加客观,适合批量自动化处理。
The automation level of classification for remote sensing image need to be improved to satisfy the timeliness and high-precision requirements in disaster emergency monitoring and assessment.But,the artificial selection of typical samples restricts the automatic interpretation of disaster information,a problem particularly acute for the development of business operation systems.This paper implements a totally automatic object-oriented land cover classification system based on automatic sample selection.First,the candidate object samples are acquired by fuzzy clustering.Second,image features and land type features are extracted from imagery and prior knowledge,respectively.Afterward,samples can be selected by applying preset thresholds on these features.Distance metric learning is then used not only for further sample selection,but also for more accurate supervised classification.Zhouqu Debris flow disaster images are computed by this method.Results show that the classification outcomes with samples selected automatically are very close to those samples selected by hand.Our results are more stable and objective than those produced manually.Moreover,it is more convenient to batch process images automatically.
The automation level of classification for remote sensing image need to be improved to satisfy the timeliness and high-precision requirements in disaster emergency monitoring and assessment.But,the artificial selection of typical samples restricts the automatic interpretation of disaster information,a problem particularly acute for the development of business operation systems.This paper implements a totally automatic object-oriented land cover classification system based on automatic sample selection.First,the candidate object samples are acquired by fuzzy clustering.Second,image features and land type features are extracted from imagery and prior knowledge,respectively.Afterward,samples can be selected by applying preset thresholds on these features.Distance metric learning is then used not only for further sample selection,but also for more accurate supervised classification.Zhouqu Debris flow disaster images are computed by this method.Results show that the classification outcomes with samples selected automatically are very close to those samples selected by hand.Our results are more stable and objective than those produced manually.Moreover,it is more convenient to batch process images automatically.
引文
[1]范一大,杨思全,王磊,等.汶川地震应急监测评估方法研究[J].遥感学报,2008,12(6):858-864
[2]杨思全,刘三超,吴玮,等.青海玉树地震遥感监测应用研究[J].航天器工程,2011,20(2):90-96
[3]Keuchel J,Naumann S,Heiler M,et al.AutomaticLand Cover Analysis for Tenerife by SupervisedClassification Using Remotely Sensed Data[J].Re-mote Sensing of Environment,2003,86(4):530-541
[4]Jordi I.Automatic Recognition of Man-made Ob-jects in High Resolution Optical Remote SensingImages by SVM Classification of Geometric ImageFeatures[J].ISPRS Journal of Photogrammetry andRemote Sensing,2007,62(3):236-248
[5]刘璞,张远,周斌,等.基于SAM和多源信息的土地利用/覆盖自动分类[J].浙江大学学报(工学版),2009,43(9):1 574-1 579
[6]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
[7]骆剑承,王钦敏,周成虎,等.基于自适应共振模型的遥感影像分类方法研究[J].测绘学报,2002,31(2):145-150
[8]王卫红,夏列钢,骆剑承,等.面向对象的遥感影像多层次迭代分类方法研究[J].武汉大学学报.信息科学版,2011,36(10):1154-1158
[9]Kuemmerle T,Radeloff V C,Perzanowski K,etal.Cross-border Comparison of Land Cover andLandscape Pattern in Eastern Europe Using a Hy-brid Classification Technique[J].Remote Sensing ofEnvironment,2006,103(4):449-464
[10]Ghosh A,Mishra N S,Ghosh S.Fuzzy ClusteringAlgorithms for Unsupervised Change Detection inRemote Sensing Images[J].Information Sciences,2011,181(4):699-715
[11]Mennis J,Guo D.Spatial Data Mining and Geo-graphic Knowledge Discovery—An Introduction[J].Computers,Environment and Urban Systems,2009,33(6):403-408
[12]Li J,Narayanan R M.Integrated Spectral and Spa-tial Information Mining in Remote Sensing Imagery[J].IEEE Transactions on Geoscience and RemoteSensing,2004,42(3):673-685
[13]Tang Hong,Fang Tao,Shi Pengfei.Spectral Simi-larity Measure Based on Fuzzy Feature ContrastModel[J].Optics Communications,2004,238(1/3):123-137
[14]van der Werff,van der Meer.Shape-based Classifi-cation of Spectrally Identical Objects[J].ISPRSJournal of Photogrammetry and Remote Sensing,2008,63(2):251-258
[15]Atkinson P M,Lewis P.Geostatistical Classifica-tion for Remote Sensing:An Introduction[J].Com-puters&Geosciences,2000,26(4):361-371
[16]吴一全,武燕燕.利用NSCT和Krawtchouk矩进行图像检索[J].武汉大学学报.信息科学版,2011,36(6):691-694
[17]Xing E P,Ng A Y,Jordan M I,et al.DistanceMetric Learning with Application to Clustering withSide-information[J].Advances in Neural Informa-tion Processing Systems,2003,15:521-528
[18]Weinberger K Q,Saul L K.Distance Metric Learn-ing for Large Margin Nearest Neighbor Classifica-tion[J].Journal of Machine Learning Research,2009,10:207-244
[19]国家测绘地理信息局.舟曲县灾前灾后权威高清影像对比图[EB/OL].http://www.sbsm.gov.cn/article/chyw/201008/20100800071142.shtml,2010
[2]杨思全,刘三超,吴玮,等.青海玉树地震遥感监测应用研究[J].航天器工程,2011,20(2):90-96
[3]Keuchel J,Naumann S,Heiler M,et al.AutomaticLand Cover Analysis for Tenerife by SupervisedClassification Using Remotely Sensed Data[J].Re-mote Sensing of Environment,2003,86(4):530-541
[4]Jordi I.Automatic Recognition of Man-made Ob-jects in High Resolution Optical Remote SensingImages by SVM Classification of Geometric ImageFeatures[J].ISPRS Journal of Photogrammetry andRemote Sensing,2007,62(3):236-248
[5]刘璞,张远,周斌,等.基于SAM和多源信息的土地利用/覆盖自动分类[J].浙江大学学报(工学版),2009,43(9):1 574-1 579
[6]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
[7]骆剑承,王钦敏,周成虎,等.基于自适应共振模型的遥感影像分类方法研究[J].测绘学报,2002,31(2):145-150
[8]王卫红,夏列钢,骆剑承,等.面向对象的遥感影像多层次迭代分类方法研究[J].武汉大学学报.信息科学版,2011,36(10):1154-1158
[9]Kuemmerle T,Radeloff V C,Perzanowski K,etal.Cross-border Comparison of Land Cover andLandscape Pattern in Eastern Europe Using a Hy-brid Classification Technique[J].Remote Sensing ofEnvironment,2006,103(4):449-464
[10]Ghosh A,Mishra N S,Ghosh S.Fuzzy ClusteringAlgorithms for Unsupervised Change Detection inRemote Sensing Images[J].Information Sciences,2011,181(4):699-715
[11]Mennis J,Guo D.Spatial Data Mining and Geo-graphic Knowledge Discovery—An Introduction[J].Computers,Environment and Urban Systems,2009,33(6):403-408
[12]Li J,Narayanan R M.Integrated Spectral and Spa-tial Information Mining in Remote Sensing Imagery[J].IEEE Transactions on Geoscience and RemoteSensing,2004,42(3):673-685
[13]Tang Hong,Fang Tao,Shi Pengfei.Spectral Simi-larity Measure Based on Fuzzy Feature ContrastModel[J].Optics Communications,2004,238(1/3):123-137
[14]van der Werff,van der Meer.Shape-based Classifi-cation of Spectrally Identical Objects[J].ISPRSJournal of Photogrammetry and Remote Sensing,2008,63(2):251-258
[15]Atkinson P M,Lewis P.Geostatistical Classifica-tion for Remote Sensing:An Introduction[J].Com-puters&Geosciences,2000,26(4):361-371
[16]吴一全,武燕燕.利用NSCT和Krawtchouk矩进行图像检索[J].武汉大学学报.信息科学版,2011,36(6):691-694
[17]Xing E P,Ng A Y,Jordan M I,et al.DistanceMetric Learning with Application to Clustering withSide-information[J].Advances in Neural Informa-tion Processing Systems,2003,15:521-528
[18]Weinberger K Q,Saul L K.Distance Metric Learn-ing for Large Margin Nearest Neighbor Classifica-tion[J].Journal of Machine Learning Research,2009,10:207-244
[19]国家测绘地理信息局.舟曲县灾前灾后权威高清影像对比图[EB/OL].http://www.sbsm.gov.cn/article/chyw/201008/20100800071142.shtml,2010
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心