利用成像匹配与成分分离的航空图像云检测
|
摘要
研究了遥感图像云检测,提出了基于几何成像匹配与成分分离的航空图像云检测算法。由于传统的云检测算法未考虑云具有半透明性的特点,直接提取的云纹理包含有多余的下垫面纹理。该算法依据线性光谱混合模型,将一幅遥感图像看作是由下垫面与云的光谱线性地构成的,考虑像素间的局部平滑,进行云成分分离,采用局部二值模型(LBP)特征描述云的纹理,进而通过构建支持向量机(SVM)分类器进行云检测。对比实验结果表明,本文方法对于航空图像云检测具有一定的效果,对于薄云区域以及边缘区域也能有效检测。
Cloud detection of remote sensing images is studied,and an aerial imaging cloud detection algorithm using ge-ometric imaging matching and component separation is proposed. Directly extracted cloud textures by the existingcloud detection algorithm contain redundant underlying surface textures duo to the ignoring of translucency of theclouds. The proposed algorithm regards that a remote sensing image is composed of the spectrum of the underlyingsurface and the clouds linearly,according to linear spectral mixture model. Considering the local smoothing be-tween pixels,the cloud component separation is processed. After that,local binary pattern( LBP) features areused to describe the textures of the clouds,and the clouds are detected by constructing support vector machine( SVM) classifier. The experimental results show that this method is effective for aerial image cloud detection,aswell as the thin cloud area and the edge region.
Cloud detection of remote sensing images is studied,and an aerial imaging cloud detection algorithm using ge-ometric imaging matching and component separation is proposed. Directly extracted cloud textures by the existingcloud detection algorithm contain redundant underlying surface textures duo to the ignoring of translucency of theclouds. The proposed algorithm regards that a remote sensing image is composed of the spectrum of the underlyingsurface and the clouds linearly,according to linear spectral mixture model. Considering the local smoothing be-tween pixels,the cloud component separation is processed. After that,local binary pattern( LBP) features areused to describe the textures of the clouds,and the clouds are detected by constructing support vector machine( SVM) classifier. The experimental results show that this method is effective for aerial image cloud detection,aswell as the thin cloud area and the edge region.
引文
[1] Hagolle O,Huc M,Pascual D V,et al. A multi-temporalmethod for cloud detection,applied to FORMOSAT-2,VENμS,LANDSAT and SENTINEL-2 images[J]. Re-mote Sensing of Environment,2010,114(8):1747-1755
[2] Nakajima T Y,Tsuchiya T,Ishida H,et al. Cloud detec-tion performance of spaceborne visible-to-infrared multi-spectral imagers[J]. Applied Optics,2011,50(17):2601-2616
[3]谭凯,张永军,童心,等.国产高分辨率遥感卫星影像自动云检测[J].测绘学报,2016,45(5):581-591
[4]胡根生,陈长春,梁栋.联合云量自动评估和加权支持向量机的Landsat图像云检测[J].测绘学报,2014,43(8):848-854
[5]高贤君,万幼川,郑顺义,等.航空摄影过程中云的实时自动检测[J].光谱学与光谱分析,2014,34(7):1909-1913
[6]单娜,郑天垚,王贞松.快速高准确度云检测算法及其应用[J].遥感学报,2009,13(6):1147-1162
[7]陶淑苹,金光,张贵祥,等.实现遥感相机自主辨云的小波SCM算法[J].测绘学报,2011,40(5):598-603
[8] Ojala T,Pietikainen M,MaenpaaT. Multiresolutiongray-scale and rotation invariant texture classification withlocal binary patterns[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2002,24(7):971-987
[9] Ishida H,Nakajima T Y. Development of an unbiasedcloud detection algorithm for a spaceborne multispectralimager[J]. Journal of Geophysical Research:Atmos-pheres,2009,114(D7):206-221
[10] Ameur Z,Ameur S,Adane A,et al. Cloud classificationusing the textural features of Meteosatimages[J]. Interna-tional journal of remote sensing,2004,25(21):4491-4503
[11]张剑清,潘励,王树根.摄影测量学[M].武汉:武汉大学出版社,2009
[12]刘琼.导引概率图与显著特征相结合的行人目标检测[J].高技术通讯,2016,26(5):464-474
[13]胡正平,张敏姣,李淑芳,等.基于局部最大概率特征和映射模型学习的行人再识别[J].高技术通讯,2018,28(3):185-193
[14]储茂祥,王安娜,巩荣芬.一种改进的最小二乘孪生支持向量机分类算法[J].电子学报,2014,42(5):998-1003
[15]苑玮琦,朱立军,张波.基于形态学与支持向量机的虹膜坑洞纹理检测[J].仪器仪表学报,2017,38(3):664-671
[16]邹冲,蔡敦波,刘莹,等.组合SVM分类器在行人检测中的研究[J].计算机科学,2017,44(Z6):188-191
[2] Nakajima T Y,Tsuchiya T,Ishida H,et al. Cloud detec-tion performance of spaceborne visible-to-infrared multi-spectral imagers[J]. Applied Optics,2011,50(17):2601-2616
[3]谭凯,张永军,童心,等.国产高分辨率遥感卫星影像自动云检测[J].测绘学报,2016,45(5):581-591
[4]胡根生,陈长春,梁栋.联合云量自动评估和加权支持向量机的Landsat图像云检测[J].测绘学报,2014,43(8):848-854
[5]高贤君,万幼川,郑顺义,等.航空摄影过程中云的实时自动检测[J].光谱学与光谱分析,2014,34(7):1909-1913
[6]单娜,郑天垚,王贞松.快速高准确度云检测算法及其应用[J].遥感学报,2009,13(6):1147-1162
[7]陶淑苹,金光,张贵祥,等.实现遥感相机自主辨云的小波SCM算法[J].测绘学报,2011,40(5):598-603
[8] Ojala T,Pietikainen M,MaenpaaT. Multiresolutiongray-scale and rotation invariant texture classification withlocal binary patterns[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2002,24(7):971-987
[9] Ishida H,Nakajima T Y. Development of an unbiasedcloud detection algorithm for a spaceborne multispectralimager[J]. Journal of Geophysical Research:Atmos-pheres,2009,114(D7):206-221
[10] Ameur Z,Ameur S,Adane A,et al. Cloud classificationusing the textural features of Meteosatimages[J]. Interna-tional journal of remote sensing,2004,25(21):4491-4503
[11]张剑清,潘励,王树根.摄影测量学[M].武汉:武汉大学出版社,2009
[12]刘琼.导引概率图与显著特征相结合的行人目标检测[J].高技术通讯,2016,26(5):464-474
[13]胡正平,张敏姣,李淑芳,等.基于局部最大概率特征和映射模型学习的行人再识别[J].高技术通讯,2018,28(3):185-193
[14]储茂祥,王安娜,巩荣芬.一种改进的最小二乘孪生支持向量机分类算法[J].电子学报,2014,42(5):998-1003
[15]苑玮琦,朱立军,张波.基于形态学与支持向量机的虹膜坑洞纹理检测[J].仪器仪表学报,2017,38(3):664-671
[16]邹冲,蔡敦波,刘莹,等.组合SVM分类器在行人检测中的研究[J].计算机科学,2017,44(Z6):188-191