一种频域高分辨率遥感图像线状特征检测方法
|
摘要
高分辨率遥感图像中线状信息过分细节化,相应的对特征的提取带来很大干扰。引入一种基于方向和频率特征的遥感图像频域线状特征检测方法,并通过傅氏变换将图像变换到频率域。在详细分析线状特征和谱线的关系、线状特征和图像频率之间关系的基础上,由分析得到的方向和频率的参数构造Gabor滤波器进行图像线状特征的提取。高分辨率遥感图像试验结果表明,该方法能较好地提取图像的线状特征。
Information of linear features in high-resolution remotely sensed imagery is very prolific,consequently the immoderate specializations and influence of noise disturbances are serious,this made it difficult to detect and recognize the linear features of landscape targets.A method is discussed based on direction and frequency characters through designing frequency domain filter for the linear feature automatic recognition and detection,owed to Fourier transformation method transform the image into frequency domain.Analyzed the relations between the linear feature and its spectrum line,also the relations between the linear feature and its frequency character,according of the direction and frequency selected above the Gabor wave filter the image linearity feature extracting was constructed.The relevant extract experiments were tested with QuickBird image,the result indicates that the linearity characteristic extracted in this method is fairly good,it provided a reference for high-resolution remotely sensed imagery linear feature extracting.
Information of linear features in high-resolution remotely sensed imagery is very prolific,consequently the immoderate specializations and influence of noise disturbances are serious,this made it difficult to detect and recognize the linear features of landscape targets.A method is discussed based on direction and frequency characters through designing frequency domain filter for the linear feature automatic recognition and detection,owed to Fourier transformation method transform the image into frequency domain.Analyzed the relations between the linear feature and its spectrum line,also the relations between the linear feature and its frequency character,according of the direction and frequency selected above the Gabor wave filter the image linearity feature extracting was constructed.The relevant extract experiments were tested with QuickBird image,the result indicates that the linearity characteristic extracted in this method is fairly good,it provided a reference for high-resolution remotely sensed imagery linear feature extracting.
引文
[1]WILKI NSON G G.Recent Development in Remote SensingTechnology and the Importance of Computer Vision AnalysisTechniques[C]∥Proceedings of Concerted Action MAVI-RIC.Kingston:[s.n.],1999.
[2]CHENG Chengqi,MA Ting.Automatic Recognition ofLandscape Linear Features from High-resolution SatelliteI mages[J].Journal of Remote Sensing,2003,7(1):26-30.(程承旗,马廷.高分辨率卫星影像上地物线性特征的自动识别[J].遥感学报,2003,7(1):26-30.)
[3]SOBELI.Neighbourhood Coding of Binary I mages for FastContour Following and General Array Binary Processing[J].Computer Graphics and I mage Processing,1978,8:127-135.
[4]CANNYJ F.AComputational Approachto Edge Detection[J].IEEE Transactions on Pattern Analysis and MachineIntelligence,1986,8(6):679-698.
[5]HOSSEI NI R.Fourier Analysis of Plain Weave FabricAppearance[J].Textile Research Journal,1995,65(11):676-683.
[6]KOVESI P.Invariant Measures of I mage Features fromPhase Information[D].Perth:University of WesternAustralia,1996.
[7]DELENNE C,RABATEL G,AGURTO V,et al.VinePlot Detectionin Aerial I mages Using Fourier Analysis[C]∥Proceedings of 1st International Conference OBI A.Salzburg:ISPRS,2006.
[8]WASSENAAR T,ROBBEZ-MASSONJ M,ANDRIEUXP,et al.Vineyard Identification and Description of SpatialCrop Structure by Per-field Frequency Analysis[J].Inter-national Journal of Remote Sensing,2002,23(17):3311-3325.
[9]LUO Y,SAUDI A,MARHOON M.Generalized HilbertTransform and Its Applications in Geophysics[J].TheLeading Eedge,2003,22(3):198-202.
[10]MORRONE M C,BURR D C,ROSS J,et al.MachBands Are Phase Dependent[J].Nature,1986,324:250-253.
[11]XI AO Pengfeng,FENG Xuezhi,ZHAO Shuhe,et al.Segmentation of High-resolution Remotely Sensed I mage-ry Based on Phase Congruency[J].Acta Geodaetica etCartographica Sinica,2007,36(2):146-151.(肖鹏峰,冯学智,赵书河,等.基于相位一致的高分辨率遥感图像分割方法[J].测绘学报,2007,36(2):146-151.)
[12]KOVESI P.I mage Features from Phase Congruency[J].Journal of Computer Vision Research,1999,1(3):1-26.
[13]BEZALEL E,EFRON U.Efficient Face RecognitionMethod Using a Combined Phase Congruency/GaborWavelet Technique[C]∥Proceedings of the SPIE.SanDiego:SPIE,2005:437-444.
[14]CHEN Shupeng.Investigation Studies on Geo-informaticTupu[M].Beijing:Commercial Press,2001.(陈述彭.地学信息图谱探索研究[M].北京:商务印书馆,2001.)
[15]TAN T N.Texture Edge Detection by Modeling VisualCortical Channels[J].Pattern Recognition,1995,28(9):1283-1298.
[16]KI M D S,LEE S U.I mage Vector Quantizer Based on aClassificationin the DCT Domain[J].IEEE Transactionson Communications,1991,39(4):549-556.
[17]TAN Yanying,DONGZhixin.Spectrum Analyzing of I m-age Edges and Applicationin Transform Coding[J].ActaElectronica Sinica,1995,23(9):61-65.(谭雁英,董志信.图像边缘信息的谱分析及其在变换编码中的应用[J].电子学报,1995,23(9):61-65.)
[18]CHEN Xuehua,HE Zhenhua,HUANG Deji.Seismic DataEdge Detection Based on Higher-order Pseudo HilbertTransform[J].Progress in Geophysics,2008,23(4):1106-1110.(陈学华,贺振华,黄德济.地震资料的高阶伪希尔伯特变换边缘检测[J].地球物理学进展,2008,23(4):1106-1110.)
[19]DELENNE C,RABATEL G,DESHAYES M.AnAutomatizedFrequency Analysis for Vine Plot Detectionand Delineation in Remote Sensing[J].IEEE Geoscienceand Remote Sensing Letters,2008,5(3):341-345.
[20]FU Yipping,LI Zhineng,YUAN Ding.Edge Detectionwith Opti mized Gabor Filter[J].Journal of Computer-aided Design and Computer Graphics,2004,16(4):481-486.(傅一平,李志能,袁丁.基于优化设计Gabor滤波器的边缘提取方法[J].计算机辅助设计与图形学学报,2004,16(4):481-486.)
[21]WU Gaohong ZHANG Yujin LI N Xinggang.Opti malGabor Filter Design for Bi-textured I mage Segmentation[J].Acta Electronica Sinica,2001,29(1):48-50.(吴高洪,章毓晋,林行刚.分割双纹理图像的最佳Gabor滤波器设计方法[J].电子学报,2001,29(1):48-50.)
[2]CHENG Chengqi,MA Ting.Automatic Recognition ofLandscape Linear Features from High-resolution SatelliteI mages[J].Journal of Remote Sensing,2003,7(1):26-30.(程承旗,马廷.高分辨率卫星影像上地物线性特征的自动识别[J].遥感学报,2003,7(1):26-30.)
[3]SOBELI.Neighbourhood Coding of Binary I mages for FastContour Following and General Array Binary Processing[J].Computer Graphics and I mage Processing,1978,8:127-135.
[4]CANNYJ F.AComputational Approachto Edge Detection[J].IEEE Transactions on Pattern Analysis and MachineIntelligence,1986,8(6):679-698.
[5]HOSSEI NI R.Fourier Analysis of Plain Weave FabricAppearance[J].Textile Research Journal,1995,65(11):676-683.
[6]KOVESI P.Invariant Measures of I mage Features fromPhase Information[D].Perth:University of WesternAustralia,1996.
[7]DELENNE C,RABATEL G,AGURTO V,et al.VinePlot Detectionin Aerial I mages Using Fourier Analysis[C]∥Proceedings of 1st International Conference OBI A.Salzburg:ISPRS,2006.
[8]WASSENAAR T,ROBBEZ-MASSONJ M,ANDRIEUXP,et al.Vineyard Identification and Description of SpatialCrop Structure by Per-field Frequency Analysis[J].Inter-national Journal of Remote Sensing,2002,23(17):3311-3325.
[9]LUO Y,SAUDI A,MARHOON M.Generalized HilbertTransform and Its Applications in Geophysics[J].TheLeading Eedge,2003,22(3):198-202.
[10]MORRONE M C,BURR D C,ROSS J,et al.MachBands Are Phase Dependent[J].Nature,1986,324:250-253.
[11]XI AO Pengfeng,FENG Xuezhi,ZHAO Shuhe,et al.Segmentation of High-resolution Remotely Sensed I mage-ry Based on Phase Congruency[J].Acta Geodaetica etCartographica Sinica,2007,36(2):146-151.(肖鹏峰,冯学智,赵书河,等.基于相位一致的高分辨率遥感图像分割方法[J].测绘学报,2007,36(2):146-151.)
[12]KOVESI P.I mage Features from Phase Congruency[J].Journal of Computer Vision Research,1999,1(3):1-26.
[13]BEZALEL E,EFRON U.Efficient Face RecognitionMethod Using a Combined Phase Congruency/GaborWavelet Technique[C]∥Proceedings of the SPIE.SanDiego:SPIE,2005:437-444.
[14]CHEN Shupeng.Investigation Studies on Geo-informaticTupu[M].Beijing:Commercial Press,2001.(陈述彭.地学信息图谱探索研究[M].北京:商务印书馆,2001.)
[15]TAN T N.Texture Edge Detection by Modeling VisualCortical Channels[J].Pattern Recognition,1995,28(9):1283-1298.
[16]KI M D S,LEE S U.I mage Vector Quantizer Based on aClassificationin the DCT Domain[J].IEEE Transactionson Communications,1991,39(4):549-556.
[17]TAN Yanying,DONGZhixin.Spectrum Analyzing of I m-age Edges and Applicationin Transform Coding[J].ActaElectronica Sinica,1995,23(9):61-65.(谭雁英,董志信.图像边缘信息的谱分析及其在变换编码中的应用[J].电子学报,1995,23(9):61-65.)
[18]CHEN Xuehua,HE Zhenhua,HUANG Deji.Seismic DataEdge Detection Based on Higher-order Pseudo HilbertTransform[J].Progress in Geophysics,2008,23(4):1106-1110.(陈学华,贺振华,黄德济.地震资料的高阶伪希尔伯特变换边缘检测[J].地球物理学进展,2008,23(4):1106-1110.)
[19]DELENNE C,RABATEL G,DESHAYES M.AnAutomatizedFrequency Analysis for Vine Plot Detectionand Delineation in Remote Sensing[J].IEEE Geoscienceand Remote Sensing Letters,2008,5(3):341-345.
[20]FU Yipping,LI Zhineng,YUAN Ding.Edge Detectionwith Opti mized Gabor Filter[J].Journal of Computer-aided Design and Computer Graphics,2004,16(4):481-486.(傅一平,李志能,袁丁.基于优化设计Gabor滤波器的边缘提取方法[J].计算机辅助设计与图形学学报,2004,16(4):481-486.)
[21]WU Gaohong ZHANG Yujin LI N Xinggang.Opti malGabor Filter Design for Bi-textured I mage Segmentation[J].Acta Electronica Sinica,2001,29(1):48-50.(吴高洪,章毓晋,林行刚.分割双纹理图像的最佳Gabor滤波器设计方法[J].电子学报,2001,29(1):48-50.)
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