基于相干系数-马尔可夫随机场的高分辨率SAR图像建筑物分割算法（英文）
|
摘要
高分辨率合成孔径雷达(Synthetic aperture radar,SAR)图像的建筑物分割问题一直是重要的研究课题之一。由于斑点噪声和多路径效应的存在以及建筑物几何结构的影响,建筑物区域内部会产生强散射斑点,像素强度值的差异较大,给建筑物分割和提取带来了困难。针对这个问题,本文提出了一种基于相干系数-马尔科夫随机场(coherence-coefficient-based Markov random field,CCMRF)的高分辨率SAR建筑物分割算法,该方法将干涉合成孔径雷达(interferometric synthetic aperture radar,InSAR)的相干系数引入到传统马尔可夫随机场(Markov random field,MRF)的邻域能量中,使得相干信息和空间上下文信息得到更充分的利用。根据Hammersley-Clifford定理,图像分割的最大后验(Maximum a posteriori,MAP)问题被转化为最小化似然能量和邻域能量之和的问题,最后采用迭代条件模型(Iterative condition model, ICM)得到最优解。实验结果表明,该方法与传统的马尔可夫方法和K均值聚类方法相比,可以有效地对SAR建筑物进行分割并获得更准确的结果。
Building segmentation from high-resolution synthetic aperture radar(SAR) images has always been one of the important research issues. Due to the existence of speckle noise and multipath effect, the pixel values change drastically, causing the large intensity differences in pixels of building areas. Moreover, the geometric structure of buildings can cause strong scattering spots, which brings difficulties to the segmentation and extraction of buildings. To solve of these problems, this paper presents a coherence-coefficient-based Markov random field(CCMRF) approach for building segmentation from high-resolution SAR images. The method introduces the coherence coefficient of interferometric synthetic aperture radar(InSAR) into the neighborhood energy based on traditional Markov random field(MRF), which makes interferometric and spatial contextual information more fully used in SAR image segmentation. According to the Hammersley-Clifford theorem, the problem of maximum a posteriori(MAP) for image segmentation is transformed into the solution of minimizing the sum of likelihood energy and neighborhood energy. Finally, the iterative condition model(ICM) is used to find the optimal solution. The experimental results demonstrate that the proposed method can segment SAR building effectively and obtain more accurate results than the traditional MRF method and K-means clustering.
Building segmentation from high-resolution synthetic aperture radar(SAR) images has always been one of the important research issues. Due to the existence of speckle noise and multipath effect, the pixel values change drastically, causing the large intensity differences in pixels of building areas. Moreover, the geometric structure of buildings can cause strong scattering spots, which brings difficulties to the segmentation and extraction of buildings. To solve of these problems, this paper presents a coherence-coefficient-based Markov random field(CCMRF) approach for building segmentation from high-resolution SAR images. The method introduces the coherence coefficient of interferometric synthetic aperture radar(InSAR) into the neighborhood energy based on traditional Markov random field(MRF), which makes interferometric and spatial contextual information more fully used in SAR image segmentation. According to the Hammersley-Clifford theorem, the problem of maximum a posteriori(MAP) for image segmentation is transformed into the solution of minimizing the sum of likelihood energy and neighborhood energy. Finally, the iterative condition model(ICM) is used to find the optimal solution. The experimental results demonstrate that the proposed method can segment SAR building effectively and obtain more accurate results than the traditional MRF method and K-means clustering.
引文
[1] Stilla U,Soergel U,Thoennessen U.Potential and limits of InSAR data for building reconstruction in built-up areas.Isprs Journal of Photogrammetry & Remote Sensing,2003,58 (1):113-123.
[2] Ferro A,Brunner D,Bruzzone L.Automatic detection and reconstruction of building radar footprints from single VHR SAR images.IEEE Transactions on Geoscience & Remote Sensing,2013,51(2):935-952.
[3] Chen S,Wang H,Xu F,et al.Automatic recognition of isolated buildings on single-aspect SAR image using range detector.IEEE Geoscience & Remote Sensing Letters,2014,12(2):219-223.
[4] Rossi C,Eineder M.High-resolution InSAR building layovers detection and exploitation.IEEE Transactions on Geoscience & Remote Sensing,2015,53(12):6457-6468.
[5] Thiele A,Cadario E,Schulz K,et al.Analysis of gable-roofed building signature in multiaspect InSAR data.IEEE Geoscience & Remote Sensing Letters,2010,7(1):83-87.
[6] Sportouche H,Tupin F,Denise L.Extraction and three-dimensional reconstruction of isolated buildings in urban scenes from high-resolution optical and SAR spaceborne images.IEEE Transactions on Geoscience & Remote Sensing,2011,49(10):3932-3946.
[7] Deng H W,Clausi D A.Unsupervised segmentation of synthetic aperture radar sea ice imagery using a novel Markov random field model.IEEE Transactions on Geoscience & Remote Sensing,2005,43(3):528-538.
[8] Moser G,Serpico S B.Combining support vector machines and markov random fields in an integrated framework for contextual image classification.IEEE Transactions on Geoscience & Remote Sensing,2013,51(5):2734-2752.
[9] Sun L,Wu Z,Liu J,et al.Supervised spectral-spatial hyperspectral image classification with weighted Markov random fields.IEEE Transactions on Geoscience & Remote Sensing,2015,53(3):1490-1503.
[10] Tarabalka Y,Fauvel M,Chanussot J,et al.SVM- and MRF-based method for accurate classification of hyperspectral images.IEEE Geoscience & Remote Sensing Letters,2010,7(4):736-740.
[11] Wu Y,Li M,Zhang P,et al.Unsupervised multi-class segmentation of SAR images using triplet Markov fields models based on edge penalty.Pattern Recognition Letters,2011,32(11):1532-1540.
[12] Voisin A,Krylov V A,Moser G,et al.Classification of very high resolution SAR images of urban areas using copulas and texture in a hierarchical Markov random field model.IEEE Geoscience & Remote Sensing Letters,2013,10(1):96-100.
[13] Boudaren M E Y,Lin A,Pieczynski W.Unsupervised segmentation of SAR images using Gaussian mixture-hidden evidential markov fields.IEEE Geoscience & Remote Sensing Letters,2016,13(12):1865-1869.
[14] Zhang H,Shi W Z,Wang Y J,et al.Spatial-attraction-based Markov random field approach for classification of high spatial resolution multispectral imagery.IEEE Geoscience & Remote Sensing Letters,2014,11(2):489-493.
[15] Solberg A H S,Taxt T,Jain A K.A Markov random field model for classification of multisource satellite imagery.IEEE Transactions on Geoscience & Remote Sensing,1996,34(1):100-113.
[16] Tison C,Nicolas J M,Tupin F,et al.A new statistical model for Markovian classification of urban areas in high-resolution SAR images.IEEE Transactions on Geoscience & Remote Sensing,2004,42(10):2046-2057.
[17] Touzi R,Lopes A,Bruniquel J,et al.Coherence estimation for SAR imagery.IEEE Transactions on Geoscience & Remote Sensing,1999,37(1):135-149.
[18] Askne J,Hagberg J O.Potential of interferometric SAR for classification of land surfaces.In:Proceedings of the IEEE International Geoscience and Remote Sensing Symposium,Tokyo,1993:985-987.
[19] Shufelt J A.Performance evaluation and analysis of monocular building extraction from aerial imagery.IEEE Trans Pattern Analysis & Machine Intelligence,1999,21(4):311-326.
[2] Ferro A,Brunner D,Bruzzone L.Automatic detection and reconstruction of building radar footprints from single VHR SAR images.IEEE Transactions on Geoscience & Remote Sensing,2013,51(2):935-952.
[3] Chen S,Wang H,Xu F,et al.Automatic recognition of isolated buildings on single-aspect SAR image using range detector.IEEE Geoscience & Remote Sensing Letters,2014,12(2):219-223.
[4] Rossi C,Eineder M.High-resolution InSAR building layovers detection and exploitation.IEEE Transactions on Geoscience & Remote Sensing,2015,53(12):6457-6468.
[5] Thiele A,Cadario E,Schulz K,et al.Analysis of gable-roofed building signature in multiaspect InSAR data.IEEE Geoscience & Remote Sensing Letters,2010,7(1):83-87.
[6] Sportouche H,Tupin F,Denise L.Extraction and three-dimensional reconstruction of isolated buildings in urban scenes from high-resolution optical and SAR spaceborne images.IEEE Transactions on Geoscience & Remote Sensing,2011,49(10):3932-3946.
[7] Deng H W,Clausi D A.Unsupervised segmentation of synthetic aperture radar sea ice imagery using a novel Markov random field model.IEEE Transactions on Geoscience & Remote Sensing,2005,43(3):528-538.
[8] Moser G,Serpico S B.Combining support vector machines and markov random fields in an integrated framework for contextual image classification.IEEE Transactions on Geoscience & Remote Sensing,2013,51(5):2734-2752.
[9] Sun L,Wu Z,Liu J,et al.Supervised spectral-spatial hyperspectral image classification with weighted Markov random fields.IEEE Transactions on Geoscience & Remote Sensing,2015,53(3):1490-1503.
[10] Tarabalka Y,Fauvel M,Chanussot J,et al.SVM- and MRF-based method for accurate classification of hyperspectral images.IEEE Geoscience & Remote Sensing Letters,2010,7(4):736-740.
[11] Wu Y,Li M,Zhang P,et al.Unsupervised multi-class segmentation of SAR images using triplet Markov fields models based on edge penalty.Pattern Recognition Letters,2011,32(11):1532-1540.
[12] Voisin A,Krylov V A,Moser G,et al.Classification of very high resolution SAR images of urban areas using copulas and texture in a hierarchical Markov random field model.IEEE Geoscience & Remote Sensing Letters,2013,10(1):96-100.
[13] Boudaren M E Y,Lin A,Pieczynski W.Unsupervised segmentation of SAR images using Gaussian mixture-hidden evidential markov fields.IEEE Geoscience & Remote Sensing Letters,2016,13(12):1865-1869.
[14] Zhang H,Shi W Z,Wang Y J,et al.Spatial-attraction-based Markov random field approach for classification of high spatial resolution multispectral imagery.IEEE Geoscience & Remote Sensing Letters,2014,11(2):489-493.
[15] Solberg A H S,Taxt T,Jain A K.A Markov random field model for classification of multisource satellite imagery.IEEE Transactions on Geoscience & Remote Sensing,1996,34(1):100-113.
[16] Tison C,Nicolas J M,Tupin F,et al.A new statistical model for Markovian classification of urban areas in high-resolution SAR images.IEEE Transactions on Geoscience & Remote Sensing,2004,42(10):2046-2057.
[17] Touzi R,Lopes A,Bruniquel J,et al.Coherence estimation for SAR imagery.IEEE Transactions on Geoscience & Remote Sensing,1999,37(1):135-149.
[18] Askne J,Hagberg J O.Potential of interferometric SAR for classification of land surfaces.In:Proceedings of the IEEE International Geoscience and Remote Sensing Symposium,Tokyo,1993:985-987.
[19] Shufelt J A.Performance evaluation and analysis of monocular building extraction from aerial imagery.IEEE Trans Pattern Analysis & Machine Intelligence,1999,21(4):311-326.