基于K-AP算法的高光谱图像波段选择方法
|
摘要
在高光谱图像分析领域中,波段选择是一种能有效减少高光谱图像维度的方法。K类仿射传播算法是一种高效的聚类算法,已成功地应用于人脸识别和数据分析等领域,但在高光谱图像分析领域还少有成功的应用。提出将K-AP算法应用于高光谱图像波段选择,对高光谱图像进行有效的数据压缩。针对K-AP算法的特点,基于KullbackLeibler散度定义了新的相似度矩阵,对波段进行度量,再使用K-AP算法进行聚类,选择最有代表性的波段。实验结果表明,与常用的波段选择方法相比,所提出的方法有更好的表现。
In hyperspectral image analysis, band selection is an effective method to reduce the dimension of hyperspectral images. K-Affinity Propagation(K-AP)algorithm is a highly efficient clustering algorithm, which is mainly used in face recognition and data analysis. However, it has not been applied in hyperspectral image analysis. In this paper, the K-AP algorithm is applied to the band selection for hyperspectral images that can be effective for data compression. According to the characteristics of K-AP algorithm, a new similarity matrix is defined based on Kullback-Leibler divergence to measure the band similarity. The K-AP algorithm is used to cluster and select the most representative bands. The experimental results show that the proposed approach generally can get better performance compared with other popular band selection methods.
In hyperspectral image analysis, band selection is an effective method to reduce the dimension of hyperspectral images. K-Affinity Propagation(K-AP)algorithm is a highly efficient clustering algorithm, which is mainly used in face recognition and data analysis. However, it has not been applied in hyperspectral image analysis. In this paper, the K-AP algorithm is applied to the band selection for hyperspectral images that can be effective for data compression. According to the characteristics of K-AP algorithm, a new similarity matrix is defined based on Kullback-Leibler divergence to measure the band similarity. The K-AP algorithm is used to cluster and select the most representative bands. The experimental results show that the proposed approach generally can get better performance compared with other popular band selection methods.
引文
[1] Jia S,Tang G,Zhu J,et al.A novel ranking-based clustering approach for hyperspectral band selection[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(1):88-102.
[2] Zhang M,Ma J,Gong M.Unsupervised hyperspectral band selection by fuzzy clustering with particle swarm optimization[J].IEEE Geoscience and Remote Sensing Letters,2017,14(5):773-777.
[3] Sun W,Jiang M,Li W,et al.A symmetric sparse representation based band selection method for hyperspectral imagery classification[J].Remote Sensing,2016,8(12):238.
[4] Cao Xianghai,Xiong Tao,Jiao Licheng.Supervised band selection using local spatial information for hyperspectral image[J].IEEE Geoscience and Remote Sensing Letters,2016,13(3):1-5.
[5] Yang H,Du Q,Su H,et al.An efficient method for supervised hyperspectral band selection[J].IEEE Geoscience and Remote Sensing Letters,2011,8(1):138-142.
[6] Taskin G,Kaya H,Bruzzone L.Feature selection based on high dimensional model representation for hyperspectral images[J].IEEE Transactions on Image Processing,2017,26(6):2918-2928.
[7] Bai X,Guo Z,Wang Y,et al.Semisupervised hyper-spectral band selection via spectral-spatial hyper-graph model[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2015,8(6):2774-2783.
[8] Chang C I,Wang S.Constrained band selection for hyperspectral imagery[J].IEEE Transactions on Geoscience&Remote Sensing,2006,44(6):1575-1585.
[9] Chang C,Du Q,Sun T,et al.A joint band prioritization and band decorrelation approach to band selection for hyperspectral image classification[J].IEEE Transactions on Geoscience and Remote Sensing,1999,37(6):2631.
[10] Martínez-Usómartinez-Uso A,Pla F,Sotoca J M,et al.Clustering-based hyperspectral band selection using information measures[J].IEEE Transactions on Geoscience and Remote Sensing,2007,45(12):4158-4171.
[11] MartínezusóA,Pla F,Sotoca J M,et al.Comparison of unsupervised band selection methods for hyperspectral imaging[C]//Iberian Conference on Pattern Recognition and Image Analysis,2007:30-38.
[12] Zhang X,Wang W,N?rvag K,et al.K-AP:generating specified K clusters by efficient affinity propagation[C]//IEEE International Conference on Data Mining,2011:1187-1192.
[13] Brusco M J,K?hn H F.Comment on “Clustering by passing messages between data points”[J].Science,2008,319:726.
[14] Qian Y,Yao F,Jia S.Band selection for hyperspectral imagery using affinity propagation[J].IET Computer Vision,2009,3(4):213-222.
[15] Chang C,Lin C J,LIBSVM:a library for support vector machines[EB/OL].(2001).http://www.csie.ntu.edu.tw/~cjlin/libsvm.
[16] Fabio R,Giorgio F.Support vector machines for remote sensing image classification[J].Proceedings of the SPIE,2001,4170:160-166.
[17] Huang C,Davis L S,Townshend J R G.An assessment of support vector machines for land cover classification[J].International Journal of Remote Sensing,2002,23(4):725-749.
[18] Melgani F,Bruzzone L.Classification of hyperspectral remote sensing images with support vector machines[J].IEEE Transactions on Geoscience&Remote Sensing,2004,42(8):1778-1790.
[19] Blanzieri E,Melgani F.Nearest neighbor classification of remote sensing images with the maximal margin principle[J].IEEE Transactions on Geoscience&Remote Sensing,2008,46(6):1804-1811.
[20] Chang C,Du Q,Sun T L,et al.A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification[J].IEEE Transactions on Geoscience&Remote Sensing,1999,37(6):2631-2641.
[21] Chen Y,Nasrabadi N M,Tran T D.Sparse representation for target detection in hyperspectral imagery[J].IEEE Journal of Selected Topics in Signal Processing,2011,5(3):629-640.
[22] Aharon M,Elad M,Bruckstein A.K-SVD:an algorithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322.
[2] Zhang M,Ma J,Gong M.Unsupervised hyperspectral band selection by fuzzy clustering with particle swarm optimization[J].IEEE Geoscience and Remote Sensing Letters,2017,14(5):773-777.
[3] Sun W,Jiang M,Li W,et al.A symmetric sparse representation based band selection method for hyperspectral imagery classification[J].Remote Sensing,2016,8(12):238.
[4] Cao Xianghai,Xiong Tao,Jiao Licheng.Supervised band selection using local spatial information for hyperspectral image[J].IEEE Geoscience and Remote Sensing Letters,2016,13(3):1-5.
[5] Yang H,Du Q,Su H,et al.An efficient method for supervised hyperspectral band selection[J].IEEE Geoscience and Remote Sensing Letters,2011,8(1):138-142.
[6] Taskin G,Kaya H,Bruzzone L.Feature selection based on high dimensional model representation for hyperspectral images[J].IEEE Transactions on Image Processing,2017,26(6):2918-2928.
[7] Bai X,Guo Z,Wang Y,et al.Semisupervised hyper-spectral band selection via spectral-spatial hyper-graph model[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2015,8(6):2774-2783.
[8] Chang C I,Wang S.Constrained band selection for hyperspectral imagery[J].IEEE Transactions on Geoscience&Remote Sensing,2006,44(6):1575-1585.
[9] Chang C,Du Q,Sun T,et al.A joint band prioritization and band decorrelation approach to band selection for hyperspectral image classification[J].IEEE Transactions on Geoscience and Remote Sensing,1999,37(6):2631.
[10] Martínez-Usómartinez-Uso A,Pla F,Sotoca J M,et al.Clustering-based hyperspectral band selection using information measures[J].IEEE Transactions on Geoscience and Remote Sensing,2007,45(12):4158-4171.
[11] MartínezusóA,Pla F,Sotoca J M,et al.Comparison of unsupervised band selection methods for hyperspectral imaging[C]//Iberian Conference on Pattern Recognition and Image Analysis,2007:30-38.
[12] Zhang X,Wang W,N?rvag K,et al.K-AP:generating specified K clusters by efficient affinity propagation[C]//IEEE International Conference on Data Mining,2011:1187-1192.
[13] Brusco M J,K?hn H F.Comment on “Clustering by passing messages between data points”[J].Science,2008,319:726.
[14] Qian Y,Yao F,Jia S.Band selection for hyperspectral imagery using affinity propagation[J].IET Computer Vision,2009,3(4):213-222.
[15] Chang C,Lin C J,LIBSVM:a library for support vector machines[EB/OL].(2001).http://www.csie.ntu.edu.tw/~cjlin/libsvm.
[16] Fabio R,Giorgio F.Support vector machines for remote sensing image classification[J].Proceedings of the SPIE,2001,4170:160-166.
[17] Huang C,Davis L S,Townshend J R G.An assessment of support vector machines for land cover classification[J].International Journal of Remote Sensing,2002,23(4):725-749.
[18] Melgani F,Bruzzone L.Classification of hyperspectral remote sensing images with support vector machines[J].IEEE Transactions on Geoscience&Remote Sensing,2004,42(8):1778-1790.
[19] Blanzieri E,Melgani F.Nearest neighbor classification of remote sensing images with the maximal margin principle[J].IEEE Transactions on Geoscience&Remote Sensing,2008,46(6):1804-1811.
[20] Chang C,Du Q,Sun T L,et al.A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification[J].IEEE Transactions on Geoscience&Remote Sensing,1999,37(6):2631-2641.
[21] Chen Y,Nasrabadi N M,Tran T D.Sparse representation for target detection in hyperspectral imagery[J].IEEE Journal of Selected Topics in Signal Processing,2011,5(3):629-640.
[22] Aharon M,Elad M,Bruckstein A.K-SVD:an algorithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322.