Please wait a minute...
浙江大学学报(工学版)
J4  2013, Vol. 47 Issue (8): 1403-1410    DOI: 10.3785/j.issn.1008-973X.2013.08.012
    
Hyperspectral image classification based on compsite kernels support vector machine
LI Xiao-run1, ZHU Jie-er1, WANG Jing1, ZHAO Liao-ying2
1 College of Electrical Engineering,Zhejiang University,Hangzhou 310027, China； 2. Institute of Computer Application Technology, HangZhou Dianzi University,Hangzhou 310018, China
Download:   PDF(0KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

To improve the utilization of spatial information when classifying hyperspectral images, this paper proposes a composite kernel SVM algorithm combining spatial and spectral information. First, the hyperspectral image was classified into a map using conventional SVM. The spatial-contextual features were then extracted based on the classified map, and combined with spectral information to construct a composite kernel SVM for classification. The spatial-contextual features were extracted again and the composite kernel SVM classified the image iteratively. The process was repeated 10 times and a proper one was chosen as the last outcome. The results show that the method increases the overall accuracy by around 10%, compared with conventional SVM. In addition, the method also demands much less training samples than usual SVM.

Published: 01 August 2013
CLC:  TP 751.1  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Cite this article:

LI Xiao-run, ZHU Jie-er, WANG Jing, ZHAO Liao-ying. Hyperspectral image classification based on compsite kernels support vector machine. J4, 2013, 47(8): 1403-1410.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2013.08.012     OR     http://www.zjujournals.com/eng/Y2013/V47/I8/1403


组合核支持向量机高光谱图像分类

为了提高高光谱遥感图像分类中空间信息的利用率,提出一种将空间邻域信息和光谱信息结合的组合核支持向量机(SVM)学习算法.用SVM进行预分类,从分类结果图提取各像素的空间邻域特征,与光谱特征结合构造组合核SVM进行分类,并再次提取空间邻域特征进行多次空-谱信息组合核SVM迭代分类,如此迭代10次,从中选择合适的结果作为最终输出.结果表明,该方法对传统支持向量机的分类精度提升幅度可达10%左右.同时,与其他组合核支持向量机相比,该算法用更少的训练样本获得了更高分类精度.

［1］ WANG Xiang-tao, FENG Yan. New method based on support vector machine in classification for hyperspectral data ［C］∥ 2008 International Symposium on Computational Intelligence and Design. Wuhan: ［s. n.］, 2008: 76-80.

［2］ MELGANI F, BRUZZONE L. Classification of hyperspectral remote sensing image with support vector machines ［J］. IEEE Transactions on Geoscience and Remote Sensing, 2004, 42(8): 1778-1790.

［3］ 高恒振,万建伟,粘永健,等.组合核函数支持向量机高光谱图像融合分类［J］.光学精密工程,2011,19(4): 878-883.

GAO Heng-zhen, WAN Jian-wei, NIAN Yong-jian, et al. Fusion classification of hyperspectral image by composite kernels support vector machine ［J］. Optics and Precision Engineering, 2011, 19(4): 878-883.

［4］ DUAN Shan. Mathematical morphology and its Application research in remote sensing image processing ［D］. Wuhan: Wuhan Univesity, 2004.

［5］ 陈君颖,田庆久.高分辨率遥感植被分类研究［J］.遥感学报,2007, 11(2): 221-227.

CHEN Jun-ying, TIAN Qing-jiu. Vegetation classification based on high-resolution satellite image ［J］. Journal of Remote Sensing, 2007, 11(2): 221-227.

［6］ TAN Kun, DU Pei-jun. Combined multi-kernel support vector machine and wavelet analysis for hyperspectral remote sensing image classification ［J］. Chinese Optics Letters, 2011, 9(1): (011003-1)-(011003-4).

［7］ BERNARD K, TARABALKA Y, ANGULO J, et al. Spectral-Spatial classification of hyperspectral data based on a stochastic minimum spanning forest approach ［J］. IEEE Transactions on Geoscience and Remote Sensing, 2012, 21(4): 2008-2021.

［8］ LI Cheng-hsuan, KUO Bor-chen, LIN Chin-teng, et al. A spatial-contextual support vector machine for remotely sensed image classification ［J］. IEEE Transactions on Geoscience and Remote Sensing, 2012, 50(3): 784-799.

［9］ BOSER B E, GUYON I M, VAPNIK V N. A training algorithm for optimal margin classifiers ［C］∥ Proc.5th Annu. Workshop Comput.Learn.Theory. ［s. l.］: ［s. n.］,1992: 144-152.

［10］ DU Qian, YANG He. Similarity-based unsupervised band selection for hyperspectral image analysis ［J］. IEEE Geoscience and Remote Sensing Letters, 2008, 5(4): 564-568.

［11］ BIOUCAS-DIAS J M, NASCIMENTO J M P. Hyperspectral Subspace Identification ［J］. IEEE Geoscience and Remote Sensing Letters, 2008, 46(8): 564-568.

［12］ CAMPS-VALLS G, GOMEZ-CHOVA L, MUNOZ-MARI J. Composite kernels for hyperspectral Image classication ［J］. IEEE Geoscience and Remote Sensing Letters, 2006, 3(1): 93-97.
[1] MA Li-gang, ZHANG Le-ping, DENG Jing-song, WANG Ya-jie, WANG Ke. Land use classification using ZY1-“02C” remote sensing images[J]. J4, 2013, 47(8): 1508-1516.
[2] LAI Xiao-bo, ZHU Shi-qiang, FANG Chun-jie. A three-dimensional reconstruction algorithm for complex
background image and its medical applications[J]. J4, 2012, 46(11): 2061-2067.
[3] CHE Hong-kun, LV Fu-zai, XIANG Zhan-qin. Time-frequency optimal feature extraction method
based on SFFS algorithm for defects recognition[J]. J4, 2011, 45(12): 2235-2239.
[4] HAN Ning, ZHANG Xiu-Yang, WANG Xiao-Meng, CHEN Li-Su, WANG Ke. Identification of distributional information Torreya Grandis Merrlllii
using high resolution imagery[J]. J4, 2010, 44(3): 420-425.