[1]舒 速,杨 明,赵振凯.基于分水岭的高光谱图像分类方法[J].南京师大学报(自然科学版),2015,38(01):91.
　Shu Su,Yang Ming,Zhao Zhenkai.Hyperspectral Image Classification Method Based on Watershed[J].Journal of Nanjing Normal University(Natural Science Edition),2015,38(01):91.
点击复制

基于分水岭的高光谱图像分类方法()
分享到：

《南京师大学报》（自然科学版）[ISSN:1001-4616/CN:32-1239/N]

卷:
第38卷
期数:
2015年01期
页码:
91
栏目:
计算机科学
出版日期:
2015-06-30

文章信息/Info

Title:
Hyperspectral Image Classification Method Based on Watershed
作者:
舒 速; 杨 明; 赵振凯
南京师范大学计算机科学与技术学院,江苏 南京 210023
Author(s):
Shu Su; Yang Ming; Zhao Zhenkai
School of Computer Science and Technology,Nanjing Normal University,Nanjing 210023,China
关键词:
分水岭; 高光谱; 图像分类
Keywords:
watershed; hyperspectral; image classification
分类号:
TP751
文献标志码:
A
摘要:
近年来,高光谱图像的分类受到了广泛的关注,许多机器学习的方法都在高光谱图像上得到了应用,如SVM、神经网络、决策树等. 为了提高分类精度,通常将图像的光谱信息与空间信息结合起来进行分类. 本文提出了如何利用分水岭分割得到的空间信息来得到更精确的分类结果. 首先利用分水岭得到图像区域信息,然后根据一个区域中是否含有训练样本而采取不同的策略得到该区域中所有点的类别. 本文在两幅图像上分别用SVM和联合稀疏表示对该方法的有效性进行验证,实验结果表明该方法优于其他一些同类方法.
Abstract:
Hyperspectral image classification has attracted a great deal of attention. Many machine learning methods have been applied in hyperspectral image classification,such as SVM,neural network and decision tree,etc. In order to increase classification performances,people usually integrate of spatial information into the classification process. In this paper,we will present how to use spatial information obtained by watershed segmentation to obtain a more accurate classification results. We obtained the regional imformation by watershed segment and then adopted different strategies to get the category of the points in an area according to the area whether it contains the training sample. SVM and the joint sparse representation are used on two images to verify the effectiveness of the proposed method. Experimental results show that our algorithm outperforming some other similar methods.

参考文献/References:

[1] Chen C H,Peter Ho P G. Statistical pattern recognition in remote sensing[J]. Pattern Recognition,2008,41(9):2 731-2 741.
[2]Bali N,Mohammad-Djafari A. Bayesian approach with hidden Markov modeling and mean field a approximation for hyperspectral data analysis[J]. IEEE Trans on Image Process,2008,17(2):217-225.
[3]Hernańdez-Espinosa C,Fernańdez-Redondo M,Torres-Sospedra J. Some experiments with ensembles of neural networks for classification of hyperspectral images[J]. ISNN,2004(1):912-917.
[4]Zhou H,Mao Z,Wang D. Classification of coastal areas by airborne hyperspectral image[J]. Proceedings of SPIE,2005,5 832:471-476.
[5]Pal M,Mather M. Support vector machines for classification in remote sensing[J]. Int J Remote Sens,2005,26(5):1 007-1 011.
[6]Heikkinen V,Tokola T,Parkkinen J,et al. Simulated multispectral imagery for tree species classification using support vector machines[J]. IEEE Trans on Geosci Remote Sens,2010,48(3):1 355-1 364.
[7]Giorgos M,Jungho I,Caesar O. Support vector machines in remote sensing:A review[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2011,66:247-259.
[8]Fauvel M. Spectral and spatial methods for the classification of urban remote sensing data[D]. Reykejavik,Iceland:France and Faculty of Electrical and Computer Engineering,University of Iceland,2007.
[9]Farag A,Mohamed R,El-Baz A. A unified framework for map estimation in remote sensing image segmentation[J]. IEEE Trans on Geosci Remote Sens,2005,43(7):1 617-1 634.
[10]Camps-Valls G,Gomez-Chova L,Mu(～overn)oz-Marí J,et al. Composite kernels for hyperspectral image classification[J]. IEEE Geosci Remote Sens Lett,2006,3(1):93-97.
[11]Tarabalka Y,Chanussot J,Benediktsson J A. Segmentation and classification of hyperspectral images using watershed transformation[J]. Pattern Recognition,2010,43(2):2 367-2 379.
[12]Vincent L,Soille P. Watersheds in digital spaces:an efficient algorithm based on immersion simulations[J]. IEEE Trans Pattern Anal Mach Intell,1991,13(6):583-598.
[13]Meyer F. Topographic distance and watershed lines[J]. Signal Process,1994,38(1):113-125.
[14]Stoev S. Rafsi—a fast watershed algorithm based on rainfalling simulation[C]//Proceedings of the Eighth International Conference on Computer Graphics,Visualization,and Interactive Digital Media. Plzen,Gzech Republic:WSCG,2000.
[15]Osma-Ruiz V,Godino-Llorente J I,Saénz-Lechoń N,et al. An improved watershed algorithm based on efficient computation of shortest paths[J]. Pattern Recognition,2007,40(3):1 078-1 090.
[16]Chen Y,Nasrabadi N M,Trac D,et al. Hyperspectral image classification using dictionary-based sparse representation[J]. IEEE Trans on Geosci Remote Sens,2011,49(10):3 973-3 985.

备注/Memo

备注/Memo:
收稿日期:2014-09-30.
基金项目:国家自然科学基金重点、面上(61432008、61272222).
通讯联系人:舒速,硕士,研究方向:机器学习、模式识别. E-mail:shusu510@126.com
更新日期/Last Update: 2015-03-30