基于改进遗传算法的高光谱图像波段选择

doi:10.7523/j.issn.2095-6134.2009.6.010

中国科学院大学学报 ›› 2009, Vol. 26 ›› Issue (6): 795-802.DOI: 10.7523/j.issn.2095-6134.2009.6.010

基于改进遗传算法的高光谱图像波段选择

赵冬^1,2, 赵光恒¹

1. 中国科学院光电研究院,北京 10019
2. 中国科学院研究生院,北京 100049

收稿日期:2009-04-10 修回日期:2009-06-04 发布日期:2009-11-15
通讯作者: 赵冬

Band selection of hyperspectral image based on improved genetic algorithm

ZHAO Dong^1,2, ZHAO Guang-Heng¹

1. Academy of Optoelectronics, Chinese Academy of Sciences, Beijing 100190, China;
2. Graduate University of the Chinese Academy of Sciences, Beijing 100049, China

Received:2009-04-10 Revised:2009-06-04 Published:2009-11-15

摘要/Abstract

摘要：

在对地观测领域,高光谱图像得到了广泛应用,但存在数据量大、波段间相关性高等问题. 针对以上问题分析了已有的波段选择方法,提出了基于信息量及类间可分离性准则的遗传算法对高光谱图像进行波段选择:构造波段互相关系数矩阵进行子空间划分;利用联合熵作为组合信息量的标准,Bhattacharyya距离作为类间可分离性标准,构造遗传算法的适应度方程,改进了遗传算法中的选择算子. 最后用AVIRIS图像对提出的算法进行试验,并利用最大似然分类法对最优波段组合进行分类,总体分类精度达到94.24%,Kappa系数达到0.94.

关键词: 高光谱图像, 波段选择, 遗传算法, 联合熵, Bhattacharyya距离

Abstract:

Hyperspectral images have been widely used in earth observation. However, there are some problems such as huge amount of data and high correlation between bands. The existing band selection algorithms have been analyzed in this article regarding to the problems mentioned above. An application of genetic algorithm based on image information content and between-class separability criteria was proposed to band selection of hyperspectral image. The correlation coefficient matrix was formed for subspace partition. The fitness functions of genetic algorithm were reconstructed by using joint entropy as criterion of information content and Bhattacharya distance as between-class separability. The selection operator in genetic algorithm was improved. Finally, the improved algorithm was tested with AVIRIS image data. The maximum likelihood classification method was implemented to classify the selected optimal band combinations. The classification results illustrate that the total classification accuracy percentage of the chosen band image is 94.24% and Kappa coefficient 0.94.

Key words: hyperspectral image, band selection, genetic algorithm, joint entropy, Bhattacharyya distance

中图分类号:

TP751.1

赵冬, 赵光恒. 基于改进遗传算法的高光谱图像波段选择[J]. 中国科学院大学学报, 2009, 26(6): 795-802.

ZHAO Dong, ZHAO Guang-Heng. Band selection of hyperspectral image based on improved genetic algorithm[J]. , 2009, 26(6): 795-802.

参考文献

[1] Zhao C H, Chen W H, Yang L.Research advances and analysis of hyperspectral remote sensing image band selection
[J].Journal of Nature Science of Heilongjiang University,2007,24(5):592-602 (in Chinese). 赵春晖,陈万海,杨雷.高光谱遥感图像最优波段选择方法的研究进展与分析
[J].黑龙江大学自然科学学报,2007,24(5):592-602.

[2] Holland J H. Adaptation in natural and artificial systems
[M].Ann Arbor:University of Michigan Press,1975.

[3] Goldberg D E.Genetic algorithms in search,optimization and machine learning
[M].Addison Wesley,1989.

[4] Farzam M,Beheshti S,Raahemifar K.Calculation of abundance factors in hyperspectral imaging using genetic algorithm //Canadian Conference on Electrical and Computer Engineering. Niagara,2008:837-842.

[5] Ma J P,Zheng Z B,Tong Q X,et al. An application on genetic algorithms on band selection for hyperspectral image classification //Proceedings of the Second International Conference on Machine Learning and Cybernetics.Xian,2003: 2810-2813.

[6] Gancarskia P, Blansche A, Wania A.Comparison between two co-evolutionary feature weighting algorithms in clustering
[J].Pattern Recognition,2008,41: 983-994.

[7] Chaichoke Vaiphasa,Andrew K Skidmore,Willem F de Boer,et al.A hyperspectral band selector for plant species discrimination
[J].Journal of Photogrammetry & Remote Sensing,2007,62:225-235.

[8] Yu S X,Steve De Backer,Paul Scheunders.Genetic feature selection combined with composite fuzzy nearest neighbor classification for hyperspectral satellite imagery
[J].Pattern Recognition Letters,2002,23:183-190.

[9] Zhuo L,Zheng J,Wang F,et al.A genetic algorithm based wrapper feature selection method for classification of hyperspectral images using SVM //The International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences.Beijing,2008,XXXVII(B7):397-402.

[10] Su H J,Sheng Y H,Du P J.Study on auto-subspace partition for band selection of hyperspectral image
[J].Geo-information Science,2007,4(9):123-128 (in Chinese). 苏红军,盛业华,杜培军.自动子空间划分在高光谱影像波段选择中的应用
[J].地球信息科学,2007,4(9):123-128.

[11] Lu J,Peng M,Lu X.Relativity of remote sensing images and entropy calculation
[J].Geomatics and Information Science of Wuhan University,2006,6 (31):476-480 (in Chinese). 卢健,彭馒,卢听.遥感图像相关性及其墒计算
[J].武汉大学学报(信息科学版),2006,6(31):476-480.

[12] Ren Q S,Ye Z X,Zeng J.Analysis of genetic operators
[J].Acta Electronica Sinica,2000,28(5):113-114 (in Chinese). 任庆生,叶中行,曾进,等.遗传算法中常用算子的分析
[J].电子学报,2000,28(5):113-114.

基于改进遗传算法的高光谱图像波段选择

Band selection of hyperspectral image based on improved genetic algorithm

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 12

编辑推荐

Metrics

本文评价

访问统计

联系我们

[1]	明卫鹏, 马广彬, 章文毅. 基于基因表达式编程的多星成像任务规划算法[J]. 中国科学院大学学报, 2020, 37(4): 532-538.
[2]	宋晗, 杨炜暾, 耿修瑞, 赵永超. 基于卷积神经网络与主动学习的高光谱图像分类[J]. 中国科学院大学学报, 2020, 37(2): 169-176.
[3]	杨随心, 耿修瑞, 杨炜暾, 赵永超, 卢晓军. 一种基于谱聚类算法的高光谱遥感图像分类方法[J]. 中国科学院大学学报, 2019, 36(2): 267-274.
[4]	桑睿, 吴杰, 许华, 郭强. 基于改进的遗传算法的后非线性盲源分离[J]. 中国科学院大学学报, 2012, 29(5): 707-713.
[5]	陈刚, 王宏琦, 孙显. 基于核函数原型和自适应遗传算法的 SVM模型选择方法[J]. 中国科学院大学学报, 2012, 29(1): 62-69.
[6]	刘丽峰, 张树清. 人工免疫算法规划复杂环境下三维飞行航迹[J]. 中国科学院大学学报, 2011, 28(6): 746-751.
[7]	王翠平, 郭立, 王昱洁, 陈运必. 音频DCT系数分布函数的建模[J]. 中国科学院大学学报, 2011, 28(6): 752-758.
[8]	徐菡;　张敏洪. 大规模逆向物流网络非线性优化模型的研究[J]. 中国科学院大学学报, 2007, 24(6): 749-755.
[9]	罗德林; 沈春林; 王彪 ; 吴文海. 基于混合自适应遗传算法的协同多目标攻击空战决策（英文）[J]. 中国科学院大学学报, 2006, 23(3): 382-389.
[10]	贵刚, 于军, 苏丽杰, 聂义勇. 压力铸造充型过程多工艺参数的优化选择[J]. 中国科学院大学学报, 2004, 21(4): 532-537.
[11]	甘国辉, 刘长岐, 杨丹. 基于遗传算法的农业用地结构优化研究——以北京市通州区为例[J]. 中国科学院大学学报, 2004, 21(1): 50-55.
[12]	孔祥蕾, 张先燚, 罗晓琳, 李海洋. 一种引入强制变异的改进遗传算法[J]. 中国科学院大学学报, 2003, 20(3): 316-320.