[1]田洪晨,王立国,赵亮,等.结合波段选择的差分进化高光谱图像分类[J].应用科技,2019,46(05):45-50.[doi:10.11991/yykj.201811019]
　TIAN Hongchen,WANG Liguo,ZHAO Liang,et al.Differential evolution hyperspectral image classification combined with band selection[J].Applied science and technology,2019,46(05):45-50.[doi:10.11991/yykj.201811019]
点击复制

结合波段选择的差分进化高光谱图像分类(/HTML)
分享到：

《应用科技》[ISSN:1009-671X/CN:23-1191/U]

卷:
第46卷
期数:
2019年05期
页码:
45-50
栏目:
现代电子技术
出版日期:
2019-09-05

文章信息/Info

Title:
Differential evolution hyperspectral image classification combined with band selection
作者:
田洪晨;  王立国;  赵亮;  陈春雨
哈尔滨工程大学 信息与通信工程学院, 黑龙江 哈尔滨 150001
Author(s):
TIAN Hongchen;  WANG Liguo;  ZHAO Liang;  CHEN Chunyu
College of Information and Communications Engineering, Harbin Engineering University, Harbin 150001, China
关键词:
高光谱; 高光谱图像分类; 半监督分类; 波段选择; 差分进化; 遗传算法; 支持向量机; 机器学习
Keywords:
hyperspectral; hyperspectral image classification; semi-supervised classification; band selection; differential evolution; genetic algorithm; support vector machine; machine learning
分类号:
TP753;TP751
DOI:
10.11991/yykj.201811019
文献标志码:
A
摘要:
高光谱图像具有数据维数高、有标签样本少的特点，影响了现有分类方法的效果。针对这一情况，提出一种结合波段选择的半监督分类算法。该算法首先通过波段选择方法，去除高光谱图像中的冗余信息，进而降低复杂度和提高泛化能力；然后通过差分进化算法交叉变异无标记样本，选取置信度高的样本扩充入标记样本群以提高分类精度。实验结果表明，该算法能够有效地提升在标记样本有限的情况下分类器的分类精度与分类速度。
Abstract:
Hyperspectral images have the characteristics of high dimension of data and few labeled samples, which affects the effect of existing classification methods. Aiming at this situation, we propose a semi-supervised classification algorithm combining band selection. Firstly, the band selection method is used to remove the redundant information in the hyperspectral image, reducing the complexity and improving the generalization ability. Then, the differential evolution algorithm is used to cross-mutates the unlabeled samples, and expand the labeled sample group by selecting the samples with high confidence. The experimental results show that the proposed method can effectively improve the classification accuracy and speed of the classifier under the condition of limited label samples.

参考文献/References:

[1] 童庆禧, 张兵, 郑兰芬. 高光谱遥感[M]. 北京:高等教育出版社, 2006.
[2] 朱艳, 刘晓莉, 杨哲海. 高光谱数据的降维及Tabu搜索算法的应用[J]. 测绘科学技术学报, 2007, 24(1):22-25, 29
[3] ZHONG Yanfei, MA Ailong, ONG Y S, et al. Computational intelligence in optical remote sensing image processing[J]. Applied soft computing, 2018, 64:75-93.
[4] CHENG Gong, HAN Junwei. A survey on object detection in optical remote sensing images[J]. ISPRS journal of photogrammetry and remote sensing, 2016, 117:11-28.
[5] 杜培军, 夏俊士, 薛朝辉, 等. 高光谱遥感影像分类研究进展[J]. 遥感学报, 2016, 20(2):236-256
[6] WANG Jianyu, XUE Yongqi, RONG Shu, et al. Airborne hyperspectral and infrared remote sensing technology and application[C]//Proceedings of 2006 Joint 31st International Conference on Infrared Millimeter Waves and 14th International Conference on Teraherz Electronics. Shanghai, China, 2006.
[7] HUNG C C, KULKARNI S, KUO B C. A new weighted fuzzy c-means clustering algorithm for remotely sensed image classification[J]. IEEE journal of selected topics in signal processing, 2011, 5(3):543-553.
[8] VILLA A, BENEDIKTSSON J A, CHANUSSOT J, et al. Hyperspectral image classification with independent component discriminant analysis[J]. IEEE transactions on geoscience and remote sensing, 2011, 49(12):4865-4876.
[9] 周千琪. 基于稀疏表示和半监督主动学习的小样本高光谱数据分类问题研究[D]. 银川:北方民族大学, 2016.
[10] 高飞, 朱福利. 基于样本类别确定度的半监督分类[J]. 北京航空航天大学学报, 2018, 44(9):1941-1951
[11] CORTES C, VAPNIK V. Support vector networks[J]. Machine learning, 1995, 20(3):273-297.
[12] KUMAR M A, GOPAL M. Least squares twin support vector machines for pattern classification[J]. Expert systems with applications, 2009, 36(4):7535-7543.
[13] 陈国良, 王煦法, 庄镇泉, 等. 遗传算法及其应用[M]. 北京:人民邮电出版社, 1996.
[14] 张明阳. 基于进化优化学习的高光谱特征选择与提取[D]. 西安:西安电子科技大学, 2018.
[15] STORN R, PRICE K. Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of global optimization, 1997, 11(4):341-359.
[16] 刘明广. 差异演化算法及其改进[J]. 系统工程, 2005, 23(2):108-111

相似文献/References:

[1]赵春晖,刘凡.基于改进自组织竞争神经网络的高光谱图像分类[J].应用科技,2009,36(08):8.[doi:10.3969/j.issn.1009-671X.2009.08.003]
　ZHAO Chun-hui,LIU Fan.An improved hyperspectral remote sensing image classification method based on SOFM neural network[J].Applied science and technology,2009,36(05):8.[doi:10.3969/j.issn.1009-671X.2009.08.003]
[2]王立国,吴国峰.结合FDA与NMF的高光谱数据解混方法[J].应用科技,2011,38(12):20.[doi:10.3969/j.issn.1009-671X.2011.12.005]
　WANG Liguo,WU Guofeng.Combining Fisher discriminant analysis with nonnegative matrix factorization by perspectral data unmixing[J].Applied science and technology,2011,38(05):20.[doi:10.3969/j.issn.1009-671X.2011.12.005]
[3]房森,焦淑红.分组寻优的多端元高光谱图像解混方法[J].应用科技,2019,46(06):20.[doi:10.11991/yykj.201903005]
　FANG Sen,JIAO Shuhong.Multi-endmember hysperspectral image unmixing based on group optimization[J].Applied science and technology,2019,46(05):20.[doi:10.11991/yykj.201903005]
[4]池辛格,王立国.结合局部全局一致性和支持向量机的半监督分类方法[J].应用科技,2021,48(1):48.[doi:10.11991/yykj.202009014]
　CHI Xinge,WANG Liguo.Semi-supervised classification method combining local and global consistency and support vector machine[J].Applied science and technology,2021,48(05):48.[doi:10.11991/yykj.202009014]

备注/Memo

备注/Memo:
收稿日期:2018-12-04。
基金项目:国家自然基金项目（61675051）
作者简介:田洪晨,女,硕士研究生;王立国,男,教授,博士生导师
通讯作者:王立国,E-mail:wangliguo@hrbeu.edu.cn
更新日期/Last Update: 2019-08-29