扫 描 看 全 文
引用
康孝岩,张爱武,胡少兴,肖青,柴沙驼.2020.高光谱图像的JM变换自适应降维.遥感学报,24(1): 67-75
KANG Xiaoyan,ZHANG Aiwu,HU Shaoxing,XIAO Qing,CHAI Shatuo. 2020. Hyperspectral images adaptive dimensionality reduction optimized byJM transform. Journal of Remote Sensing(Chinese). 24(1): 67-75
在无需先验标签样本的情况下，非监督降维可以有效简化高光谱图像的特征空间，避免目标分类中的霍夫效应。本文提出JM非线性变换优化的自适应降维模型来研究面向图像目标分类的高光谱波段选择问题。该方法考虑波段的信息量和独立性等两个重要因子，针对其测度方法的差异性问题，引入JM变换函数进行规范化优化。选用线阵高光谱和面阵显微光谱等两个图像数据集，在,k,最邻近和随机森林分类器下，进行了多组监督分类实验，结果表明，在Kappa系数、总体分类精度和平均分类精度上，本文方法均优于3种非监督方法MABS、InfFS和LSFS。说明本文提出的JM变换的自适应降维模型能够有效降低特征维度，满足高光谱图像分类的高精度要求。
Hyperspectral remote sensing images, which collect rich spectral and spatial information of observed targets, usually contain dozens to hundreds of narrow bands with wavelengths ranging from the visible light region to the near-infrared spectra. With such an abundant number of spectral features, hyperspectral images (HSI) allow us to distinguish different of objects or targets by rule and line. Unfortunately, such high-dimensionality data pose a challenge in data transmission, storage, and processing. Specifically, those HSIs with high redundancy information and strong correlation are prone to a Hughes phenomenon during the image classification process. Therefore, dimensionality reduction is necessary for targets classification. Moreover, without using prior label samples, unsupervised dimensionality reduction can effectively simplify the HSI feature space, and prevent the Hughes phenomenon in the targets classification.,In this paper, the Jeffries–Matusita (JM) modified adaptive band selection (JM,2,ABS) method is proposed to extract proper features from HSI datasets. Generally speaking, a band that contains many information and demonstrates strong independence is a very important feature that helps unsupervised band selection methods to classify targets. The JM,2,ABS method considers both the information capacity and independence of HSI bands. Given the significant differences in the measurements of a band’s information capacity and its independence, we introduce the JM transform function to normalize the distributions of the information capacity and the independence of HSI data. Thus JM,2,ABS shows that both the information capacity and the independence are equally important in unsupervised dimensionality reduction.,We also compare our proposed JM,2,ABS method against three typical methods, namely, the modified adaptive band selection method, the Laplacian score feature selection method, and the infinite feature selection method. By using random training samples, we perform supervised classification experiments on two kinds of HSI public datasets (linear and planar arrays). The results demonstrate that JM,2,ABS outperforms the other three typical methods in terms of Kappa value, overall classification accuracy, and average classification accuracy. Moreover, under a small number of bands, JM,2,ABS can reach a high and stable level regardless of the different datasets and different classifiers used.,The proposed JM,2,ABS can select the proper features of HSI datasets for their classification application. The JM transform function is a kind of nonlinear distribution function that can standardize variables from different scales to the same. To demonstrate the feasibility of JM transform optimization, we set the same weight for the information content and the independence. In our future work, we will explore the similarities and differences between the information capacity and the independence in dimensionality reduction.
遥感JM变换规范化自适应降维非监督波段选择高光谱图像
remote sensingJeffries-Matusita transformnormalizationadaptive dimensionality reductionunsupervised band selectionhyperspectral image
Bajcsy P andGroves P. 2004. Methodology for hyperspectral band selection. Photogrammetric Engineering and Remote Sensing, 70(7): 793-802 [DOI: 10.14358/PERS.70.7.793http://dx.doi.org/10.14358/PERS.70.7.793]
Cao X H, Wu B, Tao D C and Jiao L C. 2016. Automatic band selection using spatial-structure information and classifier-based clustering. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 9(9): 4352-4360 [DOI: 10.1109/JSTARS.2015.2509461http://dx.doi.org/10.1109/JSTARS.2015.2509461]
He X F, Cai D and Niyogi P. 2005. Laplacian score for feature selection//Proceedings of the 18th International Conference on Neural Information Processing Systems. Vancouver, British Columbia, Canada: MIT Press: 507-514
Ioffe S and Szegedy C. 2015. Batch normalization: accelerating deep network training by reducing internal covariate shift//Proceedings of the 32nd International Conference on Machine Learning. Lille, France: 448-456
Li S J and Qi H R. 2011. Sparse representation based band selection for hyperspectral images//Proceedings of the 18th IEEE International Conference on Image Processing. Brussels, Belgium: IEEE: 2693-2696 [DOI: 10.1109/ICIP.2011.6116223http://dx.doi.org/10.1109/ICIP.2011.6116223]
Liu C H, Zhao C H and Zhang L Y. 2005. A new method of hyperspectral remote sensing image dimensional reduction. Journal of Image and Graphics, 10(2): 218-222
刘春红, 赵春晖, 张凌雁. 2005. 一种新的高光谱遥感图像降维方法. 中国图象图形学报, 10(2): 218-222 [DOI: 10.3969/j.issn.1006-8961.2005.02.015http://dx.doi.org/10.3969/j.issn.1006-8961.2005.02.015]
Matusita K. 1955. Decision rules, based on the distance, for problems of fit, two samples, and estimation. The Annals of Mathematical Statistics, 26(4): 631-640 [DOI: 10.1214/aoms/1177728422http://dx.doi.org/10.1214/aoms/1177728422]
Qin F P, Zhang A W, Wang S M, Meng X G, Hu S X and Sun W D. 2015. Hyperspectral band selection based on spectral clustering and inter-class separability factor. Spectroscopy and Spectral Analysis, 35(5): 1357-1364
秦方普, 张爱武, 王书民, 孟宪刚, 胡少兴, 孙卫东. 2015. 基于谱聚类与类间可分性因子的高光谱波段选择. 光谱学与光谱分析, 35(5): 1357-1364) [DOI: 10.3964/j.issn.1000-0593(201505-1357-08http://dx.doi.org/10.3964/j.issn.1000-0593(2015)05-1357-08]
Richards J A. 2013. Remote Sensing Digital Image Analysis. 5th ed. Berlin, Heidelberg: Springer: 350-352
Roffo G, Melzi S and Cristani M. 2015. Infinite feature selection//IEEE International Conference on Computer Vision. Santiago, Chile: IEEE: 4202-4210 [DOI: 10.1109/ICCV.2015.478http://dx.doi.org/10.1109/ICCV.2015.478]
Roffo G and Melzi S. 2017. Ranking to learn: feature ranking and selection via eigenvector centrality//Appice A, Ceci M, Loglisci C, Masciari E and Raś Z, eds. New Frontiers in Mining Complex Patterns. Berlin Heidelberg: Springer: 19-35 [DOI: 10.1007/978-3-319-61461-8_2http://dx.doi.org/10.1007/978-3-319-61461-8_2]
Shukla U P and Nanda S J. 2018. A Binary Social Spider Optimization algorithm for unsupervised band selection in compressed hyperspectral images. Expert Systems with Applications, 97: 336-356 [DOI: 10.1016/j.eswa.2017.12.034http://dx.doi.org/10.1016/j.eswa.2017.12.034]
Sui C H, Tian Y, Xu Y P and Xie Y. 2015. Unsupervised band selection by integrating the overall accuracy and redundancy. IEEE Geoscience and Remote Sensing Letters, 12(1):185-189 [DOI: 10.1109/LGRS.2014.2331674http://dx.doi.org/10.1109/LGRS.2014.2331674]
Sun W W, Zhang D F, Yang G and Li W Y. 2018. Band selection for hyperspectral imagery based on weighted probabilistic archetypal analysis. Journal of Remote Sensing, 22(1): 110-118
孙伟伟, 张殿发, 杨刚, 李巍岳. 2018. 加权概率原型分析的高光谱影像波段选择. 遥感学报, 22(1): 110-118 [DOI: 10.11834/jrs.20186446http://dx.doi.org/10.11834/jrs.20186446]
Sun W W, Zhang L P, Du B, Li W Y and Lai Y M. 2015. Band selection using improved sparse subspace clustering for hyperspectral imagery classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8(6): 2784-2797 [DOI: 10.1109/JSTARS.2015.2417156http://dx.doi.org/10.1109/JSTARS.2015.2417156]
Swain P H and Davis S M. 1978. Remote Sensing: the Quantitative Approach. New York: McGraw-Hill
Tong Q X, Zhang B and Zhang L F. 2016. Current progress of hyperspectral remote sensing in China. Journal of Remote Sensing, 20(5): 689-707
童庆禧, 张兵, 张立福. 2016. 中国高光谱遥感的前沿进展. 遥感学报, 20(5): 689-707 [DOI: 10.11834/jrs.20166264http://dx.doi.org/10.11834/jrs.20166264]
Tong Q X, Zhang B and Zheng L F. 2006. Hyperspectral Remote Sensing: the Principle, Technology and Application. Beijing: Higher Education Press: 135-136
童庆禧, 张兵, 郑兰芬. 2006. 高光谱遥感: 原理、技术与应用. 北京: 高等教育出版社: 135-136
Wang L G and Wei F J. 2013. Band selection for hyperspectral imagery based on combination of genetic algorithm and ant colony algorithm. Journal of Image and Graphics, 18(2): 235-242
王立国, 魏芳洁. 2013. 结合遗传算法和蚁群算法的高光谱图像波段选择. 中国图象图形学报, 18(2): 235-242
Yu X C, Feng W F, Yang G P and Chen W. 2013. Analysis and Application for Hyperspectral Imagery. Beijing: Science Press: 1-6
余旭初, 冯伍法, 杨国鹏, 陈伟. 2013. 高光谱影像分析与应用. 北京: 科学出版社: 1-6
Zhang A W, Du N, Kang X Y and Guo C F. 2017. Hyperspectral adaptive band selection method through nonlinear transform and information adjacency correlation. Infrared and Laser Engineering, 46(5): 213-221
张爱武, 杜楠, 康孝岩, 郭超凡. 2017. 非线性变换和信息相邻相关的高光谱自适应波段选择. 红外与激光工程, 46(5): 213-221 [DOI: 10.3788/IRLA201746.0538001http://dx.doi.org/10.3788/IRLA201746.0538001]
Zhang K, Hei B Q, Zhou Z and Li S Y. 2018. CNN with coefficient of variation-based dimensionality reduction for hyperspectral remote sensing images classification. Journal of Remote Sensing, 22(1): 87-96
张康, 黑保琴, 周壮, 李盛阳. 2018. 变异系数降维的CNN高光谱遥感图像分类. 遥感学报, 22(1): 87-96 [DOI: 10.11834/jrs.20187075http://dx.doi.org/10.11834/jrs.20187075]
相关作者
相关机构
京公网安备11010802024621
