浏览全部资源
扫码关注微信
收稿：2019-08-13，
修回：2019-10-12，
录用：2019-10-19，
纸质出版：2020-05-16
移动端阅览
目的
2
高光谱图像具有高维度的光谱结构，而且邻近波段之间往往存在大量冗余信息，导致在随机样本选择策略和图像分类过程中出现选择波段算法复杂度较高和不适合小样本的现象。针对该问题，在集成学习算法的基础上，考虑不同波段在高光谱图像分类过程中的作用不同，提出一种融合累积变异比和超限学习机的高光谱图像分类算法。
方法
2
定义波段的累积变异比函数来确定各波段在分类算法的贡献程度。基于累积变异比函数剔除低效波段，并结合空谱特征进行平均分组加权随机选择策略进行数据降维。为了进一步提高算法的泛化能力，对降维后提取的空谱特征进行多次样本重采样，训练得到多个超限学习机弱分类器，再将多个弱分类器的结果通过投票表决法得到最后的分类结果。
结果
2
实验使用Indian Pines、Pavia University scene和Salinas这3种典型的高光谱图像作为实验标准数据集，采用支持向量机（support vector machine，SVM），超限学习机（extreme learning machine，ELM），基于二进制多层Gabor超限学习机（ELM with Gabor，GELM），核函数超限学习机（ELM with kernel，KELM），GELM-CK（GELM with composite kernel），KELM-CK（KELM with composite kernel）和SS-EELM（spatial-spectral and ensemble ELM）为标准检测算法验证本文算法的有效性，在样本比例较小的实验中，本文算法的总体分类精度在3种数据集中分别为98.0%、98.9%和97.9%，比其他算法平均分别高出9.6%和4.7%和4.1%。本文算法耗时在3种数据集中分别为15.2 s、60.4 s和169.4 s。在同类目标空谱特性差异较大的情况下，相比于分类精度较高的KELM-CK和SS-EELM算法减少了算法耗时，提高了总体分类精度；在同类目标空谱特性相近的情况下，相比于其他算法，样本数量的增加对本文算法的耗时影响较小。
结论
2
本文算法通过波段的累积变异比函数优化了平均分组波段选择策略，针对各类地物目标分布较广泛并且同类目标空谱特性差异较大的高光谱数据集，能够有效提取特征光谱维度的差异性，确定参数较少，总体分类效果较好。
Objective
2
Hyperspectral remote sensing has become a promising research field and is applied to various aspects. Hyperspectral image classification has become a key part of hyperspectral image processing. However
high-dimensional data structures bring new challenges for hyperspectral image classification. In particular
problems may occur in the feature extraction and classification process of a hyperspectral image dataset
e.g.
the Hughes phenomenon
because of the unbalance between the high-dimensionality of the data and the limited number of training samples. To improve the accuracy of hyperspectral image classification
we propose a hyperspectral image classification algorithm based on ensemble extreme learning machine (ELM) with cumulative variation quotient
referred to as EELM with cumulative variation quotient (CVQ-EELM).
Method
2
In this study
the coefficient of variation is usually used as the index to show the data dispersion. Compared with the standard deviation
its main advantage is that it is not affected by the measurement scale. In particular
the coefficient of variation takes into account the influence of the average value of the data. The coefficient of variation is improved and applied to the dimensionality reduction of the HIS dataset. First
the cumulative variation functions of the intraclass and the interclass and the cumulative variation quotient are proposed. In actual operation
some pixels may contain multiple ground objects
while the gray values of the intraclass are quite different. Therefore
the cumulative variation function of the interclass and the cumulative variation function of the intraclass should be comprehensively considered to define the cumulative variation quotient function of bands. On the premise of the same band
the quotient of the norm of the interclass' cumulative variation function and the sum of the norm of the intraclass' cumulative variation function is called the cumulative variation quotient of the band. If the cumulative variation quotient of the band is far from 1
it means that the classification effect is better by using this band. If the cumulative variation quotient of band is close to 1
it means that the classification effect is poor by using this band. The inefficient bands are eliminated on the basis of the cumulative variation quotient function. Second
to provide the input information of hyperspectral bands for ELM and considering the strong correlation relationship between neighboring bands
average grouping is performed for the remaining effective bands after eliminating the inefficient bands. A certain number of bands are then selected by the weighted-random-selecting-based approach to reduce the dimension of the hyperspectral image dataset. Specifically
the hyperspectral bands are grouped on average and then the weights of each group are calculated based on the cumulative variation quotient. The bands of each group are selected randomly according to their weights. Finally
the spatial spectral features extracted after dimensionality reduction are sampled repeatedly to train several weak ELM classifiers. The results of several weak classifiers are majority voted to build a strong classifier.
Result
2
Three well-known HIS datasets (Indian Pines
Pavia University scene
and Salinas) are used to verify the effectiveness of the proposed method. SVM (support vector machine)
ELM
GELM (ELM with Gabor)
KELM (ELM with kernel)
GELM-CK (GELM with composite kernel)
KELM-CK (KELM with composite kernel)
and SS-EELM (spatial-spectral and ensemble ELM) serve as the benchmark algorithms to measure the performance of the proposed CVQ-EELM. SVM
ELM
GELM
and KELM methods use only spectral features. Although SVM and KELM introduce the kernel function and increase the computational cost
SVM and KELM have better classification performance than ELM and GELM. GELM-CK
KELM-CK
SS-EELM
and CVQ-EELM methods incorporate the spatial information into the spectral information. The four spatial-spectral-feature-based methods show better classification performance than the four spectral-feature-based methods. Further analysis shows that the classification capacities of KELM-CK
SS-EELM
and CVQ-EELM are better than that of GELM-CK. However
KELM-CK
SS-EELM
and CVQ-EELM are amenable to more time cost than GELM-CK. For example
in three typical HIS datasets
KELM-CK method consumes classification times up to 15.8 s
143 s
and 54.6 s
respectively. Although SS-EELM avoids referencing kernel functions
SS-EELM based on the ensemble extreme learning machines also has a large operation time
equal to 32.4 s
85.5 s
and 171 s. The proposed CVQ-EELM only needs 15.2 s
60.4 s
and 169.4 s to do so. The time-consuming characteristic of the SS-EELM and CVQ-EELM algorithms is related to the number of weak classifiers. Specifically
the greater the number of weak classifiers
the more time-consuming the algorithm will take. Compared with KELM-CK
the time-consuming growth rates of SS-EELM and CVQ-EELM are smaller with an increasing number of samples. When the spatial-spectral features of the same category are quite different
the proposed CVQ-EELM outperforms KELM-CK and SS-EELM. For example
in two typical HIS datasets (Indian Pines and Pavia University scene)
overall accuracy (OA) of the proposed CVQ-EELM is 98.0% and 98.9%
respectively; OA of KELM-CK is 97.8% and 98.8%
respectively; and OA of SS-EELM is 97.2% and 98.6%
respectively. The computational cost of CVQ-EELM is also still lower than that of SS-EELM in two typical HIS datasets. According to the above experimental comparison
the computational cost of CVQ-EELM is similar to that of KELM-CK in the Indian Pines dataset. However
the classification accuracy of CVQ-EELM is higher than that of KELM-CK. Especially in the Pavia University dataset
the computational cost of CVQ-EELM is still low
approximately 2.5 times faster than KELM-CK. Moreover
the classification accuracy of CVQ-EELM is higher than KELM-CK. Therefore
the proposed CVQ-EELM has the best classification performance among all the classification algorithms.
Conclusion
2
The conclusion shows that the proposed algorithm optimizes the band selection strategy of average grouping through the cumulative variation quotient function. For hyperspectral data sets with a wide distribution of various objects and a large difference in spatial-spectral features of similar objects
the characteristics of spectral differences can be extracted effectively. The proposed CVQ-EELM has the advantages of few adjustable parameters and fast training speed. It also outperforms various state-of-the-art hyperspectral image classification counterparts in terms of classification accuracy.
Cao F X, Yang Z J, Ren J C, Jiang M Y and Ling W K. 2017. Linear vs. nonlinear extreme learning machine for spectral-spatial classification of hyperspectral image. Sensors, 17(11):#2603[DOI:10.3390/s17112603]
Cervellera C and Macciò D. 2017. An extreme learning machine approach to density estimation problems. IEEE Transactions on Cybernetics, 47(10):3254-3265[DOI:10.1109/TCYB.2017.2648261]
Chen Y J, Ma C Y, Sun L and Zhan T M. 2019. Edge-modified superpixel based spectral-patial kernel method for hyperspectral image classification. Acta Electronica Sinica, 47(1):73-81
陈允杰, 马辰阳, 孙乐, 詹天明. 2019.基于边缘修正的高光谱图像超像素空谱核分类方法.电子学报, 47(1):73-81[DOI:10.3969/j.issn.0372-2112.2019.01.010]
Fauvel M, Chanussot J and Benediktsson J A. 2012. A spatial-spectral kernel-based approach for the classification of remote-sensing images. Pattern Recognition, 45(1):381-392[DOI:10.1016/j.patcog.2011.03.035]
Gevaert C M, Tang J and García-Haro F J. 2014. Combining hyperspectral UAV and multispectral Formosat-2 imagery for precision agriculture applications//Proceedings of the 6th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing. Lausanne: IEEE: 1-4[ DOI: 10.1109/WHISPERS.2014.8077607 http://dx.doi.org/10.1109/WHISPERS.2014.8077607 ]
Gu Y, Xu Y and Guo B F. 2018. Hyperspectral image classification by combination of spatial-spectral features and ensemble extreme learning machines. Acta Geodaetica et Cartographica Sinica, 47(9):1238-1249
谷雨, 徐英, 郭宝峰. 2018.融合空谱特征和集成超限学习机的高光谱图像分类.测绘学报, 47(9):1238-1249[DOI:10.11947/j.AGCS.2018.20170476]
Huang G B, Zhou H M, Ding X J and Zhang R. 2012. Extreme learning machine for regression and multiclass classification. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 42(2):513-529[DOI:10.1109/TSMCB.2011.2168604]
Li J J, Du Q, Li W and Li Y S. 2015. Optimizing extreme learning machine for hyperspectral image classification. Journal of Applied Remote Sensing, 9(1):097296[DOI:10.1117/1.JRS.9.097296]
Li W, Wu G D, Zhang F and Du Q. 2017. Hyperspectral image classification using deep pixel-pair features. IEEE Transactions on Geoscience and Remote Sensing, 55(2):844-853[DOI:10.1109/TGRS.2016.2616355]
Lin Y P, Xie Z G, Xu K, Chen F and Liu L G. 2018. Fast cancer diagnosis based on extreme learning machine. Journal of University of Science and Technology of China, 48(2):154-160
林宇鹏, 谢智歌, 徐凯, 陈飞, 刘利刚. 2018.基于超限学习机的快速癌症检测方法.中国科学技术大学学报, 48(2):154-160[DOI:10.3969/j.issn.0253-2778.2018.02.010]
Liu Y X, Fang J J, Zhang X J and Sun J. 2015. Application of extreme learning machine in the nonlinear error compensation of magnetic compass. Chinese Journal of Scientific Instrument, 36(9):1921-1927
刘艳霞, 方建军, 张晓娟, 孙建. 2015.超限学习机在磁罗盘非线性误差补偿中的应用.仪器仪表学报, 36(9):1921-1927[DOI:10.3969/j.issn.0254-3087.2015.09.001]
Liu Y Y, Zhang J, Gao X J, Zhang X D and Gao J. 2017. 3D object recognition via convolutional-recursive neural network and kernel extreme learning machine. Pattern Recognition and Artificial Intelligence, 30(12):1091-1099
刘阳阳, 张骏, 高欣健, 张旭东, 高隽. 2017.基于卷积递归神经网络和核超限学习机的3D目标识别.模式识别与人工智能, 30(12):1091-1099[DOI:10.16451/j.cnki.issn1003-6059.201712004]
Lyu F and Han M. 2018. Hyperspectral remote sensing image classification based on deep extreme learning machine. Journal of Dalian University of Technology, 58(2):166-173
吕飞, 韩敏. 2018.基于深度超限学习机的高光谱遥感影像分类研究.大连理工大学学报, 58(2):166-173[DOI:10.7511/dllgxb201802009]
Mei Y and Lu C B. 2019. Adaptive weighted online extreme learning machine for imbalance data steam. Pattern Recognition and Artificial Intelligence, 32(2):144-150
梅颖, 卢诚波. 2019.面向不平衡数据流的自适应加权在线超限学习机算法.模式识别与人工智能, 32(2):144-150[DOI:10.16451/j.cnki.issn1003-6059.201902006]
Qiao T, Ren J C, Wang Z and Zabalza J. 2017. Effective denoising and classification of hyperspectral images using curvelet transform and singular spectrum analysis. IEEE Transactions on Geoscience and Remote Sensing, 55(1):119-133[DOI:10.1109/TGRS.2016.2598065]
Ran Q, Yu H Y, Gao L R, Li W and Zhang B. 2018. Superpixel and subspace projection-based support vector machines for hyperspectral image classification. Journal of Image and Graphics, 23(1):95-105
冉琼, 于浩洋, 高连如, 李伟, 张兵. 2018.结合超像元和子空间投影支持向量机的高光谱图像分类.中国图象图形学报, 23(1):95-105[DOI:10.11834/jig.170201]
Samat A, Du P J, Liu S C, Li J and Cheng L. 2014. E 2 LMs:ensemble extreme learning machines for hyperspectral image classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(4):1060-1069[DOI:10.1109/JSTARS.2014.2301775 ]
Schölkopf B and Smola A J. 2002. Learning with Kernels:Support Vector Machines, Regularization, Optimization, and Beyond. Cambridge:MIT Press
Tang Y D, Huang S C and Xue A J. 2017. Sparse representation based band selection for hyperspectral imagery target detection. Acta Electronica Sinica, 45(10):2368-2374
唐意东, 黄树彩, 薛爱军. 2017.面向目标检测基于稀疏表示的波段选择方法.电子学报, 45(10):2368-2374[DOI:10.3969/j.issn.0372-2112.2017.10.009]
Windrim L, Ramakrishnan R, Melkumyan A and Murphy R J. 2018. A physics-based deep learning approach to shadow invariant representations of hyperspectral images. IEEE Transactions on Image Processing, 27(2):665-677[DOI:10.1109/TIP.2017.2761542]
Yao K, Guo X D, Nan Y, Li K, Jiang S F and Sun T T. 2016. Research progress of hyperspectral remote sensing monitoring of vegetation biomass assessment. Science of Surveying and Mapping, 41(8):48-53
姚阔, 郭旭东, 南颖, 李坤, 江淑芳, 孙婷婷. 2016.植被生物量高光谱遥感监测研究进展.测绘科学, 41(8):48-53[DOI:10.16251/j.cnki.1009-2307.2016.08.010]
Zabalza J, Ren J C, Zheng J B, Zhao H M, Qing C M, Yang Z J, Du P J and Marshall S. 2016. Novel segmented stacked autoencoder for effective dimensionality reduction and feature extraction in hyperspectral imaging. Neurocomputing, 185:1-10[DOI:10.1016/j.neucom.2015.11.044]
Zhang H K, Li Y and Jiang Y N. 2018. Deep learning for hyperspectral imagery classification:the state of the art and prospects. Acta Automatica Sinica, 44(6):961-977
张号逵, 李映, 姜晔楠. 2018.深度学习在高光谱图像分类领域的研究现状与展望.自动化学报, 44(6):961-977[DOI:10.16383/j.aas.2018.c170190]
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.20187075]
Zhang X H, Yang G, Huang J H and Yang Y B. 2017. Hyperspectral images based on segmented column-and-line 2D-PCA. Computer Engineering, 43(9):256-262
张筱晗, 杨桄, 黄俊华, 杨永波. 2017.基于分段行列2D-PCA的高光谱图像数据降维方法.计算机工程, 43(9):256-262[DOI:10.3969/j.issn.1000-3428.2017.09.045]
Zhao W Z and Du S H. 2016. Spectral-spatial feature extraction for hyperspectral imageclassification:a dimension reduction and deep learning approach. IEEE Transactions on Geoscience and Remote Sensing, 54(8):4544-4554[DOI:10.1109/TGRS.2016.2543748]
Zhou Y C, Peng J T and Chen C L P. 2015. Extreme learning machine with composite kernels for hyperspectral image classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8(6):2351-2360[DOI:10.1109/JSTARS.2014.2359965]
相关作者
相关机构
京公网安备11010802024621
