Semi-supervised hyperspectral image classification based on focal loss

doi:10.11772/j.issn.1001-9081.2019081390

Abstract

Abstract: Concerning the difficult acquisition of training data in HyperSpectral Image(HSI),a new semi-supervised classification framework for HSI was adopted,in which both limited labeled data and abundant unlabeled data were used to train deep neural networks. At the same time,the unbalanced distribution of hyperspectral samples leads to huge differences in the classification difficulty of different samples,and the original cross-entropy loss function is unable to describe this distribution feature,so the classification effect is not ideal. To address this problem,a multi-classification objective function based on focal loss was proposed in the semi-supervised classification framework. Finally,considering the influence of spatial information of HSI on classification,combined with Markov Random Field(MRF),the sample space features were used to further improve the classification effect. The proposed method was compared with various classical methods on two commonly used HSI datasets. Experimental results show that the proposed method can obtain classification results superior to other comparison methods.

Key words: HyperSpectral Image (HSI) classification, semi-supervised, focal loss, deep learning, Convolutional Neural Network (CNN)

摘要： 针对高光谱图像（HSI）训练数据获取困难的问题，采用了一种新的HSI半监督分类框架，该框架利用有限的标记数据和丰富的未标记数据来训练深度神经网络。同时，由于高光谱样本分布是不平衡的，导致不同样本分类难度存在巨大差异，采用原始交叉熵损失函数无法刻画这种分布特征，因而分类效果不理想。为了解决这个问题，在半监督分类框架中提出一种基于焦点损失的多分类目标函数。最后，考虑到HSI的空间信息对分类的影响，结合马尔可夫随机场（MRF），利用样本空间特征进一步改善分类效果。在两个常用的HSI数据集上，将所提方法与多种典型算法进行了实验对比分析，实验结果表明所提方法能够产生优于其他对比方法的分类效果。

关键词: 高光谱图像分类, 半监督, 焦点损失, 深度学习, 卷积神经网络

CLC Number:

TP751.1

ZHANG Kailin, YAN Qing, XIA Yi, ZHANG Jun, DING Yun. Semi-supervised hyperspectral image classification based on focal loss[J]. Journal of Computer Applications, 2020, 40(4): 1030-1037.

张凯琳, 阎庆, 夏懿, 章军, 丁云. 基于焦点损失的半监督高光谱图像分类[J]. 计算机应用, 2020, 40(4): 1030-1037.

References

[1] FENG F,LI W,DU Q,et al. Dimensionality reduction of hyperspectral image with graph-based discriminant analysis considering spectral similarity[J]. Remote Sensing,2017,9(4):No. 323.
[2] LEE H,KWON H. Going deeper with contextual CNN for hyperspectral image classification[J]. IEEE Transactions on Image Processing,2017,26(10):4843-4855.
[3] PETROPOULOS G P,KALAITZIDIS C,VADREVU K P. Support vector machines and object-based classification for obtaining land-use/cover cartography from Hyperion hyperspectral imagery[J]. Computers and Geosciences,2012,41:99-107.
[4] LIAO W,BELLENS R,PIZURICA A,et al. Classification of hyperspectral data over urban areas using directional morphological profiles and semi-supervised feature extraction[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2012,5(4):1177-1190.
[5] BIOUCAS-DIAS J M,PLAZA A,CAMPS-VALLS G,et al. Hyperspectral remote sensing data analysis and future challenges[J]. IEEE Geoscience and Remote Sensing Magazine,2013,1(2):6-36.
[6] YANG L,WANG M,YANG S,et al. Sparse spatio-spectral LapSVM with semisupervised kernel propagation for hyperspectral image classification[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2017,10(5):2046-2054.
[7] ZHENG Z,ZHENG L,YANG Y. Pedestrian alignment network for large-scale person re-identification[J]. IEEE Transactions on Circuits and Systems for Video Technology,2019,29(10):3037-3045.
[8] LONG J,SHELHAMER E,DARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2015:3431-3440.
[9] CHEN Y,LIN Z,ZHAO X,et al. Deep learning-based classification of hyperspectral data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2094-2107.
[10] CHEN Y,ZHAO X,JIA X. Spectral-spatial classification of hyperspectral data based on deep belief network[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2015,8(6):2381-2392.
[11] CHEN Y,JIANG H,LI C,et al. Deep feature extraction and classification of hyperspectral images based on convolutional neural networks[J]. IEEE Transactions on Geoscience and Remote Sensing,2016,54(10):6232-6251.
[12] LI Y,ZHANG H,SHEN Q. Spectral-spatial classification of hyperspectral imagery with 3D convolutional neural network[J]. Remote Sensing,2017,9(1):No. 67.
[13] ZHANG H,LI Y,ZHANG Y,et al. Spectral-spatial classification of hyperspectral imagery using a dual-channel convolutional neural network[J]. Remote Sensing Letters,2017,8(5):438-447.
[14] MOU L,GHAMISI P,ZHU X. Deep recurrent neural networks for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2017,55(7):3639-3655.
[15] LEE D H. Pseudo-label:the simple and efficient semi-supervised learning method for deep neural networks[EB/OL].[2019-04-21]. http://pdfs.semanticscholar.org/798d/9840d2439a0e5d47bcf5d164aa46d5e7dc26.pdf.
[16] WU H,PRASAD S. Semi-supervised deep learning using pseudo labels for hyperspectral image classification[J]. IEEE Transactions on Image Processing,2018,27(3):1259-1270.
[17] 李绣心, 凌志刚, 邹文. 基于卷积神经网络的半监督高光谱图像分类[J]. 电子测量与仪器学报,2018,32(10):95-102.((LI X X,LING Z G,ZOU W. Semi-supervised learning via convolutional neural network for hyperspectral image classification[J]. Journal of Electronic Measurement and Instrumentation,2018,32(10):95-102.))
[18] KIPF T N,WELLING M. Semi-supervised classification with graph convolutional networks[EB/OL].[2019-04-21]. https://arxiv.org/pdf/1609.02907.pdf.
[19] MA X,WANG H,WANG J. Semisupervised classification for hyperspectral image based on multi-decision labeling and deep feature learning[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2016,120:99-107.
[20] YAN Q,DING Y,XIA Y,et al. Class probability propagation of supervised information based on sparse subspace clustering for hyperspectral images[J]. Remote Sensing, 2017, 9(10):No. 1017.
[21] ZHAI H,ZHANG H,ZHANG L,et al. A new sparse subspace clustering algorithm for hyperspectral remote sensing imagery[J]. IEEE Geoscience and Remote Sensing Letters,2017,14(1):43-47.
[22] WEN Y, ZHANG K, LI Z, et al. A discriminative feature learning approach for deep face recognition[C]//Proceedings of the 2016 European Conference on Computer Vision,LNCS 9911. Cham:Springer,2016:499-515.
[23] LIU W,WEN Y,YU Z,et al. Large-margin softmax loss for convolutional neural networks[C]//Proceedings of the 33rd International Conference on Machine Learning. New York:JMLR. org, 2016:507-516.
[24] LIN T Y,GOYAL P,GIRSHICK R,et al. Focal loss for dense object detection[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway:IEEE,2017:2999-3007.
[25] LI J,BIOUCAS-DIAS J M,PLAZA A. Semisupervised hyperspectral image segmentation using multinomial logistic regression with active learning[J]. IEEE Transactions on Geoscience and Remote Sensing,2010,48(11):4085-4098.
[26] CAO X,XU L,MENG D,et al. Integration of 3-dimensional discrete wavelet transform and Markov random field for hyperspectral image classification[J]. Neurocomputing,2017,226:90-100.
[27] XU J,XU K,CHEN K,et al. Reweighted sparse subspace clustering[J]. Computer Vision and Image Understanding,2015, 138:25-37.
[28] PENG X,ZHANG L,YI Z. Scalable sparse subspace clustering[C]//Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2013:430-437.
[29] MA L,WANG C,XIAO B,et al. Sparse representation for face recognition based on discriminative low-rank dictionary learning[C]//Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2012:2586-2593.
[30] RAO S R,TRON R,VIDAL R,et al. Motion segmentation via robust subspace separation in the presence of outlying,incomplete,or corrupted trajectories[C]//Proceedings of the 2008 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2008:1-8.
[31] CHENG B,LIU G,WANG J,et al. Multi-task low-rank affinity pursuit for image segmentation[C]//Proceedings of the 2011 International Conference on Computer Vision. Piscataway:IEEE, 2011:2439-2446.
[32] 管皓, 薛向阳, 安志勇. 深度学习在视频目标跟踪中的应用进展与展望[J]. 自动化学报,2016,42(6):834-847.((GUAN H, XUE X Y,AN Z Y. Advances on application of deep learning for video object tracking[J]. Acta Automatica Sinica,2016,42(6):834-847.))
[33] CHAIB S,LIU H,GU Y,et al. Deep feature fusion for VHR remote sensing scene classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2017,55(8):4775-4784.
[34] CHÉRON G,LAPTEV I,SCHMID C. P-CNN:pose-based CNN features for action recognition[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Piscataway:IEEE, 2015:3218-3226.
[35] GIRSHICK R,DONAHUE J,DARRELL T,et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2014:580-587.
[36] REDMON J,FARHADI A. YOLO9000:better,faster,stronger[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2017:6517-6525.
[37] YU S,JIA S,XU C. Convolutional neural networks for hyperspectral image classification[J]. Neurocomputing,2017,219:88-98.