Please wait a minute...
 
Remote Sensing for Natural Resources    2023, Vol. 35 Issue (1) : 66-73     DOI: 10.6046/zrzyyg.2022028
|
Remote sensing image classification based on DenseNet feature hashing
LI Guoxiang1,2(), XIA Guo’en2,3(), BAI Liming3, MA Wenbin1,2
1. Department of Academic Affairs Guangxi University of Finance and Economics, Nanning 530003, China
2. Guangxi Engineering Research Center of Big Data Analysis of Finance and Taxation, Nanning 530003, China
3. School of Business Administration, Guangxi University of Finance and Economics, Nanning 530003, China
Download: PDF(4939 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks    
Abstract  

To achieve accurate remote sensing scene classification, this study proposed a classification algorithm based on DenseNet feature hashing. First, dimension reduction was conducted for high-level semantic features output by a DenseNet through a fully connected layer. Then, normalized feature vectors were generated as the input of the classification layer using an activation function, and an end-to-end classification network was formed. Using the trained network as a feature extractor, the features of the activation layer of test data were mapped into binary hash codes. Finally, the remote sensing scene classification was conducted using support vector machine. The new algorithm was validated on public data sets UC Merced, WHU, and NWPU-RESISC45, and its classification effect was compared with that of multiple algorithms at three levels, namely the conventional local feature descriptor, transfer learning, and depth feature coding. The experimental results are as follows. The new algorithm had significantly higher classification accuracy than conventional algorithms based on mid- and low-level semantic features. Compared with the algorithm based on transfer learning, the proposed algorithm has fine-scale DenseNet feature mapping and accumulates elements used to determine core categories of images and, thus, is more suitable for the feature distribution of remote sensing images. Compared with the depth feature coding algorithm, the new algorithm has a simple feature structure, high classification accuracy, and strong transferability and extensibility and, thus, can meet the classification requirements of different remote sensing scenarios.
Keywords transfer learning      feature coding      DenseNet      hash code     
ZTFLH:  TP393  
Issue Date: 20 March 2023
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Guoxiang LI
Guo’en XIA
Liming BAI
Wenbin MA
Cite this article:   
Guoxiang LI,Guo’en XIA,Liming BAI, et al. Remote sensing image classification based on DenseNet feature hashing[J]. Remote Sensing for Natural Resources, 2023, 35(1): 66-73.
URL:  
https://www.gtzyyg.com/EN/10.6046/zrzyyg.2022028     OR     https://www.gtzyyg.com/EN/Y2023/V35/I1/66
Fig.1  Visualization of different deep features
Fig.2  Illustration of dense feature hashing algorithm
Fig.3  Classification accuracy of different nodes number
方法 UCM WHU NWPU
BOW 63.80 63.96 40.11
BOW+SPM[18] 67.72 66.29 44.78
BOW+PCA 65.29 66.06 43.12
VLAD[19] 80.20 84.29 64.43
DRFV[20] 88.54 91.12 81.06
本文方法 98.21 99.20 96.49
Tab.1  Comparison of traditional feature descriptors(%)
预训练网络 UC Merced WHU NWPU
AlexNet 89.80 93.16 77.81
VGG19[21] 92.00 95.03 85.84
GoogleNet[22] 92.50 93.18 88.16
ResNet-50[2] 95.50 95.56 93.68
NetVlad[23] 91.15 95.98 82.62
DenseNet[3] 96.70 96.88 94.78
本文算法 98.21 99.02 96.49
Tab.2  Comparison of transfer learning algorithm (%)
深度特征 迁移学习 Bilinear BOVW VLAD FV 编码方法平均精度
AlexNet 89.80 80.40 89.50 90.30 92.50 88.18
VGG19 92.00 90.70 92.60 93.30 93.40 92.50
Google 92.50 90.30 84.00 89.50 87.70 87.88
Resnet50 95.50 93.30 90.00 89.90 90.40 90.90
DenseNet 96.70
DenseBlock1 54.10 78.86 84.57 87.81 76.34
DenseBlock2 71.62 85.24 87.14 92.38 84.10
DenseBlock3 90.76 91.90 92.76 95.05 92.62
DenseBlock4 95.24 93.90 92.29 93.14 93.64
本文算法 98.21
Tab.3  Feature coding of different CNN models(%)
Fig.4  Scatter diagram of various coding features
Fig.5  UCM confusion matrix
Fig.6  Probability analysis of misjudgment image
训练样本 测试样本 分类精度/% 时间/s
UCM UCM 98.21 87.27
NWPU 58.12 837.69
WHU 77.00 26.29
WHU UCM 71.30 55.56
NWPU 58.23 735.37
WHU 99.02 26.33
NWPU UCM 88.62 54.43
NWPU 96.49 715.13
WHU 91.86 25.03
Tab.4  Training and Testing in different data sets
[1] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]// Proceedings of the International Conference on Neural Information Processing Systems. Lake Tahoe: NIPS, 2012:1097-1105.
[2] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016:770-778.
[3] Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu:IEEE, 2017:4700-4708.
[4] Wang G, Fan B, Xiang S, et al. Aggregating rich hierarchical features for scene classification in remote sensing imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(9):4104-4115.
doi: 10.1109/JSTARS.4609443 url: https://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=4609443
[5] Razavian A S, Sullivan J, Carlsson S, et al. Visual instance retrieval with deep convolutional networks[J]. ITE Transactions on Media Technology and Applications, 2016, 4(3):251-258.
doi: 10.3169/mta.4.251 url: https://www.jstage.jst.go.jp/article/mta/4/3/4_251/_article
[6] Zheng L, Yang Y, Tian Q. SIFT meets CNN:A decade survey of instance retrieval[J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 2018, 40(5):1224-1244.
doi: 10.1109/TPAMI.2017.2709749 url: https://ieeexplore.ieee.org/document/7935507/
[7] Hu F, Xia G S, Hu J, et al. Transferring deep convolutional neural networks for the scene classification of high-resolution remote sensing imagery[J]. Remote Sensing, 2015, 7(11):14680-14707.
doi: 10.3390/rs71114680 url: http://www.mdpi.com/2072-4292/7/11/14680
[8] Cheng G, Li Z, Yao X, et al. Remote sensing image scene classification using bag of convolutional features[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(10):1735-1739.
doi: 10.1109/LGRS.2017.2731997 url: http://ieeexplore.ieee.org/document/8008758/
[9] 王鑫, 李可, 宁晨, 等. 基于深度卷积神经网络和多核学习的遥感图像分类方法[J]. 电子与信息学报, 2019, 41(5):1098-1105.
[9] Wang X, Li K, Ning C, et al. Remote sensing image classification method based on deep convolution neural network and multi-kernel learning[J]. Journal of Electronics & Information Technology, 2019, 41(5):1098-1105.
[10] He N, Fang L, Li S, et al. Remote sensing scene classification using multilayer stacked covariance pooling[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(12):6899-6910.
doi: 10.1109/TGRS.2018.2845668 url: https://ieeexplore.ieee.org/document/8408558/
[11] Cheng G, Yang C, Yao X, et al. When deep learning meets metric learning:Remote sensing image scene classification via learning discriminative CNNs[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(5):2811-2821.
doi: 10.1109/TGRS.2017.2783902 url: http://ieeexplore.ieee.org/document/8252784/
[12] 刘异, 庄姊琪, 闫利, 等. 联合Fisher核编码和卷积神经网络的影像场景分类[J]. 遥感信息, 2018, 33(4):8-15.
[12] Liu Y, Zhuang Z Q, Yan L, et al. Combined Fisher kernel coding framework with convolutional neural network for remote sensing scene classification[J]. Remote Sensing Information, 2018, 33(4):8-15.
[13] 余东行, 张保明, 赵传, 等. 联合卷积神经网络与集成学习的遥感影像场景分类[J]. 遥感学报, 2020, 24(6):717-727.
[13] Yu D H, Zhang B M, Zhao C, et al. Scene classification of remote sensing image using ensemble convolutional neural network[J]. Journal of Remote Sensing, 2020, 24(6):717-727.
[14] 李亚飞, 董红斌. 基于卷积神经网络的遥感图像分类研究[J]. 智能系统学报, 2018, 13(4):550-556.
[14] Li Y F, Dong H B. Classification of remote-sensing image based on convolutional neural network[J]. CAAI Transactions on Intelligent Systems, 2018, 13(4):550-556.
[15] 张哲益, 曹卫华, 朱蕊, 等. 基于脉冲卷积神经网络稀疏表征的高分辨率遥感图像场景分类方法[J]. 控制与决策, 2022, 37(9):2305-2313.
[15] Zhang Z Y, Cao W H, Zhu R, et al. Sparse representation with spike convolutional neural networks for scene classification of remote sensing images of high resolution[J]. Control and Decision, 2022, 37(9):2305-2313.
[16] Selvaraju R R, Cogswell M, Das A, et al. Grad-cam:Visual explanations from deep networks via gradient-based localization[C]// Proceedings of the IEEE International Conference on Computer Vision. Venice:IEEE, 2017:618-626.
[17] Lin K, Yang H F, Hsiao J H, et al. Deep learning of binary hash codes for fast image retrieval[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. Boston:IEEE, 2015:27-35.
[18] Lazebnik S, Schmid C, Ponce J. Beyond bags of features:Spatial pyramid matching for recognizing natural scene categories[C]// 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06). New York: IEEE, 2006:2169-2178.
[19] Arandjelovic R, Zisserman A. All about VLAD[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Portland:IEEE, 2013:1578-1585.
[20] 李国祥, 马文斌, 王继军. 稠密特征编码的遥感场景分类算法[J]. 小型微型计算机系统, 2021, 42(4):766-772.
[20] Li G X, Ma W B, Wang J J. Remote sensing image classification based on dense feature coding[J]. Journal of Chinese Computer Systems, 2021, 42(4):766-772.
[21] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J/OL]. arxiv, 2014[2022-04-24]. https://arxiv.org/pdf/1409.1556.pdf.
url: https://arxiv.org/pdf/1409.1556.pdf
[22] Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston:IEEE, 2015:1-9.
[23] Arandjelovic R, Gronat P, Torii A, et al. NetVLAD:CNN architecture for weakly supervised place recognition[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016:5297-5307.
[24] Lin T, Roychowdhury A, Maji S. Bilinear convolutional neural networks for fine-grained visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(6):1309-1322.
doi: 10.1109/TPAMI.2017.2723400 url: https://ieeexplore.ieee.org/document/7968351/
[25] Sánchez J, Perronnin F, Mensink T, et al. Image classification with the fisher vector:Theory and practice[J]. International Journal of Computer Vision, 2013, 105(3):222-245.
doi: 10.1007/s11263-013-0636-x url: http://link.springer.com/10.1007/s11263-013-0636-x
[1] LIU Guangjin, WANG Guanghui, BI Weihua, LIU Huijie, YANG Huachao. Cloud detection algorithm of remote sensing image based on DenseNet and attention mechanism[J]. Remote Sensing for Natural Resources, 2022, 34(2): 88-96.
[2] CAI Zhiling, WENG Qian, YE Shaozhen, JIAN Cairen. Remote sensing image scene classification based on Inception-V3[J]. Remote Sensing for Land & Resources, 2020, 32(3): 80-89.
[3] Zhiwei LIN, Weihao TU, Jiahang HUANG, Qilu DING, Zhengwen ZHOU, Jinfu LIU. Tree species recognition of UAV aerial images based on FC-DenseNet[J]. Remote Sensing for Land & Resources, 2019, 31(3): 225-233.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
京ICP备05055290号-2
Copyright © 2017 Remote Sensing for Natural Resources
Support by Beijing Magtech