High-resolution remote sensing image classification based on improved DeepLabV3 network

doi:10.13474/j.cnki.11-2246.2021.0108

Abstract

Abstract: Since high-resolution images usually have features such as rich details and large category differences, current methods for remote sensing classification of convolutional neural network generally perform poorly in classification accuracy and object boundary detection. In this paper, we propose an image classification model based on enhanced DeepLabV3 network. Firstly, R-MCN is built by combing residual networks and convolution kernels of different sizes, which is used to extract multi-scale, multi-level feature information of shallow networks. Then a learnable upsampling method is used to restore image size, and fuse the features extracted by R-MCN with high-level semantic information. Finally, a loss function named Mloss is built to achieve classification results of remote sensing images. The experimental results demonstrate that the proposed method can refine object boundaries,improve classification performance,and obtain higher accuracy of image classification compared with traditional convolutional neural network.

Key words: image classification, fusion, upsampling, loss function, DeepLabV3

CLC Number:

P237

YE Yuanxin, TAN Xin, SUN Miaomiao, WANG Mengmeng. High-resolution remote sensing image classification based on improved DeepLabV3 network[J]. Bulletin of Surveying and Mapping, 2021, 0(4): 40-44.

References

[1] 徐慧敏,齐华,南柯,等.结合nDSM的高分辨率遥感影像深度学习分类方法[J].测绘通报,2019(8):63-67.
[2] 苏健民,杨岚心,景维鹏.基于U-Net的高分辨率遥感图像语义分割方法[J].计算机工程与应用,2019,55(7):212-218.
[3] LONG J, SHELHAMER E, DARRELL T. Fully convolution-al networks for semantic segmentation[C]//Proceedings of 2015 rence on Computer Vision and Pattern Recognition. New York:IEEE Computer Society,2015:3431-3440.
[4] 李欣,唐文莉,杨博.利用深度残差网络的高分遥感影像语义分割[J].应用科学学报, 2019, 37(2):136-144.
[5] RONNEBERGER O, FISCHER P, BROX T. U-net:Convolu-tional networks for biomedical imagesegmentation[M].[S.l.]:Springer International Publishing,2015:234-241.
[6] BADRINARAYANAN V, KENDALL A, CIPOLLA R.SegNet:a deep convolutional encoder-decoder architecture for image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2481-2495.
[7] CHEN L C, PAPANDREOU G, KOKKINOS I, et al.Semantic image segmentation with deep convolutional nets and fully connected CRFs[J].Computer Science,2014(4):357-361.
[8] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab:semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2016,40(4):834-848.
[9] CHEN L C, ZHU Y, PAPANDREOU G, et al.Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of 2018 European Conference on Computer Vision.[S.l.]:Springer International Publishing,2018:801-818.
[10] HE K, ZHANG X, REN S, et al.Identity mappings in deep residual networks[C]//European conference on computer vision.Switzerland:Springer,2016:630-645.
[11] 陈天华,郑司群,于峻川.采用改进DeepLab网络的遥感图像分割[J].测控技术,2018,37(11):40-45.
[12] WANG P, CHEN P, YUAN Y, et al. Understanding convolution for semantic segmentation[C]//Proceedings of 2018 IEEE Winter Conference on Applications of Computer Vision (WACV).New York:IEEE,2018:1451-1460.
[13] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017(99):2999-3007.
[14] TONG W, PI-LIAN H. The classification algorithm based on the cross entropy rule and new activetion function in fuzzy neural network[C]//Proceedings of 2005 International Conference on Machine Learning and Cybernetics.New York:IEEE, 2005,8:4631-4635.

[1]	YUE Shengjie, WANG Hongqi, LIU Qunpo, ZHAO Rongliang. Adaptive pose estimation for robot based on extended Kalman filter and point-line iterative closest point [J]. Bulletin of Surveying and Mapping, 2022, 0(7): 49-53.
[2]	YANG Jun, LIU Linghui. Feature extraction of spatial and temporal change of land in functional areas based on weighted full-polarization SAR image classification [J]. Bulletin of Surveying and Mapping, 2022, 0(7): 138-142,153.
[3]	REN Yuejuan, GE Xiaosan. An road synthesis extraction method of remote sensing image based on improved DeepLabV3+ network [J]. Bulletin of Surveying and Mapping, 2022, 0(6): 55-61.
[4]	XIE Hong, CHEN Libo, NIE Qian, WU Wei, ZHANG Pei. Air-ground image fusion technology with point cloud registration [J]. Bulletin of Surveying and Mapping, 2022, 0(6): 82-87.
[5]	HAN Shijing, MIAO Shufeng, HAO Xiangyang, CHEN Runze. Review on the development of surveillance video and geographic information fusion technology [J]. Bulletin of Surveying and Mapping, 2022, 0(5): 1-6.
[6]	CHEN Lanlan, XIA Yiqiang, XIAO Haiping, LIU Xiaosheng. Research status and progress of slope stability monitoring methods in open-pit mines [J]. Bulletin of Surveying and Mapping, 2022, 0(5): 7-13.
[7]	WEN Nu, GUO Renzhong, HE Biao, WAN Yuan. Multi-Patch multi-frame incremental traffic video object detection method based on YOLO v4 [J]. Bulletin of Surveying and Mapping, 2022, 0(5): 38-44.
[8]	RUAN Minghao, LU Yonghua, LIU Yuxian, TANG Shengjun, WANG Weixi, ZHANG Yuqi, LI You. Registration and seamless fusion of point cloud from LiDAR and oblique photography [J]. Bulletin of Surveying and Mapping, 2022, 0(4): 6-10.
[9]	LIU Xiaoyu, LIU Yang, DU Mingyi, ZHANG Min, JIA Jingjue, YANG Heng. Research on construction and demolition waste stacking point identification based on DeeplabV3+ [J]. Bulletin of Surveying and Mapping, 2022, 0(4): 16-19,43.
[10]	DOU Shiqing, CHEN Zhiyu, XU Yong, ZHENG Hegang, MIAO Linlin, SONG Yingying. Hyperspectral image classification based on multi-feature fusion and dimensionality reduction algorithms [J]. Bulletin of Surveying and Mapping, 2022, 0(4): 32-36,50.
[11]	GAO Sha, YUAN Xiping, GAN Shu, YANG Minglong, YUAN Xinyue, LUO Weidong. Fusion of UAV image and LiDAR point cloud to study the detection technology of mountain surface cover landscape characteristics [J]. Bulletin of Surveying and Mapping, 2022, 0(1): 110-115.
[12]	GAO Yu, HU Zhaoling, FAN Ru. Effect of high-resolution image fusion algorithm on the classification precision of land utilization in coastal wetland [J]. Bulletin of Surveying and Mapping, 2022, 0(1): 116-120.
[13]	SUN Baoyan, ZHOU Xin, QIN Yucheng, WEI Longhua, GE Guanghao. Application of air ground multi-data mutual auxiliary fusion technology in reconstruction of large-scale ancient buildings [J]. Bulletin of Surveying and Mapping, 2022, 0(1): 121-127,132.
[14]	HAN Zhisheng, SUN Pichuan, TANG Chao. Generative adversarial network super-resolution reconstruction for single remote sensing image [J]. Bulletin of Surveying and Mapping, 2021, 0(8): 106-110.
[15]	GAO Sha, YUAN Xiping, GAN Shu, YANG Yafu, LI Xia. Application of airborne LiDAR and UAV image fusion for complicated terrain [J]. Bulletin of Surveying and Mapping, 2021, 0(7): 65-69.