Research on Improved Semantic Segmentation of Remote Sensing

doi:10.3778/j.issn.1002-8331.2011-0021

Abstract

Abstract: In recent years, with the continuous development of remote sensing technology, remote sensing images have a great application prospect in the fields of urban planning, agricultural planning and military training. In this paper, a semantic segmentation method of remote sensing images based on improved Deeplav3 is proposed. By improving the single up-sampling layer, multi-layer up-sampling is carried out by using the residuals obtained from the backbone network to ensure the semantic integrity of the image on resolution; at the same time, the dilated rate of 4-layer dilated convolution in ASPP layer is modified to make the network have a better effect on small object segmentation. The experimental results show that, the proposed algorithm achieves 94.92% and 98.01% in mIou and pixel accuracy, which is 3.77 percentage points and 2.40 percentage points higher than the original algorithm. The proposed algorithm not only has a higher accuracy, but also has a better robustness to various terrain segmentation. Meanwhile, the proposed algorithm is also superior to state-of-art algorithm such as U-net, SegNet, HR-Net and DANet.

Key words: semantic segmentation, remote sensing, deep learning, up-sampling, dilated rate

摘要： 近年来，随着遥感技术的不断发展，遥感图像在城市规划、农业规划及军事训练等领域有着极大的应用前景，对遥感图像的语义分割研究也变得尤为重要。提出一种基于改进Deeplabv3的遥感图像语义分割方法，通过改进单一的上采样层，利用主干网络中得到的残差进行多层上采样，保证图像在分辨率上的语义完整；同时，修改ASPP层中4层膨胀卷积的膨胀率，使得网络对小物体分割有更好的效果。实验结果表明：改进的Deeplabv3语义分割算法在自制的数据集上mIou和像素准确率达到了94.92%和98.01%，较原算法分别提高了3.77个百分点和2.40个百分点，不仅拥有更高的准确性，而且对各类地形的分割有更好的鲁棒性；同时，提出的分割方法也优于U-net、SegNet、HR-Net和DANet等当前主流分割方法。

关键词: 语义分割, 遥感, 深度学习, 上采样, 膨胀率

XIONG Fengguang, ZHANG Xin, HAN Xie, KUANG Liqun, LIU Huanle, JIA Jionghao. Research on Improved Semantic Segmentation of Remote Sensing[J]. Computer Engineering and Applications, 2022, 58(8): 185-190.

熊风光, 张鑫, 韩燮, 况立群, 刘欢乐, 贾炅昊. 改进的遥感图像语义分割研究[J]. 计算机工程与应用, 2022, 58(8): 185-190.

References

[1] ASRAR G R.Advances in quantitative earth remote sensing：past，prsent and future[J].Sensors，2019，19（24）：5399.
[2] 李志欣，裴玉东，郑璐.遥感处理技术应用浅述[J].现代信息科技，2020，4（11）：51-53.
LI Z X，PEI Y D，ZHENG L.Application of remote sensing processing technology[J].Modern Information Technology，2020，4（11）：51-53.
[3] LEI Z，ZHENG W，MIAO Y，et al.Level set for semantic segmentation with edge compensation[J].Journal of Physics：Conference Series，2020，1449（1）：012041.
[4] 张日升，原明亭，丁军航，等.基于图像阈值分割的浒苔图像提取[J].自动化技术与应用，2020，39（2）：83-86.
ZHANG R S，YUAN M T，DING J H，et al.Enteromorpha image extraction based on image threshold segmentation[J].Techniques of Automation and Applications，2020，39（2）：83-86.
[5] 桑高丽，郑增国，闫超.基于区域分割的表情鲁棒三维人脸识别方法[J].计算机应用研究，2020，37（3）：914-918.
SANG G L，ZHENG Z G，YAN C.Region-based expression invariant 3D face recognition[J].Application Research of Computers，2020，37（3）：914-918.
[6] CHEN L C，PAPANDREOU G，SCHROFF F，et al.Rethinking atrous convolution for semantic image segmentation[J].arXiv：1706.05587，2017.
[7] 胥智杰，杨小兵，何灵敏，等.多尺度遥感语义分割网络[J].计算机工程与应用，2020，56（21）：210-217.
XU Z J，YANG X B，HE L M，et al.Multiscale remote sensing semantic segmentation network[J].Computer Engineering and Applications，2020，56（21）：210-217.
[8] 熊伟，蔡咪，吕亚飞，等.基于神经网络的遥感图像海陆语义分割方法[J].计算机工程与应用，2020，56（15）：221-227.
XIONG W，CAI M，LYU Y F，et al.Sea-land semantic segmentation method of remote sensing image based on neural network[J].Computer Engineering and Applications，2020，56（15）：221-227.
[9] 祖宝开，夏克文，牛文佳，等.分块低秩图的遥感影像半监督分类应用[J].计算机科学与探索，2019，13（7）：1217-1226.
ZU B K，XIA K W，NIU W J，et al.Semi-supervised classification application of remote sensing image based on block low-rank images[J].Journal of Frontiers of Computer Science and Technology，2019，13（7）：1217-1226.
[10] 许玥，冯梦如，皮家甜，等.基于深度学习模型的遥感图像分割方法[J].计算机应用，2019，39（10）：2905-2914.
XU Y，FENG M R，PI J T，et al.Remote sensing image segmentation method based on deep learning model[J].Journal of Computer Applications，2019，39（10）：2905-2914.
[11] RONNEBERGER O，FISCHER P，BROX T.U-net：convolutional networks for biomedical image segmentation[C]//Proceedings of MICCAI，2015.
[12] 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 and Machine Intelligence，2018，40（4）：834-848.
[13] SUN K，XIAO B，LIU D，et al.Deep high-resolution representation learning for human pose estimation[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），2020.
[14] FU J，LIU J，TIAN H，et al.Dual attention network for scene segmentation[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），2020.
[15] CHEN J，HUANG H，PENG J，et al.Convolution neural network architecture learning for remote sensing scene classification[J].arXiv：2001.09614，2020.
[16] SHAO Z，ZHOU W，DENG X，et al.Multilabel remote sensing image retrieval based on fully convolutional network[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2020，13（1）：318-328.
[17] HAO H E，WANG S，WANG S，et al.A road extraction method for remote sensing image based on encoder-decoder network[J].Journal of Geodesy and Geoinformation Science，2020，3（2）：16-25.
[18] WANG Y，DONG M，SHEN J，et al.Dilated convolutions with lateral inhibitions for semantic image segmentation[J].arXiv：2006.03708，2020.