Classification of high-resolution remote sensing images based on enhanced DeepLab algorithm and adaptive loss function

Zeyu XU; Zhanfeng SHEN; Yang LI; Lifang ZHAO; Yingming KE; Lingling LI; Qi WEN

doi:10.11834/jrs.20209200

Remote Sensing Intelligent Interpretation | Views : 0 下载量: 737 CSCD: 4 更多指标

PDF
Export
Share
Collection
Album

Classification of high-resolution remote sensing images based on enhanced DeepLab algorithm and adaptive loss function
Vol. 26, Issue 2, Pages: 406-415(2022)
Published： 07 February 2022 ，
DOI： 10.11834/jrs.20209200

扫描看全文

许泽宇，沈占锋，李杨，赵丽芳，柯映明，李苓苓，温奇.2022.增强型DeepLab算法和自适应损失函数的高分辨率遥感影像分类.遥感学报，26（2）： 406-415

Xu Z Y，Shen Z F，Li Y，Zhao L F，Ke Y M，Li L L and Wen Q. 2022. Classification of high-resolution remote sensing images based on enhanced DeepLab algorithm and adaptive loss function. National Remote Sensing Bulletin， 26（2）：406-415
许泽宇，沈占锋，李杨，赵丽芳，柯映明，李苓苓，温奇.2022.增强型DeepLab算法和自适应损失函数的高分辨率遥感影像分类.遥感学报，26（2）： 406-415 DOI： 10.11834/jrs.20209200.

Xu Z Y，Shen Z F，Li Y，Zhao L F，Ke Y M，Li L L and Wen Q. 2022. Classification of high-resolution remote sensing images based on enhanced DeepLab algorithm and adaptive loss function. National Remote Sensing Bulletin， 26（2）：406-415 DOI： 10.11834/jrs.20209200.

摘要

高分辨率遥感影像地物复杂，分类难度大，而深度学习方法可以提取地物更多更深层次的特征信息，适用于高分辨率遥感影像的地物分类。本文研究对高分辨率影像中不透水地面、建筑、低矮植被、树、车辆等地物的高精度分类。结合遥感多地物分类的特点，以DeepLab v3+网络模型为基础，提出E-DeepLab网络模型。主要改进为：（1）改进编码器和解码器的结合方式，使用简洁有效的加成连接方式。（2）缩小单次上采样倍数，增加上采样层，提高编码器与解码器连接的紧密性。（3）使用改进的自适应权重损失函数，自动调节地物损失权重。同时根据数据特点，提出结合DSM、NDVI数据等多通道训练方式。使用两个地区数据进行实验，结果表明，两地区精度均明显优于原始DeepLab v3+模型和其他相关模型，Potsdam地区总体提取精度达到93.2%，建筑物提取精度达到97.8%，Vaihingen地区总体提取精度达到90.7%，建筑物提取精度达到96.3%。目视对比分类图和标准标记图，两者具有高度的一致性。本文所提出的E-DeepLab网络在高分辨率遥感影像地物高精度提取和分类中有较好的应用价值。

Abstract

High-resolution remote sensing images are complex and difficult to classify. Deep learning methods can extract more and deeper information of the features

which is suitable for the classification of high-resolution remote sensing images. This paper studies the high-precision classification of impervious ground

buildings

low vegetation

trees

vehicles

and other features in high-resolution images. An E-DeepLab network model is proposed by combining the characteristics of remote sensing multiground feature classification based on DeepLab v3+ network model. The main improvements are as follows: (1) Improving the combination of encoder and decoder modules and using a simple and effective addition connection method. (2) Reducing the single upsampling multiple

increasing the upsampling layer

and improving the tightness of the connection between the encoder and the decoder. (3) Using the improved adaptive weight loss function to automatically adjust the weight of losses. In accordance with the characteristics of the data

a multichannel training method combining digital surface model and normalized difference vegetation index data is proposed. Using the data from the two regions to conduct experiments

the overall extraction accuracy in Potsdam region reaches 93.2%

the extraction accuracy in buildings reaches 97.8%

the overall extraction accuracy in Vaihingen region reaches 90.7%

and the accuracy of building extraction reaches 96.3%. The accuracy of the two regions is significantly better than the original DeepLab v3+ model and other related models. The visual effect of the classification results is extremely close to the standard map by comparing the classification map and the standard marker map. Results show that the E-DeepLab network has good application value in the feature extraction and classification of high-precision remote sensing images.

关键词

遥感高分辨率影像深度学习E-DeepLab自适应权重损失函数

Keywords

remote sensinghigh-resolution imagesdeep learningE-DeepLabadaptive weight loss function

references

Audebert N, Le Saux B and Lefèvre S. 2018. Beyond RGB: very high resolution urban remote sensing with multimodal deep networks. ISPRS Journal of Photogrammetry and Remote Sensing, 140: 20-32 [DOI: 10.1016/j.isprsjprs.2017.11.011http://dx.doi.org/10.1016/j.isprsjprs.2017.11.011]

Badrinarayanan V, Kendall A and Cipolla R. 2017. SegNet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12): 2481-2495 [DOI: 10.1109/TPAMI.2016.2644615http://dx.doi.org/10.1109/TPAMI.2016.2644615]

Chen K Q, Fu K, Yan M L, Gao X, Sun X and Wei X. 2018a. Semantic segmentation of aerial images with shuffling convolutional neural networks. IEEE Geoscience and Remote Sensing Letters, 15(2): 173-177 [DOI: 10.1109/LGRS.2017.2778181http://dx.doi.org/10.1109/LGRS.2017.2778181]

Chen L C, Papandreou G, Kokkinos I, Murphy K and Yuille A L. 2017a. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4): 834-848 [DOI: 10.1109/TPAMI.2017.2699184http://dx.doi.org/10.1109/TPAMI.2017.2699184]

Chen L C, Papandreou G, Schroff F and Adam H. 2017b. Rethinking atrous convolution for semantic image segmentation. arXiv: 1706.05587

Chen L C, Zhu Y K, Papandreou G, Schroff F and Adam H. 2018b. Encoder-decoder with atrous separable convolution for semantic image segmentation//15th European Conference on Computer Vision. Munich, Germany: Springer: 833-851 [DOI: 10.1007/978-3-030-01234-2_49http://dx.doi.org/10.1007/978-3-030-01234-2_49]

Chollet F. 2017. Xception: deep learning with depthwise separable convolutions//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI, USA: IEEE: 1800-1807 [DOI: 10.1109/CVPR.2017.195http://dx.doi.org/10.1109/CVPR.2017.195]

Gong J Y and Ji S P. 2017. From Photogrammetry to computer vision. Geomatics and Information Science of Wuhan University, 42(11): 1518-1522, 1615

龚健雅, 季顺平. 2017. 从摄影测量到计算机视觉. 武汉大学学报(信息科学版), 42(11): 1518-1522, 1615 [DOI: 10.13203/j.whugis20170283http://dx.doi.org/10.13203/j.whugis20170283]

Gong J Y and Ji S P. 2018. Photogrammetry and deep learning. Acta Geodaetica et Cartographica Sinica, 47(6): 693-704

龚健雅, 季顺平. 2018. 摄影测量与深度学习. 测绘学报, 47(6): 693-704 [DOI: 10.11947/j.AGCS.2018.20170640http://dx.doi.org/10.11947/j.AGCS.2018.20170640]

Guo H D, Chen R S, Xu Z W, Sun J J, Bi J, Wang L Z, Luo J J, Shen H W, Gu D X, Liang D, Shen W Q, Zhang X, Spiess H W and Lengauer T. 2016. Big data in natural sciences, humanities and social sciences-review of the 6th exploratory round table conference. Bulletin of Chinese Academy of Sciences, 31(6): 707-716

郭华东, 陈润生, 徐志伟, 孙建军, 毕军, 王力哲, 骆健俊, 沈华伟, 顾东晓, 梁栋, 沈文庆, 张旭, Spiess H W, Lengauer T. 2016. 自然科学与人文科学大数据——第六届中德前沿探索圆桌会议综述. 中国科学院院刊, 31(6): 707-716 [DOI: 10.16418/j.issn.1000-3045.2016.06.014http://dx.doi.org/10.16418/j.issn.1000-3045.2016.06.014]

He K M, Zhang X Y, Ren S Q and Sun J. 2016. Deep residual learning for image recognition//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE: 770-778 [DOI: 10.1109/CVPR.2016.90http://dx.doi.org/10.1109/CVPR.2016.90]

Kussul N, Lavreniuk M, Skakun S and Shelestov A. 2017. Deep learning classiﬁcation of land cover and crop types using remote sensing data. IEEE Geoscience and Remote Sensing Letters, 14(5): 778-782 [DOI: 10.1109/LGRS.2017.2681128http://dx.doi.org/10.1109/LGRS.2017.2681128]

Li D R, Tong Q X, Li R X, Gong J Y and Zhang L P. 2012. Current issues in high-resolution Earth observation technology. Science China Earth Sciences, 55(7): 1043-1051

李德仁, 童庆禧, 李荣兴, 龚健雅, 张良培. 2012. 高分辨率对地观测的若干前沿科学问题. 中国科学: 地球科学, 42(6): 805-813 [DOI: 10.1360/zd-2012-42-6-805http://dx.doi.org/10.1360/zd-2012-42-6-805]

Li G, Yun I, Kim J and Kin J. 2019. DABNet: depth-wise asymmetric bottleneck for real-time semantic segmentation. arXiv: 1907.11357v2

Li Z Q. 2019. Research on Urban Building Extraction from Deep Learning Approach. Beijing: Beijing University of Civil Engineering and Architecture

李志强. 2019. 基于深度学习的城市建筑物提取方法研究. 北京: 北京建筑大学

Liu X. 2018. Road Extraction from Remote Sensing Image based on Deep Learning. Xuzhou: China University of Mining and Technology

刘笑. 2018. 基于深度学习算法模型的遥感影像道路提取技术. 徐州: 中国矿业大学

Mnih V. 2013. Machine Learning for Aerial Image Labeling. Canada: University of Toronto

Paszke A, Chaurasia A, Kim S and Culurciello E. 2016. ENet: a deep neural network architecture for real-time semantic segmentation. arXiv: 1606.02147v1

Peng B. 2018. Research on Road Information Extraction of Remote Sensing Image based on Deep Learning. Chengdu: University of Electronic Science and Technology of China

彭博. 2018. 基于深度学习的遥感图像道路信息提取算法研究. 成都: 电子科技大学

Ronneberger O, Fischer P and Brox T. 2015. U-Net: convolutional networks for biomedical image segmentation//18th International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich, Germany: Springer: 234-241 [DOI: 10.1007/978-3-319-24574-4_28http://dx.doi.org/10.1007/978-3-319-24574-4_28]

Shelhamer E, Long J and Darrell T. 2017. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(4): 640-651 [DOI: 10.1109/TPAMI.2016.2572683http://dx.doi.org/10.1109/TPAMI.2016.2572683]

Simonyan K and Zisserman A. 2015. Very deep convolutional networks for large-scale image recognition. arXiv: 1409.1556

Szegedy C, Liu W, Jia Y Q, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V and Rabinovich A. 2015. Going deeper with convolutions//2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA: IEEE: 1-9 [DOI: 10.1109/CVPR.2015.7298594http://dx.doi.org/10.1109/CVPR.2015.7298594]

Xia M, Cao G, Wang G Y and Shang Y F. 2017. Remote sensing image classification based on deep learning and conditional random fields. Journal of Image and Graphics, 22(9): 1289-1301

夏梦, 曹国, 汪光亚, 尚岩峰. 2017. 结合深度学习与条件随机场的遥感图像分类. 中国图象图形学报, 22(9): 1289-1301 [DOI: 10.11834/jig.170122http://dx.doi.org/10.11834/jig.170122]

Alert me when the article has been cited

提交

Optical-signal token guided change detection network for heterogeneous remote sensing image

Research progress on hyperspectral anomaly detection

Hyperspectral remote sensing image classification based on multidirectional adaptive aware network

Weakly supervised scale adaptation data augmentation for scene classification of high-resolution remote sensing images

Urban cinnamomum camphora crown detection research using RGB-DSM images and deep learning