面向高分遥感影像道路提取的轻量级双注意力和特征补偿残差网络模型

doi:10.12082/dqxxkx.2022.210597

地球信息科学学报 ›› 2022, Vol. 24 ›› Issue (5): 949-961.doi: 10.12082/dqxxkx.2022.210597

面向高分遥感影像道路提取的轻量级双注意力和特征补偿残差网络模型

陈振¹^,²^,³(), 陈芸芝¹^,²^,³^,^*(), 吴婷¹^,²^,³, 李佳优¹^,²^,³

1.福州大学数字中国研究院（福建）,福州 350108
2.卫星空间信息技术综合应用国家地方联合工程研究中心,福州 350108
3.福州大学空间数据挖掘与信息共享教育部重点实验室,福州 350108

收稿日期:2021-09-30 修回日期:2021-12-03 出版日期:2022-05-25 发布日期:2022-07-25
通讯作者: * 陈芸芝（1982— ）,女,福建福州人,博士,副研究员,主要研究方向为环境遥感和林业遥感等。 E-mail: chenyunzhi@fzu.edu.cn
作者简介:陈振（1996— ）,男,福建南平人,硕士生,主要研究方向为遥感影像信息提取。E-mail: n195527003@fzu.edu.cn
基金资助:
国家自然科学基金项目(42071446);联通(福建)产业互联网有限公司委托项目(JC21-3502-2020-000559)

A Lightweight Dual Attention and Feature Compensated Residual Network Model for Road Extraction from High-Resolution Remote Sensing Images

CHEN Zhen¹^,²^,³(), CHEN Yunzhi¹^,²^,³^,^*(), WU Ting¹^,²^,³, LI Jiayou¹^,²^,³

1. The Academy of Digital China(Fujian), Fuzhou University, Fuzhou 350108, China
2. National and Local Joint Engineering Research Center for the Comprehensive Application of Satellite Space Information Technology, Fuzhou 350108, China
3. Key Laboratory of Spatial Data Mining and Information Sharing, Ministry of Education, Fuzhou University, Fuzhou 350108, China.

Received:2021-09-30 Revised:2021-12-03 Online:2022-05-25 Published:2022-07-25
Supported by:
National Natural Science Foundation of China(42071446);Commissioning Project of Unicom (Fujian) Industrial Internet Co., Ltd.,(JC21-3502-2020-000559)

摘要/Abstract

摘要：

针对高分辨率遥感影像背景复杂,道路提取容易受阴影、建筑物和铁路等背景信息干扰的问题,提出一种带有轻量级双注意力和特征补偿机制的DAFCResUnet模型。该模型在ResUnet的基础上,通过增加轻量级的双注意力和特征补偿模块实现模型在性能和时空复杂度上的平衡。其中,双注意力模块可以增强模型的特征提取能力,特征补偿模块可以融合网络中来自深浅层的道路特征。在DeepGlobe和GF-2道路数据集上的实验结果表明,DAFCResUnet模型的IoU和F1-score可以达到0.6713、0.8033和0.7402、0.8507,模型的整体精度优于U-Net、ResUnet和VNet模型。与U-Net和ResUnet模型相比,DAFCResUnet模型仅增加了少量的计算量和参数量,但IoU和F1-score均有较大幅度的提高;与VNet模型相比,DAFCResUnet模型在计算量和参数量远低于VNet的情况下取得了更高的精度,模型在精度和时空复杂度两方面均有优势。相比其他对比模型,DAFCResUnet模型具有更强的特征提取和抗干扰能力,能更好解决道路上的干扰物、与道路特征相似地物、树荫或阴影遮挡等造成的道路空洞、误提和漏提现象。

关键词: 深度学习, 道路提取, 高分辨率遥感影像, 残差网络, U-Net模型, 双注意力模块, 编解码器, 特征补偿

Abstract:

Aiming at the problem that the background of high-resolution remote sensing images is complex and road extraction is easily disturbed by background information such as shadows, buildings, and railroads, the DAFCResUnet model with lightweight dual attention and feature compensation mechanism is proposed in this study. The model is based on ResUnet and achieves a balance between model performance and spatiotemporal complexity by adding lightweight dual attention and feature compensation modules. The dual attention module enhances the feature extraction capability of the model, and the feature compensation module can fuse the road features from deep and shallow layers in the network. The experimental results using DeepGlobe and GF-2 road datasets show that the IoU of the DAFCResUnet model can reach 0.6713, 0.8033, respectively, and the F1-score is 0.7402, 0.8507, respectively. The overall accuracy of the model is higher than that of U-Net, ResUnet, and VNet models. Compared with the U-Net and ResUnet models, the DAFCResUnet model only increases a small amount of computation and number of parameters, but the IoU and F1-score are improved substantially. Compared with the VNet model, the DAFCResUnet model achieves a higher accuracy with much lower computation and smaller number of parameters, and the model has advantages in both accuracy and spatiotemporal complexity. Compared with the other models, the DAFCResUnet model has stronger feature extraction and anti-interference ability, which can better solve the commission and omission caused by interfering objects on the road, ground features similar to roads, tree shade or shadow shading, etc.

Key words: deep learning, road extraction, high-resolution remote sensing images, residual network, U-Net model, dual attention module, codecs, feature compensation

陈振, 陈芸芝, 吴婷, 李佳优. 面向高分遥感影像道路提取的轻量级双注意力和特征补偿残差网络模型[J]. 地球信息科学学报, 2022, 24(5): 949-961.DOI:10.12082/dqxxkx.2022.210597

CHEN Zhen, CHEN Yunzhi, WU Ting, LI Jiayou. A Lightweight Dual Attention and Feature Compensated Residual Network Model for Road Extraction from High-Resolution Remote Sensing Images[J]. Journal of Geo-information Science, 2022, 24(5): 949-961.DOI:10.12082/dqxxkx.2022.210597

图/表 12

图1

表1

DAFCResUnet网络各层的参数及输出特征图尺寸

编码器				解码器
层号	网络层	步长	输出特征图尺寸	层号	网络层	步长	输出特征图尺寸
1	Conv 3 $×$ 3	1	256 $×$ 256 $×$ 32	1	T-Conv 2 $×$ 2	2	32 $×$ 32 $×$ 256
2	Conv 3 $×$ 3	1	256 $×$ 256 $×$ 32	2	Conv 3 $×$ 3	1	32 $×$ 32 $×$ 256
3	Conv 3 $×$ 3	2	128 $×$ 128 $×$ 64	3	Conv 3 $×$ 3	1	32 $×$ 32 $×$ 256
4	Conv 3 $×$ 3	1	128 $×$ 128 $×$ 64	4	T-Conv 2 $×$ 2	2	64 $×$ 64 $×$ 128
5	Conv 3 $×$ 3	2	64 $×$ 64 $×$ 128	5	Conv 3 $×$ 3	1	64 $×$ 64 $×$ 128
6	Conv 3 $×$ 3	1	64 $×$ 64 $×$ 128	6	Conv 3 $×$ 3	1	64 $×$ 64 $×$ 128
7	Conv 3 $×$ 3	2	32 $×$ 32 $×$ 256	7	T-Conv 2 $×$ 2	2	128 $×$ 128 $×$ 64
8	Conv 3 $×$ 3	1	32 $×$ 32 $×$ 256	8	Conv 3 $×$ 3	1	128 $×$ 128 $×$ 64
9	Conv 3 $×$ 3	2	16 $×$ 16 $×$ 512	9	Conv 3 $×$ 3	1	128 $×$ 128 $×$ 64
10	Conv 3 $×$ 3	1	16 $×$ 16 $×$ 512	10	T-Conv 2 $×$ 2	2	256 $×$ 256 $×$ 32
				11	Conv 3 $×$ 3	1	256 $×$ 256 $×$ 32
				12	Conv 3 $×$ 3	1	256 $×$ 256 $×$ 32
				13	Conv 1 $×$ 1	1	256 $×$ 256 $×$ 1

表1

图2

图3

图4

图5

图6

图7

图8

表2

图9

表3

参考文献 28

[1]	赫晓慧, 李代栋, 李盼乐, 等. 基于EDRNet模型的高分辨率遥感影像道路提取[J]. 计算机工程, 2021, 47(9):297-303,312.
	[ He X H, Li D D, Li P L, et al. Road extraction from high resolution remote sensing images based on EDRNet model[J]. Computer Engineering, 2021, 47(9):297-303,312. ] DOI: 10.19678/j.issn.1000-3428.0058876. doi: 10.19678/j.issn.1000-3428.0058876
[2]	贺浩, 王仕成, 杨东方, 等. 基于Encoder-Decoder网络的遥感影像道路提取方法[J]. 测绘学报, 2019, 48(3):330-338.
	[ He H, Wang S C, Yang D F, et al. An road extraction method for remote sensing image based on Encoder-Decoder network[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(3):330-338. ] DOI: 10.11947/j.AGCS.2019.20180005. doi: 10.11947/j.AGCS.2019.20180005
[3]	罗庆洲, 尹球, 匡定波. 光谱与形状特征相结合的道路提取方法研究[J]. 遥感技术与应用, 2007(3):339-344.
	[ Luo Q Z, Yi Q, Kuang D B. Research on extracting road based on its spectral feature and shape feature[J]. Remote Sensing Technology and Application, 2007(3):339-344. ] DOI: 10.11873/j.issn.1004-0323.2007.3.339. doi: 10.11873/j.issn.1004-0323.2007.3.339
[4]	吴学文, 徐涵秋. 一种基于水平集方法提取高分辨率遥感影像中主要道路信息的算法[J]. 宇航学报, 2010, 31(5):1495-1502.
	[ Wu X W, Xu H Q, et al. Level set method major roads information extract from high-resolution remote-sensing imagery[J]. Journal of Astronautics, 2010, 31(5):1495-1502. ] DOI: 10.3873/j.issn.1000-1328.2010.05.038 doi: 10.3873/j.issn.1000-1328.2010.05.038
[5]	赵文智, 雒立群, 郭舟, 等. 光谱特征分析的城市道路提取[J]. 光谱学与光谱分析, 2015, 35(10):2814-2819.
	[ Zhao W Z, Luo L Q, Guo Z, et al. Road extraction in remote sensing images based on spectral and edge analysis[J]. Spectroscopy and Spectral Analysis, 2015, 35(10):2814-2819.] DOI: 10.3964/j.issn.1000-0593(2015)10-2814-06 doi: 10.3964/j.issn.1000-0593(2015)10-2814-06
[6]	王建华, 秦其明, 高中灵, 等. 加入空间纹理信息的遥感图像道路提取[J]. 湖南大学学报(自然科学版), 2016, 43(4):153-156.
	[ Wang J H, Qing Q M, Gao Z L, et al. Road extraction from high-resolution remote sensing imagery by including spatial texture feature[J]. Journal of Hunan University(Natural Sciences), 2016, 43(4):153-156. ] DOI: 10.16339/j.cnki.hdxbzkb.2016.04.021 doi: 10.16339/j.cnki.hdxbzkb.2016.04.021
[7]	Mnih V, Hinton G E. Learning to detect roads in high-resolution aerial images [C]//European Conference on Computer Vision. Springer, Berlin, Heidelberg, 2010:210-223. DOI: 10.1007/978-3-642-15567-3_16 doi: 10.1007/978-3-642-15567-3_16
[8]	Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[J]. Advances in Neural Information Processing Systems, 2012, 25:1097-1105. DOI: 10.1145/3065386 doi: 10.1145/3065386
[9]	Li Y, Zhang H, Xue X, et al. Deep learning for remote sensing image classification: A survey[J]. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2018, 8(6):e1264. DOI: 10.1002/widm.1264 doi: 10.1002/widm.1264
[10]	Zhang Z X, Liu Q J, Wang Y H. Road Extraction by Deep Residual U-Net[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(5):749-753. DOI: 10.1109/LGRS.2018.2802944 doi: 10.1109/LGRS.2018.2802944
[11]	He K M, Zhang X, Ren S, et al. Deep residual learning for image recognition [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016:770-778. DOI: 10.1109/CVPR.2016.90. doi: 10.1109/CVPR.2016.90
[12]	Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation [C]// International Conference on Medical Image Computing and Computer-assisted Intervention. Springer, Cham, 2015:234-241. DOI: 10.1007/978-3-319-24574-4_28 doi: 10.1007/978-3-319-24574-4_28
[13]	Zhou L C, Zhang C, Wu M. D-linknet: Linknet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2018:182-186. DOI: 10.1109/CVPRW.2018.00034 doi: 10.1109/CVPRW.2018.00034
[14]	Chaurasia A, Culurciello E. Linknet: Exploiting encoder representations for efficient semantic segmentation [C]//2017 IEEE Visual Communications and Image Processing (VCIP). IEEE, 2017:1-4. DOI: 10.1109/VCIP.2017.8305148 doi: 10.1109/VCIP.2017.8305148
[15]	Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions. [C]//4th International Conference on Learning Representations. San Juan, Puerto Rico: ICLR, 2016.
[16]	金飞, 王龙飞, 刘智, 等. 一种双U-Net的遥感影像道路提取方法[J]. 测绘科学技术学报, 2019,36(4):377-381+387.
	[ Jin F, Wang L F, Liu Z, et al. Double U-Net remote sensing image road extraction method[J]. Journal of Geomatics Science and Technology, 2019, 36(4):377-381,387. ] DOI: 10.3969/j.issn.1673-6338.2019.04.009 doi: 10.3969/j.issn.1673-6338.2019.04.009
[17]	Abdollahi A, Pradhan B, Alamri A. VNet: An end-to-end fully convolutional neural network for road extraction from high-resolution remote sensing data[J]. IEEE Access, 2020, 8:179424-179436. DOI: 10.1109/ACCESS.2020.3026658 doi: 10.1109/ACCESS.2020.3026658
[18]	Zhou M, Sui H, Chen S, et al. BT-RoadNet: A boundary and topologically-aware neural network for road extraction from high-resolution remote sensing imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 168:288-306. DOI: 10.1016/j.isprsjprs.2020.08.019 doi: 10.1016/j.isprsjprs.2020.08.019
[19]	肖昌城, 吴锡. 基于门控卷积残差网络的卫星图像道路提取[J]. 计算机应用研究, 2021, 38(12):3820-3825.
	[ Xiao C C, Wu X. Road extraction from satellite image based on gated convolutional residual network[J]. Application Research of Computer, 2021, 38(12):3820-3825. ] DOI: 10.19734/j.issn.1001-3695.2021.03.0123 doi: 10.19734/j.issn.1001-3695.2021.03.0123
[20]	He K M, Zhang X, Ren S, et al. Identity mappings in deep residual networks [C]//European Conference on Computer Vision. Springer, Cham, 2016: 630-645. DOI: 10.1007/978-3-319-46493-0_38 doi: 10.1007/978-3-319-46493-0_38
[21]	Hu J, Shen L, Sun G. Squeeze-and-excitation networks [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018:7132-7141. DOI: 10.1109/TPAMI.2019.2913372 doi: 10.1109/TPAMI.2019.2913372
[22]	Wang Q, Wu B, Zhu P, et al. ECA-Net: efficient channel attention for deep convolutional neural networks, 2020 IEEE [C]//CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. 2020. DOI: 10.1109/CVPR42600.2020.01155 doi: 10.1109/CVPR42600.2020.01155
[23]	Roy A G, Navab N, Wachinger C. Concurrent spatial and channel ‘squeeze & excitation’in fully convolutional networks [C]//International Conference on Medical Image Computing and Computer-assisted Intervention. Springer, Cham, 2018:421-429. DOI: 10.1007/978-3-030-00928-1_48 doi: 10.1007/978-3-030-00928-1_48
[24]	宋廷强, 刘童心, 宗达, 等. 改进U-Net网络的遥感影像道路提取方法研究[J]. 计算机工程与应用, 2021, 57(14):209-216.
	[ Song T Q, Liu T X, Zong D, et al. Research on road extraction method from remote sensing images based on improved U-Net network[J]. Computer Engineering and Applications, 2021, 57(14):209-216. ] DOI: 10.3778/j.issn.1002-8331.2007-0392 doi: 10.3778/j.issn.1002-8331.2007-0392
[25]	李君. 基于多尺度卷积神经网络的遥感图像道路提取研究[D]. 湘潭:湘潭大学, 2019.
	[ Li J. Multi-scale Convolutional Neural Network for Road Extraction in Remote Sensing Imagery[D]. Xiangtan: Xiangtan University, 2019. ] DOI: 10.27426/d.cnki.gxtdu.2019.001567 doi: 10.27426/d.cnki.gxtdu.2019.001567
[26]	罗咏潭. 基于深度学习的遥感图像道路提取与语义分割[D]. 厦门:厦门大学, 2019.
	[ Luo Y T. Road extraction and semantic segmentation from remote sensing image based on deep learning[D]. Xiamen: Xiamen University, 2019. ]
[27]	Demir I, Koperski K, Lindenbaum D, et al. Deepglobe 2018: A challenge to parse the earth through satellite images [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2018: 172-181. DOI: 10.1109/CVPRW.2018.00031 doi: 10.1109/CVPRW.2018.00031
[28]	Wang J, Perez L. The effectiveness of data augmentation in image classification using deep learning[J]. Convolutional Neural Networks Vis. Recognit, 2017, 11:1-8.

模型	IoU	Recall	Precision	F1-score	FLOPs(G)	Params(M)
U-Net^[12]	0.6521	0.7750	0.8044	0.7894	13.73	7.77
ResUnet^[10]	0.6498	0.7748	0.8011	0.7877	14.43	8.12
VNet^[17]	0.6686	0.7938	0.8091	0.8014	44.85	36.00
DAResUnet	0.6690	0.7914	0.8122	0.8016	14.44	8.12
DAFCResUnet	0.6713	0.7982	0.8085	0.8033	14.46	8.12

模型	IoU	Recall	Precision	F1-score	FLOPs(G)	Params(M)
U-Net^[12]	0.7182	0.8050	0.8695	0.8360	13.73	7.77
ResUnet^[10]	0.7272	0.8109	0.8756	0.8420	14.43	8.12
VNet^[17]	0.7330	0.8196	0.8741	0.8460	44.85	36.00
DAResUnet	0.7383	0.8258	0.8745	0.8494	14.44	8.12
DAFCResUnet	0.7402	0.8226	0.8808	0.8507	14.46	8.12

面向高分遥感影像道路提取的轻量级双注意力和特征补偿残差网络模型

A Lightweight Dual Attention and Feature Compensated Residual Network Model for Road Extraction from High-Resolution Remote Sensing Images

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 28

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	蒯宇, 王彪, 吴艳兰, 陈搏涛, 陈兴迪, 薛维宝. 基于多尺度特征感知网络的城市植被无人机遥感分类[J]. 地球信息科学学报, 2022, 24(5): 962-980.
[2]	杨先增, 周亚男, 张新, 李睿, 杨丹. 融合边缘特征与语义信息的人工坑塘精准提取方法[J]. 地球信息科学学报, 2022, 24(4): 766-779.
[3]	龙怡灿, 雷蓉, 董杨, 李东子, 赵琛琛. 基于YOLOv5算法的飞机类型光学遥感识别[J]. 地球信息科学学报, 2022, 24(3): 572-582.
[4]	周欣昕, 吴艳兰, 李梦雅, 郑智腾. 基于特征分离机制的深度学习植被自动提取方法[J]. 地球信息科学学报, 2021, 23(9): 1675-1689.
[5]	李国清, 柏永青, 杨轩, 陈正超, 余海坤. 基于深度学习的高分辨率遥感影像土地覆盖自动分类方法[J]. 地球信息科学学报, 2021, 23(9): 1690-1704.
[6]	刘戈, 姜小光, 唐伯惠. 特征优选与卷积神经网络在农作物精细分类中的应用研究[J]. 地球信息科学学报, 2021, 23(6): 1071-1081.
[7]	戴激光, 王晓桐, 智新宇, 马榕辰, 张依蕾. 动态权重约束下的农村机耕路提取方法[J]. 地球信息科学学报, 2021, 23(5): 773-784.
[8]	许泽宇, 沈占锋, 李杨, 柯映明, 李硕, 王浩宇, 焦淑慧. 结合模糊度和形态学指数约束的深度学习建筑物提取[J]. 地球信息科学学报, 2021, 23(5): 918-927.
[9]	唐璎, 刘正军, 杨懿, 顾海燕, 杨树文. 基于特征增强和ELU的神经网络建筑物提取研究[J]. 地球信息科学学报, 2021, 23(4): 692-709.
[10]	朱盼盼, 李帅朋, 张立强, 李洋. 基于多任务学习的高分辨率遥感影像建筑提取[J]. 地球信息科学学报, 2021, 23(3): 514-523.
[11]	郭紫甜, 王春梅, 刘欣, 庞国伟, 朱梦阳, 王晋卿. 基于小流域抽样单元的中国FROM-GLC30数据精度评价[J]. 地球信息科学学报, 2021, 23(3): 524-535.
[12]	李传林, 黄风华, 胡威, 曾江超. 基于Res_AttentionUnet的高分辨率遥感影像建筑物提取方法[J]. 地球信息科学学报, 2021, 23(12): 2232-2243.
[13]	冯叶涵, 陈亮, 贺晓冬. 基于百度街景的SVF计算及其在城市热岛研究中的应用[J]. 地球信息科学学报, 2021, 23(11): 1998-2012.
[14]	储国中, 李蒙蒙, 汪小钦. 融合高度特征的高分遥感影像多尺度城市建筑类型分类[J]. 地球信息科学学报, 2021, 23(11): 2073-2085.
[15]	陆大进, 黎东, 朱笑笑, 聂胜, 周国清, 张兴忆, 杨超. 基于卷积神经网络的ICESat-2光子点云去噪分类[J]. 地球信息科学学报, 2021, 23(11): 2086-2095.