基于改进的DUNet遥感图像道路提取

doi:10.6040/j.issn.1672-3961.0.2021.325

山东大学学报 (工学版) ›› 2022, Vol. 52 ›› Issue (4): 29-37.doi: 10.6040/j.issn.1672-3961.0.2021.325

• • 上一篇

基于改进的DUNet遥感图像道路提取

侯月武¹,刘兆英¹,张婷^1*,李玉鑑^1,2,孙长明³

1. 北京工业大学信息学部, 北京 100124;2. 桂林电子科技大学人工智能学院, 广西桂林 541004;3. 新南威尔士大学计算机技术与工程学院, 新南威尔士州悉尼 201101

发布日期:2022-08-24
作者简介:侯月武(1997— ),男,山东潍坊人,硕士研究生,主要研究方向为深度学习遥感图像道路提取. E-mail:houyw@ccitrobot.com. *通信作者简介:张婷(1986— ),女,河南郑州人,讲师,主要研究方向为模式识别、深度学习以及图像处理. E-mail:zhangting@bjut.edu.cn
基金资助:
国家自然科学基金资助项目(61806013,61876010,61906005);北京市教育委员会科技计划一般资助项目(KM202110005028);北京工业大学交叉科学研究院资助项目(2021020101);北京工业大学国际科研合作种子基金资助项目(2021A01)

Road extraction from remote sensing images based on improved DUNet

HOU Yuewu¹, LIU Zhaoying¹, ZHANG Ting^1*, LI Yujian^1,2, SUN Changming³

1. Information Technology of Faulty, Beijing University of Technology, Beijing 100124, China;
2. School of Artificial Intelligence, Guilin University of Electronic Technology, Guilin 541004, Guangxi, China;
3. School of Computer Science and Engineering, University of New South Wales, Sydney 201101, New South Wales, Australia

Published:2022-08-24

摘要/Abstract

摘要： 为进一步提高遥感图像道路提取的精度,提出一种改进的DUNet遥感图像道路提取方法。在编码器部分,为使网络关注道路信息,在第3个池化层分别使用有注意力机制和没有注意力机制两个分支提取道路特征;在解码器部分,同时使用传统UNet的解码器和DUNet解码器两个分支进行上采样,最大限度减少信息丢失。试验结果表明,与其他8种常用的分割模型结果相比,此方法在Massachusetts和DeepGlobe 2018数据集上都获得最高的平均交并比和平均Dice系数,其中平均交并比最高分别提高2.90%和8.99%,平均Dice系数最高分别提高2.53%和7.66%。这表明改进的DUNet能够有效实现遥感图像的道路提取,与传统DUNet相比,在小路区域的分割效果得到提升,进一步提高了传统DUNet的分割精度。

关键词: 遥感图像, 道路提取, 多尺度上采样, 注意力机制

中图分类号:

TP18

侯月武,刘兆英,张婷,李玉鑑,孙长明. 基于改进的DUNet遥感图像道路提取[J]. 山东大学学报 (工学版), 2022, 52(4): 29-37.

HOU Yuewu, LIU Zhaoying, ZHANG Ting, LI Yujian, SUN Changming. Road extraction from remote sensing images based on improved DUNet[J]. Journal of Shandong University(Engineering Science), 2022, 52(4): 29-37.

参考文献

[1] MENA J B. State of the art on automatic road extraction for GIS update: a novel classification[J]. Pattern Recognition Letters, 2003, 24(16): 3037-3058.
[2] LI Y, XU L, PIAO H. Semi-automatic road extraction from high-resolution remote sensing image: review and prospects[C] //Proceeding of Ninth International Conference on Hybrid Intelligent Systems. Shenyang, China: IEEE, 2009: 204-209.
[3] HEIPKE C, MAYER H, WIEDEMANN C, et al. Evaluation of automatic road extraction[J]. International Archives of Photogrammetry and Remote Sensing, 1997, 32(3): 151-160.
[4] GRUEN A, LI H. Semiautomatic road extraction by dynamic programming[C] //Proceeding of ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision. Munich, Germany: SPIE, 1994: 324-332.
[5] GRUEN A, LI H. Semi-automatic linear feature extraction by dynamic programming and LSB-snakes[J]. Photogrammetric Engineering and Remote Sensing, 1997, 63(8): 985-994.
[6] ANIL P N, NATARAJAN S. A novel approach using active contour model for semi-automatic road extraction from high resolution satellite imagery[C] //Proceeding of Second International Conference on Machine Learning and Computing. Bangalore, India: IEEE, 2010: 263-266.
[7] GEMAN D, JEDYNAK B. An active testing model for tracking roads in satellite images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(1): 1-14.
[8] HU X, ZHANG Z, ZHANG J. An approach of semiautomated road extraction from aerial image based on template matching and neural network[J]. International Archives of Photogrammetry and Remote Sensing, 2000, 33(3): 994-999.
[9] CHENG G, WANG Y, GONG Y, et al. Urban road extraction via graph cuts based probability propagation[C] //Proceeding of IEEE International Conference on Image Processing. Paris, France: IEEE, 2014: 5072-5076.
[10] MAYER H, HINZ S, BACHER U, et al. A test of automatic road extraction approaches[J]. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2006, 36(3): 209-214.
[11] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Massachusetts, USA: IEEE, 2015: 3431-3440.
[12] RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segme-ntation[C] //Proceeding of International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich, Germany: Springer Press, 2015: 234-241.
[13] CHAURASIA A, CULURCIELLO E. Linknet: exploiting encoder representations for efficient semantic segmentation[C] //Proceeding of IEEE Visual Communications and Image Processing. Florida, USA: IEEE, 2017: 1-4.
[14] ZHOU L, ZHANG C, WU M. D-linknet: linknet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction[C] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops. Utah, USA: IEEE, 2018: 182-186.
[15] FAN D P, ZHOU T, JI G P, et al. Inf-net: automatic covid-19 lung infection segmentation from ct images[J]. IEEE Transactions on Medical Imaging, 2020, 39(8): 2626-2637.
[16] ZHANG Z, LIU Q, WANG Y. Road extraction by deep residual U-net[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(5): 749-753.
[17] ZHONG Z, LI J, CUI W, et al. Fully convolutional networks for building and road extraction: preliminary results[C] //Proceeding of IEEE International Geoscience and Remote Sensing Symposium. Beijing, China: IEEE, 2016: 1591-1594.
[18] MOKHTARZADE M, ZOEJ M J V. Road detection from high-resolution satellite images using artificial neural networks[J]. International Journal of Applied Earth Observation and Geoinformation, 2007, 9(1): 32-40.
[19] MNIH Volodymyr. Machine learning for aerial image labeling[D]. Ontario, Canada: Department of Computer Science, University of Toronto, 2013.
[20] 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, 2017, 40(4): 834-848.
[21] DEMIR I, KOPERSKI K, LINDENBAUM D, et al. A challenge to parse the earth through satellite images[C] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops. Utah, USA: IEEE, 2018: 18-22.
[22] WOO S, PARK J, LEE J Y, et al. Cbam: convolutional block attention module[C] //Proceedings of European Conference on Computer Vision. Glasgow, United Kingdom: Springer Press, 2018: 3-19.
[23] POWELL W B. A unified framework for stochastic optimization[J]. European Journal of Operational Research, 2019, 275(3): 795-821.
[24] FENG D, HAASE-SCHUTZ C, ROSENBAUM L, et al. Deep multi-modal object detection and semantic segmentation for autonomous driving: datasets, methods, and challenges[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(3): 1341-1360.
[25] EELBODE T, BERTELS J, BERMAN M, et al. Optimization for medical image segmentation: theory and practice when evaluating with dice score or jaccard index[J]. IEEE Transactions on Medical Imaging, 2020, 39(11): 3679-3690.
[26] 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.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于改进的DUNet遥感图像道路提取

Road extraction from remote sensing images based on improved DUNet

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

多维度评价

本文评价

推荐阅读 0

[1]	梁晔,马楠,刘宏哲. 图像依赖的显著图融合方法[J]. 山东大学学报 (工学版), 2021, 51(4): 1-7.
[2]	曹春红,段鸿轩,曹玲,张乐乐,胡凯,肖芬. 基于多级特征级联的遥感图像实时语义分割[J]. 山东大学学报 (工学版), 2021, 51(2): 19-25.
[3]	张沁洋,李旭,姚春龙,李长吾. 结合句法依存信息的方面级情感分类[J]. 山东大学学报 (工学版), 2021, 51(2): 83-89.
[4]	张俊三,程俏俏,万瑶,朱杰,张世栋. MIRGAN: 一种基于GAN的医学影像报告生成模型[J]. 山东大学学报 (工学版), 2021, 51(2): 9-18.
[5]	张月芳,邓红霞,呼春香,钱冠宇,李海芳. 融合残差块注意力机制和生成对抗网络的海马体分割[J]. 山东大学学报 (工学版), 2020, 50(6): 76-81.
[6]	廖南星,周世斌,张国鹏,程德强. 基于类激活映射-注意力机制的图像描述方法[J]. 山东大学学报 (工学版), 2020, 50(4): 28-34.
[7]	蔡国永,贺歆灏,储阳阳. 基于空间注意力和卷积神经网络的视觉情感分析[J]. 山东大学学报 (工学版), 2020, 50(4): 8-13.
[8]	肖乔,裴继红,王荔霞,龚志成. 基于多通道Gabor滤波模糊融合的遥感图像舰船检测[J]. 山东大学学报 (工学版), 2015, 45(5): 29-35.