DeepLabv3+语义分割模型的济南市防尘绿网提取及时空变化分析

刘春亭; 冯权泷; 刘建涛; 王莹; 史同广; 李毅; 龚建华; 赵辉辉

doi:10.11834/jrs.20220101

模型与方法 | 浏览量 : 0 下载量: 348 CSCD: 1 更多指标

R-PDF
PDF
导出
分享
收藏
专辑

DeepLabv3+语义分割模型的济南市防尘绿网提取及时空变化分析
Urban green plastic cover extraction and spatial pattern changes in Jinan city based on DeepLabv3+ semantic segmentation model
2022年26卷第12期页码：2518-2530
纸质出版日期： 2022-12-07 ，
DOI： 10.11834/jrs.20220101

扫描看全文

刘春亭，冯权泷，刘建涛，王莹，史同广，李毅，龚建华，赵辉辉.2022.DeepLabv3+语义分割模型的济南市防尘绿网提取及时空变化分析.遥感学报，26（12）： 2518-2530

Liu C T， Feng Q L， Liu J T， Wang Y， Shi T G， Li Y， Gong J H and Zhao H H. 2022. Urban green plastic cover extraction and spatial pattern changes in Jinan city based on DeepLabv3+ semantic segmentation model. National Remote Sensing Bulletin， 26（12）：2518-2530
刘春亭，冯权泷，刘建涛，王莹，史同广，李毅，龚建华，赵辉辉.2022.DeepLabv3+语义分割模型的济南市防尘绿网提取及时空变化分析.遥感学报，26（12）： 2518-2530 DOI： 10.11834/jrs.20220101.

Liu C T， Feng Q L， Liu J T， Wang Y， Shi T G， Li Y， Gong J H and Zhao H H. 2022. Urban green plastic cover extraction and spatial pattern changes in Jinan city based on DeepLabv3+ semantic segmentation model. National Remote Sensing Bulletin， 26（12）：2518-2530 DOI： 10.11834/jrs.20220101.

摘要

防尘防护绿网（防尘绿网）作为抑尘的主要措施被建筑工地广泛使用。快速获取防尘绿网覆盖及时空变化信息，对防尘抑尘、生态环境保护措施的制定具有重要指导意义。本文基于Sentinel-2时间序列遥感影像，使用DeepLabv3+深度学习语义分割模型生成了济南市中心城区2016年—2020年逐年防尘绿网数据，随后利用景观格局、重心-标准差椭圆等方法分析了其空间分布特征和时空扩张趋势。研究结果表明：（1）本文方法的分割精准度、召回率、F1值、IoU分别为84.05%、80.09%、0.82、69.72%，可快速准确地提取城市防尘绿网进而实现大范围、长时序的防尘绿网动态监测。（2）相较于传统的遥感分类方法和其他语义分割模型，本方法的精度最优，防尘绿网的提取结果更精细；另外对北京和天津市的提取实验也证实了本模型的迁移能力。（3）2016年—2020年济南市防尘绿网铺设范围明显扩张，斑块面积和数量不断增多，破碎程度和形状复杂度不断增强，平均斑块面积增大，斑块间凝聚度和聚集度呈波动状态，景观格局复杂且不稳定；绿网铺设范围的扩张呈现出明显的方向性。防尘绿网的分布形态与动态扩张变化受到城市规划、项目进程等人为因素影响，城市规划决定了防尘绿网的分布，而防尘绿网的使用现状在一定程度上又反映了城市建设进程。因此，利用遥感手段对城市防尘绿网进行动态监测与管理，可为城市规划、生态环境建设和城市精细管理提供数据和技术支持，对城市扩张模式与城市重建工作管理具有重要意义。

Abstract

Green plastic covers have been widely used as the primary method of dust prevention in construction sites. The rapid acquisition of green plastic cover and spatial–temporal change information has an important guiding significance for the formulation of dust prevention and ecological environmental protection measures. This work extracted the covered area of green plastic cover by using DeepLabv3+ semantic segmentation model based on Sentinel-2 remote sensing data and realized the annual green plastic cover segmentation and extraction in Jinan from 2016 to 2020. The spatial distribution characteristics and spatial–temporal expansion trend of green plastic cover were analyzed by area statistics

landscape pattern analysis

and mean center-standard deviation ellipse. Results show that: (1) According to the accuracy evaluation

the proposed architecture reaches an acceptable accuracy

with 84.05% precision

80.09% recall

0.82 F1 score

and 69.72 IoU

which could quickly and accurately extract the urban green plastic cover and realize the large-scale and time series dynamic monitoring and management. (2) The accuracy of DeepLabv3+ model used in this work is the best compared with the traditional remote sensing classification method and other sematic segmentation models

and the extraction results of green plastic cover are more precise. In addition

the extraction experiments in Beijing and Tianjin also confirmed the migration ability of the model. (3) The laying range of green plastic cover significantly expanded from 2016 to 2020. The area and number of green plastic cover patches

fragmentation degree

and shape complexity increased. The average patch area increased

the cohesion and aggregation between patches fluctuated

and the landscape pattern was complex and unstable. The expansion has an obvious direction. The distribution and dynamic expansion of green plastic cover are affected by human factors

such as urban planning and project process. Urban planning determines the distribution of green plastic cover

and the current use of green plastic cover reflects the process of urban construction to a certain extent. Therefore

the dynamic monitoring and management of urban green plastic cover by remote sensing means can provide data and technical support for urban planning

ecological environment construction

and urban accurate management

which are of great significance to the management of urban expansion mode and reconstruction.

关键词

防尘绿网遥感语义分割DeepLabv3+Sentinel-2时空变化

Keywords

green plastic coverremote sensingsemantic segmentationDeepLabv3+Sentinel-2temporal and spatial variation

references

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.2016.264 4615http://dx.doi.org/10.1109/TPAMI.2016.2644615]

Breiman L. 2001. Random forests. Machine Learning, 45(1): 5-32 [DOI: 10.1023/A:1010933404324http://dx.doi.org/10.1023/A:1010933404324]

Chen L C, Papandreou G, Kokkinos I, Murphy K and Yuille A L. 2014. Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs. arXiv preprint arXiv: 1412.7062

Chen L C, Papandreou G, Kokkinos I, Murphy K and Yuille A L. 2018a. 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, Schrof F and Adam H. 2017. Rethinking Atrous Convolution for Semantic Image Segmentation. arXiv preprint arXiv: 1706.05587

Chen L C, Zhu Y K, Papandreou G, Schrof F and Adam H. 2018b. Encoder-Decoder with atrous separable convolution for semantic image segmentation//Proceedings of 15th European Conference on Computer Vision (ECCV). Munich, Germany. 11211:833-851 [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//Proceedings of the Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE: 1800-1807 [DOI: 10.1109/CVPR.2017.195http://dx.doi.org/10.1109/CVPR.2017.195]

Cortes C and Vapnik V. 1995. Support-vector networks. Machine Learning, 20(3): 273-297 [DOI:10.1007/BF00994018http://dx.doi.org/10.1007/BF00994018]

European Space Agency. 2015. ESA Standard Document[S/OL]. https://sentinel.esa.int/documents/247904/685211/Sentinel-2_User_Handbook, 2015-07-24/2021-04-25

欧洲航天局. 2015. ESA标准文件[S/OL]. https://sentinel.esa.int/documents/247904/685211/Sentinel-2_User_Handbook, 2015-07-24/2021-04-25

Feng Q L, Niu B W, Zhu D H, Yao X C, Liu Y M, Ou C, Chen B A, Yang J Y, Guo H and Liu J T. 2021. National agricultural plastic greenhouses remote sensing classification data set in 2019. v1. Science Data Bank. http://www.dx.doi.org/10.11922/sciencedb.j00001.00230http://www.dx.doi.org/10.11922/sciencedb.j00001.00230

冯权泷, 牛博文, 朱德海, 姚晓闯, 刘逸铭, 欧聪, 陈泊安, 杨建宇, 郭浩, 刘建涛. 2021. 2019年全国农业塑料大棚遥感分类数据集. v1. Science Data Bank. http://www.dx.doi.org/10.11922/sciencedb.j000 01.00230http://www.dx.doi.org/10.11922/sciencedb.j00001.00230

Feng Q L, Yang J Y, Zhu D H, Liu J T, Guo H, Batsaikhan B and Li B G. 2019. Integrating multitemporal Sentinel-1/2 data for coastal land cover classification using a multibranch convolutional neural network: a case of the Yellow River Delta. Remote Sensing, 11(9): 1006 [DOI: 10.3390/rs11091006http://dx.doi.org/10.3390/rs11091006]

Gu X T, Gao X H, Ma H J, Shi F F, Liu X M and Cao X M. 2019. Comparison of machine learning methods for land use/land cover classification in the complicated terrain regions. Remote Sensing Technology and Application, 34(1): 57-67

谷晓天, 高小红, 马慧娟, 史飞飞, 刘雪梅, 曹晓敏. 2019. 复杂地形区土地利用/土地覆被分类机器学习方法比较. 遥感技术与应用, 34(1): 57-67 [DOI: 10.11873/j.issn.1004-0323.2019.1.0057http://dx.doi.org/10.11873/j.issn.1004-0323.2019.1.0057]

Jinan Bureau of Statistics. 2019. Statistical Communique of Jinan on 2019 National Economic and Social Development. http://jntj.jinan.gov.cnhttp://jntj.jinan.gov.cn

济南市统计局. 2019. 2019年济南市国民经济和社会发展统计公报. http://jntj.jinan.gov.cn/http://jntj.jinan.gov.cn/

Jinan Municipal Government. 2016. The Master Plan of Jinan(2011—2020). http://www.jinan.gov.cn/art/2016/8/29/art_28170_334992 7.htmlhttp://www.jinan.gov.cn/art/2016/8/29/art_28170_3349927.html

济南市人民政府. 2016. 济南市城市总体规划（2011—2020年）. http://www.jinan.gov.cn/art/2016/8/29/art_28170_3349 927.htmlhttp://www.jinan.gov.cn/art/2016/8/29/art_28170_3349927.html

Jinan Municipal Government. 2018. Jinan City Annual Report on Government Information Disclosure in 2017. http://www.jinan.gov.cn/art/2018/2/23/art_2616_2046904.htmlhttp://www.jinan.gov.cn/art/2018/2/23/art_2616_2046904.html

济南市人民政府. 2018. 济南市2017年政府信息公开工作年度报告. http://www.jinan.gov.cn/art/2018/2/23/art_2616_2046904.htmlhttp://www.jinan.gov.cn/art/2018/2/23/art_2616_2046904.html

Li D R, Yu H R and Li X. 2017. The spatial-temporal pattern analysis of city development in countries along the belt and road initiative based on nighttime light data. Geomatics and Information Science of Wuhan University, 42(6): 711-720

李德仁, 余涵若, 李熙. 2017. 基于夜光遥感影像的“一带一路”沿线国家城市发展时空格局分析. 武汉大学学报(信息科学版), 42(6): 711-720 [DOI: 10.13203/j.whugis20170100http://dx.doi.org/10.13203/j.whugis20170100]

Li P P. 2020. Dust pollution damage analysis and control methods of building construction. Environmental Science and Management, 45(11): 84-88

李萍萍. 2020. 建筑施工扬尘污染损害分析及其控制方法研究. 环境科学与管理, 45(11): 84-88 [DOI: CNKI:SUN:BFHJ.0.2020-11-018http://dx.doi.org/CNKI:SUN:BFHJ.0.2020-11-018]

Li S Q. 2020. Identification of construction waste from multi-source remote sensing image based on transfer learning. Beijing: Beijing University of Civil Engineering and Architecture (李思琦. 2020. 基于迁移学习的多源遥感影像建筑垃圾识别. 北京:北京建筑大学)

Liu J T, Feng Q L, Wang Y, Batsaikhan B, Gong J H, Li Y, Liu C T and Ma Y. 2020. Urban green plastic cover mapping based on VHR remote sensing images and a deep semi-supervised learning framework. ISPRS International Journal of Geo-Information, 9(9): 527 [DOI:10.3390/ijgi9090527http://dx.doi.org/10.3390/ijgi9090527]

Liu P H, Liu X P, Liu M X, Shi Q, Yang J X, Xu X C and Zhang Y Y. 2019. Building footprint extraction from high-resolution images via spatial residual inception convolutional neural network. Remote Sensing, 11(7):830 [DOI: 10.3390/rs11070830http://dx.doi.org/10.3390/rs11070830]

Liu W Y, Yue A Z, Ji J, Shi W H, Deng R R, Liang Y H and Xiong L H. 2020. Urban green space extraction from GF-2 remote sensing image based on DeepLabv3+ semantic segmentation model. Remote Sensing for Land & Resources, 32(2): 120-129

刘文雅, 岳安志, 季珏, 师卫华, 邓孺孺, 梁业恒, 熊龙海. 2020. 基于DeepLabv3+语义分割模型的GF-2影像城市绿地提取. 国土资源遥感, 32(2): 120-129 [DOI: 10.6046/gtzyyg.2020.02.16http://dx.doi.org/10.6046/gtzyyg.2020.02.16]

Liu Y L, Ren Y H, Wei C J, Wang A H, Zhou H Z and Chi Y B. 2009. Study on monitoring of informal open-air solid waste dumps based on Beijing-1 images. Journal of Remote Sensing, 13(2): 320-326

刘亚岚, 任玉环, 魏成阶, 汪爱华, 周会珍, 迟耀斌. 2009. 北京1号小卫星监测非正规垃圾场的应用研究. 遥感学报, 13(2): 320-326 [DOI: 10.11834/jrs.20090255http://dx.doi.org/10.11834/jrs.20090255]

Long J, Shelhamer E and Darrell T. 2015. Fully convolutional networks for semantic segmentation//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA: IEEE Computer Society: 3431-3440 [DOI: 10.1109/CVPR.2015.7298965http://dx.doi.org/10.1109/CVPR.2015.7298965]

Ming D P, Li J, Wang J Y, Zhang M. 2015. Scale parameter selection by spatial statistics for GeOBIA: Using mean-shift based multi-scale segmentation as an example. ISPRS Journal of Photogrammetry and Remote Sensing, 106: 28-41 [DOI: 10.1016/j.isprsjprs.2015.04.010http://dx.doi.org/10.1016/j.isprsjprs.2015.04.010]

Panboonyuen T, Jitkajornwanich K, Lawawirojwong S, Srestasathiern P and Vateekul P. 2019. Semantic segmentation on remotely sensed images using an enhanced global convolutional network with channel attention and domain specific transfer learning. Remote Sensing, 11(1):83 [DOI: 10.3390/rs11010083http://dx.doi.org/10.3390/rs11010083]

Rodriguez-Galiano, V F, Ghimire B, Rogan J, Chica-Olmo M and Rigol-Sanchez J P. 2012. An assessment of the effectiveness of a random forest classifier for land-cover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67: 93-104 [DOI 10.1016/j.isprsjprs.2011.11.002:http://dx.doi.org/10.1016/j.isprsjprs.2011.11.002:]

Ronneberger O, Fischer P and Brox T. 2015. U-Net: convolutional net-works for biomedical image segmentation//Proceedings of the 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]

Seydi S T, Akhoondzadeh M, Amani M and Mahdavi S. 2021. Wildfire damage assessment over Australia using Sentinel-2 imagery and MODIS land cover product within the google earth engine cloud platform. Remote Sensing, 13(2): 220-220 [DOI: 10.3390/RS13020220http://dx.doi.org/10.3390/RS13020220]

Shao Z F, Zhou Z F, Huang X and Zhang Y. 2021. MRENet: Simultaneous Extraction of Road Surface and Road Centerline inComplex Urban Scenes from Very High-Resolution Images. Remote Sensing, 13(2): 239 [DOI:10.3390/rs13020239http://dx.doi.org/10.3390/rs13020239]

Shrestha S and Vanneschi L. 2018. Improved fully convolutional network with conditional random fields for building extraction. Remote Sensing, 10(7): 1135 [DOI:10.3390/rs10071135http://dx.doi.org/10.3390/rs10071135]

Silvestri S and Omri M. 2008. A method for the remote sensing identification of uncontrolled landfills: formulation and validation. International Journal of Remote Sensing, 29(4): 975-989 [DOI:10. 1080/01431160701311317http://dx.doi.org/10.1080/01431160701311317]

Wang C, Yin S J, Meng B, Ma W L, Zhu L and Wu C Q. 2016. Analysis of the non-regular garbage sites in the region of Beijing, Tianjin and Hebei using remote sensing monitoring images. High Technology Letters, 26(Z1): 799-807

王晨, 殷守敬, 孟斌, 马万栋, 朱利, 吴传庆. 2016. 京津冀地区非正规垃圾场地遥感监测分析. 高技术通讯, 26(Z1): 799-807 [DOI: 10.3772/j.issn. 1002-0470.2016.08-09.013http://dx.doi.org/10.3772/j.issn.1002-0470.2016.08-09.013]

Wang J Q, Li J S, Zhou H C and Zhang X. 2019. Typical element extraction method of remote sensing image based on Deeplabv3+ and CRF. Computer Engineering, 45(10):206-265+271

王俊强, 李建胜, 周华春, 张旭. 2019. 基于Deeplabv3+与CRF的遥感影像典型要素提取方法. 计算机工程, 45(10): 260-265+271 [DOI: 10.19678/j.issn.1000-3428.0053359http://dx.doi.org/10.19678/j.issn.1000-3428.0053359]

Wu J G. 2000. Landscape ecology pattern process, scale and hierarchy. Beijing: High Education Press

邬建国. 2000. 景观生态学：格局、过程、尺度与等级. 北京:高等教育出版社

Wu Q, Zhong R F, Zhao W J, Song K and Du L M. 2019. Land-cover classification using GF-2 images and airborne lidar data based on random forest. International Journal of Remote Sensing, 40(5-6): 2410-2426 [DOI: 10.1080/01431161.2018.1483090http://dx.doi.org/10.1080/01431161.2018.1483090]

Yan W Y, Mahendrarajah P, Shaker A, Faisal K, Luong R and Al-Ahmad M. 2014. Analysis of multi-temporal landsat satellite images for monitoring land surface temperature of municipal solid waste disposal sites. Environmental Monitoring and Assessment, 186(12): 8161-8173 [DOI:10.1007/s10661-014-3995-zhttp://dx.doi.org/10.1007/s10661-014-3995-z]

Yi Z W, Jia L and Chen Q T. 2020. Crop classification using multi-temporal Sentinel-2 data in the Shiyang River basin of China. Remote Sensing, 12(24):4052 [DOI: 10.3390/RS12244052http://dx.doi.org/10.3390/RS12244052]

Zhang J X, Gu H Y, Yang Y, Zhang H and Li H T. 2021. Research process and trend of high-resolution remote sensing imagery intelligent interpretation. National Remote Sensing Bulletin, 25(11): 2198-2210

张继贤, 顾海燕, 杨懿, 张鹤, 李海涛. 2021. 高分辨率遥感影像智能解译研究进展与趋势. 遥感学报, 25(11): 2198-2210 [DOI:10.11834/jrs.2021038http://dx.doi.org/10.11834/jrs.2021038]

Zhang T X, Su J Y, Xu Z Y, Luo Y L and Li J Y. 2021. Sentinel-2 satellite imagery for urban land cover classification by optimized random forest classifier. Applied Sciences, 11(2): 543 [DOI: 10.3390/APP11020543http://dx.doi.org/10.3390/APP11020543]

Zhang X Q. 2020. Object oriented remote sensing information extraction of construction waste in wuhou district, Chengdu. Chengdu: Chengdu University of Technology

张兴琼. 2020. 面向对象的成都市武侯区建筑垃圾遥感信息提取. 成都:成都理工大学

Zhang Z X, Liu Q J and Wang Y H. 2018. Road extraction by deep residual U-Net. IEEE Geoscience and Remote Sensing Letters, 15(5):749-753 [DOI:10.1109/LGRS.2018.2802944http://dx.doi.org/10.1109/LGRS.2018.2802944]

Zhu Y N, Gao T, Wang S D, Zhou L and Du M Y. 2021. Automatic recognition method of construction waste based on transfer learning and retraining model and high-score remote sensing data. Remote Sensing Technology and Application, 36(2): 314-323

祝一诺, 高婷, 王术东, 周磊, 杜明义. 2021. 基于迁移学习再训练模型和高分遥感数据的建筑垃圾自动识别方法. 遥感技术与应用, 36(2): 314-323 [DOI: 10.11873/j.issn.1004-0323.2021.2.0314http://dx.doi.org/10.11873/j.issn.1004-0323.2021.2.0314]

文章被引用时，请邮件提醒。

提交

边缘增强的BECU-Net模型高分辨率遥感影像耕地提取

基于哨兵二号数据的矿区叶面降尘信息提取及尘源识别

高寒湿地分类的遥感特征优选研究

物候窗口和多源中高分辨率影像的稻虾田提取

利用Sentinel-2影像超分辨率重建的红树林冠层氮含量反演