Extraction of road blockage information for the Jiuzhaigou earthquake based on a convolution neural network and very-high-resolution satellite images
- 19 Downloads
Road blockage information extraction from a single-phase postdisaster image is difficult because roads are narrow and easily covered by vegetation. The traditional object-oriented image analysis method is restrictive, and its detection is slow. A deep learning algorithm, i.e., the convolution neural network (CNN), is applied to rapidly extract road blockage information. An algorithm for sample generation is designed to construct a typical sample library for CNN training, and an appropriate CNN structure and a complete detection process are designed to extract road blockage information. Finally, by taking the Jiuzhaigou earthquake on August 8, 2017, as an example, experimental verification is carried out. The kappa coefficient and the F1 score of the results are 77.60% and 87.95%, respectively. The extraction of road blockages can be completed with an efficiency of 14.59 km2 per hour. The requirements for disaster emergency monitoring can be met by the accuracy and efficiency of this method, which are better than those of the traditional object-oriented method.
KeywordsConvolution neural network Very-high-resolution satellite images Road blockage information Jiuzhaigou earthquake
This study was supported by the National Key Research and Development Program of China (2016YFC0803000 and 2017YFB0504100) and the Youth Innovation Promotion Association of the Chinese Academy of Sciences (CAS) (2015129). The authors wish to thank Maxar Technologies Ltd. (DigitalGlobe, https://www.digitalglobe.com/), Twenty First Century Aerospace Technology Co., Ltd. (21AT, https://www.21at.com.cn) and China Centre for Resources Satellite Data and Application (CRESDA, http://www.cresda.com/CN/gywm/zxgk/index.shtml), for providing the satellite images, National Aeronautics and Space Administration (NASA, http://reverb.echo.nasa.gov/reverb/) for providing the new global DEM (GDEM V2) data, and Environmental Systems Research Institute Inc. (ESRI, 380 New York Street, Redlands, CA 92373, USA) for providing the OSM data. The authors are grateful to the anonymous reviewers and editors for their valuable comments and suggestions that improved this manuscript.
Shixin Wang and Futao Wang conceived and designed the experiments, and Baolin Yang performed the experiments and wrote the paper. Yi Zhou provided crucial guidance and support throughout the study. Qiao Hu, Ying Chang and Qing Zhao analyzed the data.
Compliance with ethical standards
Conflict of interest
The authors declare no conflicts of interest.
- ArcGIS Editor for OSM 10.6.x Desktop. Availale online: https://www.arcgis.com/home/item.html?id=0c4b24608fc94542ba5d1130c0606802 (Jul 5, 2018)
- Audebert, N.; Le Saux, B.; Lefevre, S. Semantic segmentation of earth observation data using multimodal and multi-scale deep networks, proceedings of lecture notes in computer science, Taipei, Taiwan; Springer Verlag: 2017; 180–196Google Scholar
- Chen Z, Dou A (2018) Road damage extraction from post-earthquake uav images assisted by vector data. Proceedings ofBeijing, China; International Society for Photogrammetry and Remote Sensing:211–216Google Scholar
- Chen S, Ma; Y. Fan H, Xu F, Lian J (2008) Road damage assessment from high resolution satellite RemoteSensing imagery in Wenchuan earthquake. J Remot Sens 12:949–955 In ChineseGoogle Scholar
- Coulibaly I, Lepage R, St-Jacques M (2015) Road damage detection from high resolution satellite images based on machine learning, proceedings of IEEE 8th international workshop on the analysis of multi-temporal remote sensing images, 2015. Annecy, FranceGoogle Scholar
- Fu C, Chen Y, Tong L, Jia M, Tan L, Ji X (2014) Road damage information extraction using high-resolution SAR imagery. Proceedings of IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Quebec City, QC, Canada; Institute of Electrical and Electronics Engineers Inc:1836–1838Google Scholar
- Gong L, An L, Liu M, Zhang J (2012) Road damage detection from high-resolution RS image. Proceedings of IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Munich, Germany; Institute of Electrical and Electronics Engineers Inc:990–993Google Scholar
- Haghighattalab A, Mohammadzadeh A, Valadan Zoej MJ, Taleai M (2010) Post-earthquake road damage assessment using region-based algorithms from high-resolution satellite images. Proceedings of SPIE - The International Society for Optical Engineering, Toulouse, France; SPIE:4993–4998Google Scholar
- He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, United states; IEEE Computer Society:770–778Google Scholar
- Kingma D, Ba JA (2014) A method for stochastic optimization. Computer ScienceGoogle Scholar
- Liu X, Li X, Li J, Wang Q (2013) Object-oriented remote sensing image classification and road damage adaptive extraction. Proceedings of International Conference on Remote Sensing, Environment and Transportation Engineering (RSETE), Nanjing, China; Atlantis Press:140–143Google Scholar
- Ma H, Lu N, Ge L, Li Q, You X, Li X (2013) Automatic road damage detection using high-resolution satellite images and road maps. Proceedings of IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Melbourne, VIC, Australia; Institute of Electrical and Electronics Engineers Inc:3718–3721Google Scholar
- Muni Kumar N, Manjula R (2019) Design of multi-layer perceptron for the diagnosis of diabetes mellitus using Keras in deep learning. Proceedings ofVijayawada, India; Springer Science and Business Media Deutschland GmbH:703–711Google Scholar
- OpenStreetMap, C. OpenStreetMap. Availale online: https://www.openstreetmap.org/#map=5/51.500/-0.100 (Jul 7, 2018)
- Paisitkriangkrai S, Sherrah J, Janney P, Van Den Hengel A (2016) Semantic labeling of aerial and satellite imagery. IEEE J-STARS 9:2868–2881Google Scholar
- Samadzadegan F, Zarrinpanjeh N (2008) Earthquake destruction assessment of urban roads network using satellite imagery and fuzzy inference systems, proceedings of international archives of the photogrammetry. Remote Sensing and Spatial Information Sciences (ISPRS), Beijing, China; International Society for Photogrammetry and Remote Sensing:409–414Google Scholar
- Sghaier MO, Lepage R (2015) Road damage detection from VHR remote sensing images based on multiscale texture analysis and dempster Shafer theory. Proceedings of IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Milan, Italy; Institute of Electrical and Electronics Engineers Inc:4224–4227Google Scholar
- Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In: Proceedings of international conference on learning Representations-2015, Hilton, San Diego, pp 1–14Google Scholar
- Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A (2015) Going deeper with convolutions. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, United states; IEEE Computer Society:1–9Google Scholar
- Wang Y, Wang Y, Da Y, Liu X, Li J, Huang J (2011) An object-oriented method for road damage detection from high resolution remote sensing images. Proceedings of International Conference on Geoinformatics, Shanghai, China; IEEE Computer SocietyGoogle Scholar
- Wang J, Qin Q, Zhao J, Ye X, Qin X, Yang X, Wang J, Zheng X, Sun Y (2015b) A knowledge-based method for road damage detection using high-resolution remote sensing image. Proceedings of IEEE Geoscience and Remote Sensing Symposium, Milan, Italy; Institute of Electrical and Electronics Engineers Inc:3564–3567Google Scholar
- Wen Q, Chen S, He H, Li S, Bai R, Zhao X (2012) Application of remote sensing system of unmanned aerial vehicle in Yingjiang, Yunnan earthquake. J Nat Disaster 21:65–71 In ChineseGoogle Scholar
- Ya-lan L, Yong Z, Yu-huan R, Zhen-yu LI, Lei-qiu HU, Fu-qing Z, Hai-xia HE, Hong-tao MA (2008) C. Jiang highway damage monitoring and assessment for Wenchuan quake based on remote sensing and its information integration. J Remot Sens 933-941(In Chinese):12Google Scholar