The content of this article is from the "Surveying and Mapping Bulletin" No. 2, 2024, drawing review number: GS Jing (2024) No. 0287
A spatialization construction method of remote sensing image housing data for rapid disaster assessment after earthquake
ZHANG Ping1,2, LI Bijun3, LI Yin1,2, ZHANG Yimei1,2, TEMUQILE1,2, LIU Ke1,2, LI Zhijun4
1. Key Laboratory of Earthquake Geodesy, Institute of Seismology, China Earthquake Administration, Wuhan, Hubei 430071, China;2. Hubei Provincial Earthquake Administration, Wuhan, Hubei 430071;3. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, Hubei 430079;4. Information Center of the Ministry of Natural Resources, Beijing 100812
Funds: China Earthquake Administration Earthquake Emergency Youth Key Mission (CEAEDEM202213); Fundamental Research Funds of the Institute of Seismology, China Earthquake Administration, and Fundamental Research Funds of the Institute of Crustal Stress, China Earthquake Administration(306337-12); Basic Research Foundation of Hubei Earthquake Administration(2022HBJJ012)
Keywords: remote sensing imagery, convolutional neural network, housing data, spatialization, grid sampling
Citation format:ZHANG Ping, LI Bijun, LI Yin, et al. Spatialization Construction Method of Housing Data in Remote Sensing Images for Post-earthquake Rapid Disaster Assessment[J]. Bulletin of Surveying and Mapping, 2024(2): 69-73.doi: 10.13474/j.cnki.11-2246.2024.0212
Abstract:The convolutional neural network method can efficiently extract the house vector data on high-resolution remote sensing images, quickly obtain the spatial distribution data of the house data, and improve the update ability of the basic database of earthquake emergency. Based on the Contour Guidance and Structure Aware Codec Fully Convolutional Neural Network (CGSANet) model and the sub-regional isoscale grid sampling method, this paper obtains the spatial distribution model of building area and building structure type, and has the technical ability of spatialization of multi-type housing data in the context of complex regions. Taking Huangmei County as the research object, a spatialized dataset of 1 km × 1 km housing data was constructed, and the identification and identification of housing data of different structural types were realized. The constructed spatialized dataset of housing data can provide a data source for the basic database of earthquake emergency, which is of great significance for improving the accuracy and timeliness of housing data.
About author:ZHANG Ping (1993—), female, master, engineer, mainly engaged in remote sensing and GIS technology application research. E-mail:[email protected] Corresponding author: Li Bijun. E-mail:[email protected]
First instance: Yang Ruifang review: Song Qifan
Final Judge: Jin Jun
information