How to accurately extract large-scale urban land?Establishment of an improved fully convolutional neural network model
作者机构:Chongqing Smart City InstituteChongqing Jiaotong UniversityChongqing 400074China Chongqing Geomatics and Remote Sensing CenterChongqing 401147China
出 版 物:《Frontiers of Earth Science》 (地球科学前沿(英文版))
年 卷 期:2022年第16卷第4期
页 面:1061-1076页
核心收录:
学科分类:08[工学] 081104[工学-模式识别与智能系统] 0708[理学-地球物理学] 0811[工学-控制科学与工程]
基 金:supported by Natural Science Foundation of Chongqing in China(No.cstc2020jcyj-jqX0004) the Ministry of education of Humanities and Social Science project(No.20YJA790016) the National Natural Science Foundation of China(Grant No.42171298) We thank the patent supporting the method section of the paper(No.202110750360.1).
主 题:improved fully convolutional neural network remote sensing image classification city boundary precision evaluation
摘 要:Realizing accurate perception of urban boundary changes is conducive to the formulation of regional development planning and researches of urban sustainable development.In this paper,an improved fully convolution neural network was provided for perceiving large-scale urban change,by modifying network structure and updating network strategy to extract richer feature information,and to meet the requirement of urban construction land extraction under the background of large-scale low-resolution image.This paper takes the Yangtze River Economic Belt of China as an empirical object to verify the practicability of the network,the results show the extraction results of the improved fully convolutional neural network model reached a precision of kappa coefficient of 0.88,which is better than traditional fully convolutional neural networks,it performs well in the construction land extraction at the scale of small and medium-sized cities.