北京测绘2025，Vol.39Issue(1)：26-32,7.DOI:10.19580/j.cnki.1007-3000.2025.01.005
深度学习驱动的占用耕地建房监管平台设计
Design of supervision platform for house building via occupying farmland driven by deep learning
摘要
Abstract
In view of the disadvantages of the traditional supervision method for house building via occupying farmland,such as low efficiency,poor timeliness and accuracy,and insufficient initiative,this paper used a convolutional neural network(CNN)-based deep learning network model to interpret and process multi-temporal remote sensing images by analyzing the characteristics of house building via occupying farmland.The paper automatically identified and extracted suspected illegal patches of house building via occupying farmland and utilized Egg.js open-source framework,PostgreSQL database,geographic information system(GIS),and other supervision platforms for house building via occupying farmland.The paper employed the information management subsystem for house building via occupying farmland on a personal computer(PC)to accurately extract suspected illegal patches of house building via occupying farmland and used the application(App)for verification and proof of house building via occupying farmland to facilitate on-site information verification,effectively improving the supervision efficiency of illegal acts of house building via occupying farmland,promptly stopping illegal acts,and ensuring rectification and restoration.The supervision platform for house building via occupying farmland driven by deep learning has achieved good application results since it was put into use,and the extraction efficiency and accuracy of suspected illegal patches of house building via occupying farmland have been greatly improved.It can accurately identify and obtain information about house building via occupying farmland,providing a convenient and reliable auxiliary tool for the protection and supervision of farmland resources.关键词
深度学习/占用耕地建房/疑似违法图斑/监管平台/数据库Key words
deep learning/house building via occupying farmland/suspected illegal patch/supervision platform/database分类
天文与地球科学引用本文复制引用
洪圳材,蒋昌晶,赵洋峥..深度学习驱动的占用耕地建房监管平台设计[J].北京测绘,2025,39(1):26-32,7.基金项目
广东省科技计划(2018B020207002) （2018B020207002）
