摘要
由于农村垃圾投放处理设施落实不完善,居民环保意识淡薄,导致部分地区垃圾投放较为随意,对农村环境整洁度产生了较大影响。本文针对农村生活垃圾分布零散、隐蔽性较强等特点,以遥感影像数据为基础,采用遥感目标检测算法RSADet和深度学习相结合的方式进行垃圾目标识别提取,垃圾目标提取准确率在80%以上;同时对算法的缺陷进行了分析研究,并提出了改进方法及后续研究方向,以准确可靠的数字化、信息化技术方法为农村生活环境治理基础数据采集提供了高效解决方案。
Due to the imperfect implementation of rural garbage disposal facilities and the weak awareness of environmental protection of residents,garbage disposal in some areas is more random,which has a great impact on the cleanliness of rural environment.Focusing on the scattered distribution and strong concealment of rural household garbage,based on remote sensing image data,this paper uses remote sensing target detection algorithm RSADet and deep learning to identify and extract garbage targets.The accuracy of garbage target extraction is more than 80%.At the same time,the defects of the algorithm are analyzed and studied,the improvement methods and follow-up research directions are proposed,and the accurate and reliable digital and informatization technology methods provide efficient solutions for the basic data collection of rural living environment governance.
作者
杨磊
YANG Lei(Lunan Geological Engineering Survey Institute of Shandong Province(The Second Geological Brigade of Shandong Bureau of Geology and Mineral Resources Exploration and Development),Jining 272100,China)
出处
《测绘与空间地理信息》
2025年第10期31-34,共4页
Geomatics & Spatial Information Technology
基金
山东省鲁南地质工程勘察院(山东省地质矿产勘查开发局第二地质大队)开放基金项目(LNY202205)资助。
关键词
DOM
深度学习
智能算法
疑似垃圾目标
DOM
deep learning
intelligent algorithm
suspected garbage targets
