基于深度学习的城市遥感图像处理系统设计与实现  

Design and implementation of urban remote sensing image processing system based on deep learning

作　　者：胡欣雨 娄立都 周晓 Hu Xinyu;Lou Lidu;Zhou Xiao(School of Computer Science,University of South China,Hengyang 421000,China)

机构地区：[1]南华大学计算机学院,衡阳421000

出　　处：《现代计算机》2023年第22期25-29,共5页Modern Computer

基　　金：2023年湖南省大学生创新创业训练计划项目(D202305151441213756)。

摘　　要：城市绿化对于城市生态至关重要,因此如何监管和保护绿化资源成为环境保护、城市绿化规划等领域的重要课题。随着深度学习引入遥感技术,自动化绿化监管成为一个主流方向。基于语义分割网络模型开发了一种绿化信息提取系统,可以自动标注城市遥感图像中的绿地,并进行城市绿化面积的计算和相关数据的存储。同时使用SRGAN实现低分辨率遥感图像的4倍分辨率提升并增加纹理和细节。该系统可帮助政府或企业分析绿化分布情况,同时经济便捷地获取高分辨遥感图像。Urban greening is very important to urban ecology,so how to supervise and protect greening resources has become an important issue in the fields of environmental protection and urban greening planning.With the introduction of remote sensing technology into deep learning,automated greening supervision has become a mainstream direction.Based on the semantic segmentation network model,this study developed a greening information extraction system,which can automatically label the green space in urban remote sensing images,and calculate the urban greening area and store related data.At the same time,SRGAN is used to improve the resolution of low‑resolution remote sensing images by 4 times and increase texture and detail.The system can help the government or enterprises to analyze the distribution of greening,and obtain high‑resolution remote sensing images economically and conveniently.

关 键 词：城市绿化 城市遥感图像 语义分割 超分辨率 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置] TP18[自动化与计算机技术—控制科学与工程]