摘要
面向国家生态文明建设和领导干部自然资源资产离任审计工作业务需求,研发了一套自然资源资产审计系统,该系统基于多元遥感数据,结合地理信息技术和深度学习理论与方法,实现自然资源资产审计工作的信息化管理和智能化分析。系统通过机器深度训练、学习和验证,能够快速地对需要进行审计的图斑进行识别,自动输出审计结果。在广西壮族自治区地理信息测绘院的运行效果表明,该系统可以辅助审计人员快速地进行审计工作,对开展领导干部自然资源资产离任审计工作有重要意义。
With the aims to serve the national ecological civilization construction and the requirements of auditing outgoing officials’ management of natural resource assets. A set of natural resources assets audit system is developed. The system, which was based on multiple remote sensing data, combines both the geographical information technology and deep learning theory in the information management of natural resources assets audit work and intelligent analysis. The system can automatically figure out the type of the patches in a very quick manner based on deep learning,model training and validation. Besides, the system can automatically provide the outputs of the auditing. The performance of this system in Guangxi Institute of Geographic Information Surveying and Mapping proves that it can help auditors in their daily work, and improve the efficiency of audit. It is also of great significance in auditing outgoing officials’ management of natural resource assets.
作者
黄铭
李威
汤琢
陈泽怡
孙炎
陈奕云
HUANG Ming;LI Wei;TANG Zhuo;CHEN Zeyi;SUN Yan;CHEN Yiyun(Guangxi Institute of Geographic Information Surveying and Mapping,Liuzhou 545006,China;School of Resources and Environmental Sciences,Wuhan University,Wuhan 430079,China)
出处
《测绘地理信息》
2020年第5期97-100,共4页
Journal of Geomatics
基金
国家重点研发计划(2018YFD1100801-01)。
关键词
自然资源资产审计
多元遥感数据
卷积神经网络
natural resource asset audit
multivariate remote sensing data
convolutional neural network
