基于遥感云计算的自然资源动态监测研究

Research on dynamic monitoring of natural resources based on remote sensing cloud computing

关键词:  GEE  自然资源  机器学习  地物提取  动态监测

Key words:  GEE  natural resources  machine learning  extraction of surface features  dynamic monitoring

摘要:

目的 针对自然资源资产管理绩效评估中自然资源数据收集手段落后、要素不全、数据断档、更新频率不一致、缺乏空间信息等现状，构建了自动化、业务化自然资源遥感云计算动态监测服务平台。方法 文章以四川省理县为例，利用Landsat，MODIS和Sentinel等多源遥感数据，通过计算植被指数、水体指数和干旱指数等指标，综合运用机器学习方法识别与提取自然资源地物类别。结果 根据该文提出的高效计算方案，构建自然资源动态监测云平台，并基于多源数据信息的互补特性实现了复杂地物的高精度识别和提取，提升自然资源自动化、业务化动态变化监测能力。结论 该平台可为生态系统价值评估、县域自然资源资产管理以及生态环境质量监测工作提供思路和参考。

Abstract:

Purpose An automatic and operational natural resource remote sensing cloud computing dynamic monitoring service platform is constructed，in consideration of the status quo of backward natural resource data collection methods，incomplete elements，data interruption，inconsistent update frequency and lack of spatial information in natural resource asset management performance evaluation.Method Taking Lixian cunty of Sichuan Province as an example，this paper comprehensively conducts machine learning methods to identify and extract categories of natural resources surface features and calculates vegetation index，water index and drought index，taking advantage of multi-source remote sensing data such as Landsat，MODIS and Sentinel.Result According to the efficient computing scheme proposed in this paper，a cloud platform for dynamic monitoring of natural resources was built，the high-precision identification and extraction of complex surface features was realized basing on the complementary characteristics of multi-source data information and automatic and operational dynamic change monitoring ability of natural resources was improved.Conclusion This platform provides ideas and references for ecosystem value management and ecological environment quality monitoring.