遥感测绘手段在耕地保护规划及土地监管中的应用

Application of Remote Sensing Surveying and Mapping Methods in Farmland Protection Planning and Land Supervision

摘要: 传统耕地保护与土地监管依赖人工巡查与静态台账，存在覆盖范围有限、时效性差、漏检率高等短板，难以适配精细化管控需求。本文以多时相Sentinel-2影像的耕地“非粮化”智能监测为核心切入点，系统阐述数据预处理、时序特征提取、随机森林分类与精度验证的完整技术流程，梳理遥感技术在耕地规划与监管中的应用场景，分析当前技术应用的现实短板，提出具备可操作标准的优化路径，并以山东菏泽成武县为实证区域开展验证。结果表明，该技术方法对耕地非粮化的总体识别精度达89.2%，监测效率较人工巡查提升87%，可为县域耕地常态化管控提供可复制的技术范式。

Abstract: Traditional farmland protection and land supervision rely on manual inspections and static records, which have shortcomings such as limited coverage, poor timeliness, and high missed detection rates, making it difficult to adapt to the needs of refined control. This article takes the intelligent monitoring of farmland “non grainization” using multi temporal Sentinel-2 images as the core entry point, systematically elaborates on the complete technical process of data preprocessing, temporal feature extraction, random forest classification, and accuracy verification, sorts out the application scenarios of remote sensing technology in farmland planning and supervision, analyzes the current practical shortcomings of technology application, proposes an optimization path with operable standards, and conducts verification in Chengwu County, Heze, Shandong Province as an empirical area. The results showed that the overall recognition accuracy of this technology method for non grain cultivated land reached 89.2%, and the monitoring efficiency was improved by 87% compared to manual inspection. It can provide a replicable technical paradigm for the normalized control of cultivated land in counties.