遥感耕地监测现状与方法综述

Review on the Status and Methods of Remote Sensing Farmland Monitoring

关键词:  遥感  耕地监测  监测内容  分类体系  方法历程  主流方法  应用领域

Key words:  remote sensing  farmland monitoring  monitoring content  classification system  method history  mainstream methods  application area

摘要:

耕地资源是我国农村发展和农业现代化的根基命脉，是国家粮食安全的基石，因此迫切需要获取精细化、精准化的耕地资源分布与变化信息。对地观测技术与人工智能领域的发展为快速、精准、智能化的耕地监测提供了有力支撑。本文通过搜集国内外耕地监测领域相关的研究，梳理耕地相关监测内容，归纳遥感耕地监测分类体系，回顾遥感耕地监测领域技术方法的演化历程，总结当前主流的耕地监测方法，揭示不同农业区域、尺度区域与分类单元的应用领域现状，并指出多源遥感数据耕地监测面临的挑战及可能的发展趋势：遥感耕地监测应面向不同区域、尺度、粒度开展耕地信息提取与变化检测研究，发展与创新多源/模态数据协同监测技术，实现耕地监测的“数据-模型-知识”联合驱动，逐步提升耕地监测精准化、精细化与智能化水平。

Abstract:

Farmland resources are the lifeblood of my country's rural development and agricultural modernization, and the cornerstone of national food security.Therefore, there is an urgent need for refined and accurate information on the distribution and change of farmland resources.The development of earth observation technology and artificial intelligence provides strong support for fast, accurate and intelligent farmland monitoring.In this article, by collecting the relevant research in the field of farmland monitoring at home and abroad, the relevant content of farmland monitoring is sorted out, the classification system of remote sensing farmland monitoring is summarized, the evolution process of technical methods in the field of remote sensing farmland monitoring is reviewed, the current mainstream farmland monitoring methods is summarized, and the application status of different agricultural regions, scale regions and classification units is revealed.The challenges and possible development trends of farmland monitoring with multi-source remote sensing data are pointed out.Remote sensing farmland monitoring should carry out research on farmland information extraction and change detection for different regions, scales and granularity, develop and innovate multi-source/modal data collaborative monitoring technology, realize the joint drive of "Data-Model-Knowledge" of farmland monitoring, and gradually improve the precision, refinement and intelligence level of farmland monitoring.