耕地遥感识别研究进展与展望

快速、准确获取耕地数量及其分布信息是研究耕地时空格局和生态效应的基础，也是及时制定应对粮食问题对策的迫切需求。近年来，随着卫星遥感技术的迅猛发展，遥感以其宏观性、实时性以及经济性为耕地信息快速获取提供了可能性。本文归纳了遥感技术应用于识别耕地信息的研究进展，总结了国内外耕地信息提取研究中常用的数据源、分类算法、时相选择、分类对象，讨论了上述四大类在提取耕地信息过程中的优缺点。随着传感器数量不断增加，遥感影像时间分辨率、空间分辨率及光谱分辨率不断提高，分类算法的不断涌现，基于多源遥感数据，集成智能分类算法识别耕地将成为必然的发展趋势。

Progress and prospect of cultivated land extraction research using remote sensing

Rapidly and accurately extracted cultivated land quantity and distribution information is the basis for studying the spatiotemporal patterns and ecological effects of cultivated land, and there is an urgent need for timely formulation of corresponding countermeasures for food problems. In recent years, the advancement of remote sensing technology has provided the possibility for the rapid acquisition of cultivated land information owing to the macroscopic, real-time, and economical properties of the technology. This paper summarized the research progress of the application of remote sensing technology in the extraction of cultivated land, commonly used data sources, classification algorithms, and temporal and classification objects in the research of domestic and foreign cultivated land extraction methods, and discussed the advantages and disadvantages of the above four categories in the process of extracting cultivated land information. With the increasing number of sensors, the temporal resolution, spatial resolution, and spectral resolution of remote sensing images are continuously improving, and the classification algorithm is constantly updated. Based on multi-source remote sensing data, the integration of intelligent classification algorithms for identifying cultivated land will become an inevitable development trend.