遥感技术的发展现状及在天府粮仓建设中的应用建议

Progress of remote sensing and application recommendations in the construction of the Tianfu Granary

关键词:  遥感  耕地  农作物  产量  深度学习

Key words:  remote sensing  cropland  crops  deep learning

摘要:

【目的】 促进遥感技术在天府粮仓建设中的应用，助力农业生产高质、高效、可持续发展。【方法】 （1）文章通过文献综述，将遥感技术在农业监测中的研究分为耕地监测、农作物遥感制图、作物长势监测与产量估算。（2）在四川地区复杂应用场景下，具体分析现有遥感技术在区域监测面临的挑战。（3）提出遥感技术助力天府粮仓建设的相关建议。【结果】 遥感技术经过不断发展，建立了成熟的空天地一体化农作物遥感监测技术体系及系列数据产品，在种植结构优化、政策制定、资源有效利用等方面发挥重要作用。然而，在四川地区开展遥感监测仍面临复杂地形、地块破碎、种植制度等挑战。【结论】 为了进一步加强遥感技术在天府粮仓建设中的应用，助力农业高质量发展，提出以下建议：一是加强大模型应用，支持高标准农田建设；二是使用多源遥感数据和深度学习方法，提高景观复杂场景下的遥感制图精度；三是耦合多源遥感数据和各作物农学机理，提高作物长势监测和产量估算能力。

Abstract:

【Purpose】 In order to facilitate the application of remote sensing in the construction of the Tianfu Granary and assist in the high-quality，efficient，and sustainable development of agriculture production.【Methods】 （1）Based on literature review，the research of remote sensing technology in agricultural monitoring mainly included the cropland extraction，mapping crop area，growth monitoring and yield estimation. （2）The challenges faced by existing remote sensing technology in monitoring cropland and crops in the construction of Tianfu Granary were analyzed specifically for the complex application scenarios in the Sichuan region. （3）Several suggestions were put forward to promote the application of remote sensing technology in the construction of the Tianfu Granary.【Result】 After a long period of research，some air-heaven-earth integrated crop monitoring system had been established，with a greater degree of accuracy and intelligence，and a large number of data products had been generated，which could provide technological and data support for the government in optimizing planting structures，policy formulation，and the effective utilization of resources. However，there were still some challenges in different regions and scenarios，such as complex terrain，fragmented cropland，cropping systems.【Conclusion】 Some suggestions are proposed to promote the application of remote sensing. Firstly，strengthening the application of large-scale models to support the construction of high-standard farmland. Secondly，utilizing multisource remote sensing data and deep learning methods to improve the accuracy of remote sensing mapping in complex landscapes. Finally，integrating multisource remote sensing data with the agronomic mechanisms of various crops to enhance the capability of crop growth monitoring and yield estimation.