摘要
Abstract
Satellite remote sensing technology is now widely used to monitor the dust storm process in space and time.Fengyun-4 Meteorological Satellite(FY-4A)is a new generation of geostationary remote sensing meteorological satellites in China,and its Multi-channel Advanced Geosynchronous Radiation Imager(AGRI)plays an active role in dust identification in Asia.Several dust recognition methods based on satellite data,including dust recognition method based on RGB images,BTD(Brightness Temperature Difference),NDDI(Normalized Difference Dust Index)and Machine Learning-based dust retrieval methods,are applied to the L1 data of the FY-4A satellite AGRI to realize the identification of dust.Through individual case analysis,the experimental results are further studied and compared.The results show that most of the dust identification methods applied to the FY-4A satellite can distinguish the surface,clouds and dust,and then identify the dust.For the identification method based on physical characteristics,due to the difference in the bands of different satellites,the threshold universality is poor,and there are cases of small dust identification and misjudgment of dust in some areas.Based on the Machine Learning method,it can effectively identify the dust range,which has strong applicability and broad application prospects.Finally,the application of satellite-based dust storm identification methods is summarized,and further prospects for dust identification are given.关键词
FY-4A卫星/沙尘暴/RGB图像/BTD/NDDI/机器学习Key words
FY-4A satellite/dust storm/RGB image/BTD/NDDI/Machine Learning分类
信息技术与安全科学
