基于Google Earth Engine的海量舰船目标SAR图像处理应用研究

传统基于单机的合成孔径雷达(Synthetic Aperture Radar,SAR)图像舰船目标检测需要在本地计算机上进行数据下载、处理和分析,这极大受限于本地计算机的性能,只能对少量SAR图像进行检测。本文利用Google Earth Engine(GEE)遥感云计算平台的海量数据存储和强大运算能力,通过在云端部署SAR卫星数据、模型算法和计算机算力,在GEE平台上进行了大范围海域的海量舰船目标SAR图像处理应用研究,实现了舰船目标检测同时还可以获取舰船目标信息、统计舰船目标数量、批量下载目标检测结果图像等。通过在Sentinel-1 SAR数据上进行相关实验,结果表明本文研究可在线对海量SAR数据进行实时、高效、快速地处理,对海上舰船监视具有较高的实际应用价值。 

Research on the Application of SAR Image Processing of Massive Ship Targets Based on Google Earth Engine

The traditional ship target detection in Synthetic Aperture Radar image based on stand-alone computer requires data download, processing and analysis on the local computer, which is greatly limited by the performance of the local computer and can only be detected in a small number of SAR images. This paper makes use of the Google Earth Engine remote sensing cloud computing platform of massive data storage and powerful computing power, through the deployment of SAR satellite data, model algorithms and computer computing power in the cloud, carried out the SAR image processing application research of massive ship targets in a large range of sea areas on the GEE platform. It realizes ship target detection and can also obtain ship target information, count the number of ship targets, and download target detection result images in batches. Through related experiments on Sentinel-1 SAR data, the results show that the research in this paper can process massive SAR data online in real time, efficiently and quickly, and has high practical application value for maritime ship surveillance.