Hyperspectral Oil Spill Database (HOSD)
收藏资源简介:
Hyperspectral Oil Spill Database (HOSD)是由湖南大学电气与信息工程学院的研究团队创建的公开油污检测数据集,包含18张2010年在墨西哥湾拍摄的高光谱图像。该数据集覆盖广泛,具有大范围的覆盖面积,是首个用于油污检测的公开数据集。HOSD数据集通过AVIRIS传感器获取,提供了从可见光到红外光谱的丰富光谱信息,有助于在复杂的海洋环境中监测油污。数据集中的图像经过大气校正模型处理,每张图像的参考图均由领域专家手动标注。HOSD数据集的应用领域主要集中在油污检测,旨在通过高光谱图像分析技术,提高油污检测的准确性和效率,为海洋环境保护和资源管理提供支持。
Hyperspectral Oil Spill Database (HOSD) is a public oil spill detection dataset developed by a research team from the College of Electrical and Information Engineering, Hunan University. It consists of 18 hyperspectral images captured in the Gulf of Mexico in 2010. Boasting wide coverage and a large observation scope, it is the first public dataset tailored for oil spill detection. The HOSD dataset was acquired using the AVIRIS sensor, which provides abundant spectral information ranging from the visible to infrared bands, facilitating oil spill monitoring in complex marine environments. All images in the dataset have been processed with an atmospheric correction model, and the reference ground truth for each image was manually annotated by domain experts. The primary application field of the HOSD dataset is oil spill detection. It aims to enhance the accuracy and efficiency of oil spill detection via hyperspectral image analysis technologies, thereby providing support for marine environmental protection and resource management.

- 1Hyperspectral Remote Sensing Benchmark Database for Oil Spill Detection with an Isolation Forest-Guided Unsupervised Detector湖南大学电气与信息工程学院 · 2022年



