Geolocating and Measuring Offshore Infrastructure with Multimodal Satellite Data

Please cite this when using the dataset and code.

The offshore infrastructure is rapidly spreading in the Scottish waters to satisfy the increasing energy demand. The up-to-date knowledge of their distribution and size is critical for the development and management of marine ecosystems. With the development of remote sensing techniques, satellite data have been widely used in offshore infrastructure detection on the vast ocean. However, the automatic and accurate identification on remote sensing data is still challenging that every kind of data have limitations. Here we combine the Sentinel-1 SAR data and Sentinel-2 Multi-Spectral Instrument (MSI) imagery to propose an automatic method for the location detection and size evaluation of offshore infrastructure in Scottish waters. Specifically, three strategies (transformed median composite, 2D-SSA filtering and threshold segmentation) were designed to first extract the contour range on Sentinel-1 data. Then morphological operations were applied on Sentinel-2 true color image to obtain the precise location and size of each offshore infrastructure.

All the Sentinel-1 and Sentinel-2 data are downloaded from https://www.sentinel-hub.com/explore/eobrowser/

The file "loc_S1" is used for the contour range detection (guided area) in Sentinel-1;

The file "loc_S2" is for the specific location detection and size evaluation of oil/gas platforms and wind turbines;

使用本数据集及代码时，请引用本文。

为满足日益增长的能源需求，苏格兰海域的近海基础设施正快速扩张。精准掌握其分布与规模，对海洋生态系统的开发与管理至关重要。随着遥感技术的发展，卫星数据已被广泛应用于广袤海域的近海基础设施探测。然而，由于各类数据均存在局限性，对遥感数据进行自动且精准的识别仍颇具挑战。本研究结合哨兵-1号（Sentinel-1）合成孔径雷达（Synthetic Aperture Radar, SAR）数据与哨兵-2号（Sentinel-2）多光谱成像仪（MSI）影像，提出了一种适用于苏格兰海域近海基础设施的位置探测与规模评估自动化方法。具体而言，研究首先设计了三种策略——变换中值合成法、二维奇异谱分析（2D-SSA）滤波与阈值分割，从哨兵-1号数据中提取轮廓范围；随后对哨兵-2号真彩色影像开展形态学操作，以获取每一处近海基础设施的精确位置与规模信息。

本研究所用的所有哨兵-1号及哨兵-2号数据均下载自https://www.sentinel-hub.com/explore/eobrowser/

文件"loc_S1"用于哨兵-1号数据的轮廓范围（引导区域）探测；
文件"loc_S2"则用于油气平台与风力发电机组的精确位置探测及规模评估。