MagicBathyNet

MagicBathyNet是一个多模态遥感数据集，由柏林学习和数据基础研究所创建，旨在支持浅水区域的测深预测和基于像素的分类。该数据集包含3355个由Sentinel-2、SPOT-6和航空影像组成的图像块，以及相应的测深数据和海底类别标注。数据集覆盖了塞浦路斯的Agia Napa和波兰的Puck Lagoon两个不同海岸区域，用于评估深度学习方法在测深和海底分类任务中的性能。MagicBathyNet的创建过程涉及从多种来源收集和处理图像数据，以及手动标注海底类别，以确保数据的准确性和可靠性。该数据集主要应用于海洋科学和环境监测领域，旨在解决浅水区域测绘不足的问题，支持生态保护和灾害预防。

MagicBathyNet is a multimodal remote sensing dataset developed by the Berlin Institute for Learning and Data Foundations, aimed at supporting bathymetry prediction and pixel-based classification in shallow water areas. This dataset contains 3355 image patches composed of Sentinel-2, SPOT-6 and aerial imagery, along with corresponding bathymetric data and seafloor category annotations. It covers two distinct coastal regions: Agia Napa in Cyprus and Puck Lagoon in Poland, and is used to evaluate the performance of deep learning methods on bathymetry and seafloor classification tasks. The construction of MagicBathyNet involves collecting and processing image data from multiple sources, as well as manual annotation of seafloor categories to ensure the accuracy and reliability of the dataset. This dataset is primarily applied in the fields of marine science and environmental monitoring, aiming to address the issue of insufficient shallow water area mapping and support ecological conservation and disaster prevention.