MassMIND: Massachusetts Maritime INfrared Dataset
收藏资源简介:
MassMIND数据集是由麻省理工学院海洋自主传感系统实验室创建,包含2916张长波红外分割图像,这些图像捕捉了多种海洋环境下的场景,如复杂的海洋环境、建筑、生物体及近岸景观。数据集通过实例分割和分类,将图像分为七个类别:天空、水、障碍物、生物障碍物、桥梁、自身和背景。该数据集旨在解决海洋自主导航中的场景理解问题,特别是在极端光照条件下的应用。
The MassMIND dataset was created by the Marine Autonomous Sensing Systems Laboratory at the Massachusetts Institute of Technology (MIT). It contains 2916 long-wave infrared (LWIR) segmented images that capture scenes across diverse marine environments, including complex marine settings, man-made structures, living organisms, and nearshore landscapes. Using instance segmentation and classification, the dataset categorizes images into seven classes: sky, water, obstacles, biological obstacles, bridges, ego vehicle, and background. This dataset is designed to address scene understanding challenges in marine autonomous navigation, particularly for applications under extreme lighting conditions.




