DOORS: Dataset fOr bOuldeRs Segmentation
收藏资源简介:
DOORS数据集由米兰理工大学的Mattia Pugliatti和Francesco Topputo创建,旨在为小天体表面巨石的检测提供支持,特别是在视觉导航和危险检测等应用中。数据集包含两个子集DS1和DS2,分别包含30181和20095个样本,这些样本是通过Blender软件生成的合成图像。创建过程涉及复杂的渲染和后处理步骤,确保数据集在光照条件和巨石形状上的多样性。该数据集主要用于开发和测试基于数据驱动的图像处理算法,以解决小天体导航中的巨石识别和分割问题。
The DOORS dataset was created by Mattia Pugliatti and Francesco Topputo from Politecnico di Milano, aiming to support boulder detection on the surfaces of small celestial bodies, particularly for applications including visual navigation and hazard detection. The dataset comprises two subsets, DS1 and DS2, which hold 30181 and 20095 samples respectively. All samples are synthetic images generated via Blender software. The dataset development involves complex rendering and post-processing steps to ensure diversity in lighting conditions and boulder shapes. This dataset is primarily used for developing and testing data-driven image processing algorithms to address boulder recognition and segmentation problems in small celestial body navigation.

- 1DOORS: Dataset fOr bOuldeRs Segmentation. Statistical properties and Blender setup米兰理工大学 · 2022年



