Ev-Satellites
收藏资源简介:
Ev-Satellites数据集由西悉尼大学创建,专门用于评估神经形态卫星观测中的噪声过滤算法。该数据集包含高分辨率的卫星数据,具有高质量的地面实况,适用于各种噪声条件下的评估。数据集内容包括稀疏的卫星事件流,以及周围的热像素和背景噪声。创建过程中,通过使用CMax框架对事件进行对齐和补偿,以生成点源图像。该数据集主要应用于卫星跟踪和空间态势感知,旨在提高事件相机在复杂环境中的信号检测和跟踪能力。
The Ev-Satellites dataset was developed by Western Sydney University, specifically designed for evaluating noise filtering algorithms in neuromorphic satellite observations. This dataset features high-resolution satellite data paired with high-quality ground truth, making it suitable for evaluations under various noise conditions. The dataset includes sparse satellite event streams, along with surrounding thermal pixels and background noise. During its development, the CMax framework was employed to align and compensate for events, thus generating point-source images. This dataset is primarily applied in satellite tracking and spatial situational awareness, with the objective of enhancing the signal detection and tracking capabilities of event cameras in complex environments.
Noise Filtering Benchmark for Neuromorphic Satellites Observations
数据集概述
- 数据集名称: Noise Filtering Benchmark for Neuromorphic Satellites Observations
- 数据集用途: 用于神经形态卫星观测的噪声过滤基准测试。
环境要求
- Python版本: 3.9.x, 3.10.x
测试环境
- 操作系统: Ubuntu 22.04
- Conda版本: 23.1.0
- Python版本: 3.9.18
安装步骤
-
创建并激活Conda环境: sh conda create --name dvs_sparse_filter python=3.9 conda activate dvs_sparse_filter
-
安装依赖包: sh python3 -m pip install -e . pip install torch tqdm plotly scikit-image loris PyYAML opencv-python scikit-learn hdbscan astroquery pillow python3 -m pip install astropy requests astrometry scikit-image matplotlib-label-lines ipywidgets conda install -c conda-forge pydensecrf
运行说明
- 状态: 正在进行中...

- 1Noise Filtering Benchmark for Neuromorphic Satellites Observations西悉尼大学 · 2024年



