UAV-Rain1k
收藏资源简介:
UAV-Rain1k是由沈阳航空航天大学等机构创建的无人机航拍图像去雨滴数据集,旨在解决雨滴对无人机航拍图像质量的影响问题。该数据集包含1020张图像,通过Blender模拟雨滴形状,结合多种无人机拍摄角度收集背景图像,采用随机采样方法生成雨滴掩码。数据集不仅用于评估现有图像去雨算法,还揭示了未来研究方向,广泛应用于无人机图像处理领域,特别是在监控、测绘和环境监测等任务中。
UAV-Rain1k is a deraining dataset for unmanned aerial vehicle (UAV) aerial imagery, developed by institutions including Shenyang Aerospace University and other partners, to address the degradation of UAV aerial image quality caused by raindrops. This dataset comprises 1,020 images, which are generated by simulating raindrop shapes via Blender, collecting background images across diverse UAV shooting angles, and producing raindrop masks through random sampling. The dataset can not only serve as a benchmark for evaluating existing image deraining algorithms, but also provide insights for future research directions, and has been widely adopted in the field of UAV image processing, particularly for applications such as surveillance, mapping, and environmental monitoring.

- 1UAV-Rain1k: A Benchmark for Raindrop Removal from UAV Aerial Imagery沈阳航空航天大学 · 2024年



