ev-CIVIL
收藏资源简介:
ev-CIVIL数据集是首个基于事件相机捕捉的民用基础设施缺陷检测数据集,由丹麦技术大学的研究团队创建。该数据集包含两种类型的缺陷:裂纹和掉皮,数据来源于现场和实验室环境。现场数据集包含318个记录序列,记录了458个不同的裂纹和121个不同的掉皮实例;实验室数据集包含362个记录序列,覆盖了220个裂纹和308个掉皮实例。数据集同时包含了事件数据以及同时捕获的灰度强度图像帧,适用于民用结构缺陷检测的基准测试。
The ev-CIVIL dataset is the first civil infrastructure defect detection dataset captured by event cameras, created by the research team from the Technical University of Denmark. This dataset covers two types of defects: cracks and spalling, with data collected from both field and laboratory environments. The field dataset comprises 318 recording sequences, documenting 458 distinct crack instances and 121 distinct spalling instances; the laboratory dataset includes 362 recording sequences, covering 220 crack instances and 308 spalling instances. Additionally, the dataset contains both event data and simultaneously captured grayscale intensity image frames, making it suitable for benchmarking civil structural defect detection tasks.

- 1Event-based Civil Infrastructure Visual Defect Detection: ev-CIVIL Dataset and Benchmark丹麦技术大学 · 2025年



