EPDCL/Electrical-Lines-Defect-Detection

电气线路缺陷检测数据集，由APEPDCL公司的线路工人使用移动相机捕获，并由Sampath Balaji及其团队整理。该数据集旨在促进开源合作、可重现性和AI领域的实践学习。数据集包括YOLOv12格式的目标检测数据集和文件夹+CSV格式的分类数据集，均有训练集、验证集和测试集的划分。数据集遵循CC BY 4.0许可证，代码遵循MIT许可证。Electrical Lines Defect Detection dataset, captured by APEPDCL linemen using mobile cameras and curated by Sampath Balaji and his team. The dataset aims to foster open-source collaboration, reproducibility, and hands-on learning in AI. It includes YOLOv12 formatted object detection datasets and folder+CSV formatted classification datasets, each with train, validation, and test splits. The dataset is licensed under CC BY 4.0, and the code is under the MIT License.

搜集汇总

背景与挑战

背景概述

该数据集由EPDCL提供，专注于电力线路缺陷检测，包含图像和文本模态，数据规模为22,025行，总文件大小13.2 GB，并分为训练、验证和测试子集。它适用于计算机视觉和自然语言处理任务，用于缺陷识别和分析。