轨道扣件状态检测图像数据集

轨道扣件状态检测图像数据集以文件夹结构和标注文本文件共同组成。样本数据包括原始图像文件（.jpg）和对应的YOLO格式标注文件（.txt）。每幅图像样本用唯一文件名标识，主关键字段为图像文件名。标注文件按图像一一对应，字段包括：目标类别（0=正常，1=丢失，2=松动），目标中心的归一化x、y坐标，归一化宽w、高h。数据集文件结构分为图像（images）和标签（labels）文件，图像和标签文件夹又包括训练集（train）、验证集（val）、测试集（test）三个子文件夹，分别包含图像及相应标注文件。

The image dataset for rail fastener status detection consists of folder structures and annotated text files. The sample data includes original image files (.jpg) and corresponding YOLO-format annotation files (.txt). Each image sample is identified by a unique filename, with the image filename serving as the primary key field. The annotation files are in one-to-one correspondence with the images, and their fields include: target category (0 = normal, 1 = missing, 2 = loose), normalized x and y coordinates of the target center, as well as normalized width w and height h. The dataset's file structure is divided into image (images) and label (labels) folders, both of which contain three subfolders for the training set (train), validation set (val), and test set, respectively housing the images and their matching annotation files.

搜集汇总

背景与挑战

背景概述

该数据集是一个专门用于轨道扣件状态检测的图像集合，包含原始图像和YOLO格式的标注文件，标注类别涵盖正常、丢失和松动三种状态。数据集结构清晰，分为训练集、验证集和测试集，支持机器学习模型的训练和评估，适用于铁路维护和安全检测等应用场景。