Tunnel cable fire data set

The Tunnel Cable Fire dataset is derived experimentally, this dataset contains images of cable flames at different stages, different cable layers, and different wind speeds, with a special focus on computer vision tasks such as fire detection and segmentation. These images have been enhanced with mosaic data for a total of 1812 datasets, including single and double layer cable fire images in the case of no wind and wind speed of 2.7m/s. There are 300 images of single layer cables in the case of no wind, 280 images of single layer cable fires in the case of wind speed of 2.7m/s, and 1232 images of double layer cables fires in the case of 2.7m/s wind speed. Each case contains images before, after and during burning. The dataset is divided into a training set (70%) and a validation set (30%) to provide a standardised benchmark for evaluating tunnel cable fire detection algorithms. The dataset contributes to the development of fire detection research on tunnel cables by 1. providing images of cable fires at different stages, cable layers, and wind speeds 2. providing comprehensive pixel-level annotations validated by multi-expert cross-validation 3. supporting the development of lightweight, real-time detection systems deploying small devices.