Detection On Tracks Human Behavior Detection Dataset on Tracks

This dataset is designed for computer vision models that detect people on railway tracks. The main goal is to improve railway safety by providing high-quality annotated images that can be used to train models to detect unauthorized human intervention in real-time, thereby preventing accidents and fatalities.

The dataset contains 3,766 images of humans on railway tracks at a resolution of 1,080×1,080. Each image is annotated with bounding boxes marking the presence of humans and their behavior on the railway tracks. The annotations are provided in YOLO format (TXT) with detailed class labels. There are 4 categories for these images:

• Tracks: Human activity on railway tracks should be detected.

• Standing: One of the human behaviors on railway tracks.

• Sitting: Human sitting on railway tracks.

• Sleeping: Humans sleeping or lying down in orbit.

Datasetuse 

The dataset is well suited for training and evaluating object detection models, especially those based on Convolutional Neural Networks (CNN) and YOLO architectures. It is also suitable for research in railway safety, object detection, and real-time monitoring systems.