Automated urban flood level detection based on flooded bus dataset wading image using YOLOv8

This dataset was collected from multiple social media platforms and includes various scenarios of buses being submerged, such as water depth submerging up to the tires, complete submersion of the vehicle body, half submersion of the vehicle body, and some situations in night scenes. These images are used for annotation work in LabelImg. The images and their corresponding labels are then used as input for the model to train computer vision models.