Debris Object Detection Caused by Vehicle Accidents Using UAV and Deep Learning Techniques

Abstract

The road debris clean-up process can be improved by utilizing drones, Deep Learning, and object detection to optimize the operation and re-open roads for traffic. Common debris is unsecured items that fly out from vehicles after a vehicle accident. The cleaning procedure of the road debris after an accident is cumbersome and sensitive. It demands much workforce and a time-consuming process to haul debris properly. The paper aims to detect debris on the road using a drone to minimize the time cleaning procedure. Object detection API with the pre-trained model of SSD and Faster R-CNN is used for object detection. The accuracy graphs, evaluation matrix, and detection box score determine the efficient model for debris detection.