Hazardous Chemical Vehicles Detection Based on Path Fusion and Multi-Scale Slicing

In order to ensure the safety of highway transportation, hazardous chemical vehicles detection was carried out on vehicles engaged in the transportation of hazardous chemicals. High-speed and diverse scenes of vehicles on the road pose numerous challenges for object detection, as hazardous chemical transport vehicles often appear as small targets within the surveillance field of view. In this paper, we propose a dedicated vehicle detection algorithm based on path fusion and multi-scale slicing methods. This algorithm incorporates a dual path fusion module and a multi-scale image slice detection module to effectively enhance its feature extraction capability and fuse semantic detail information. By conducting experiments on a self-built dataset, we have demonstrated the reliability and superiority of the proposed algorithm over baseline models and mainstream object detection algorithms in terms of detection accuracy and inference speed.