YOLO-GCEF: A Novel Detection Algorithm for Timber Transport Trucks in Forest Farms

Vehicle detection technology is essential for automatic monitoring and AI-assisted driving systems. Detecting timber transport trucks presents several challenges, including a lack of datasets and complex environment. To address these issues, this paper constructs a new dataset called Timber Transport Trucks (TTT) dataset and proposes a novel vehicle detection method: YOLO-GCEF. This method incorporates the GC attention module into the backbone network to obtain critical information for vehicle detection tasks, thereby improving detection accuracy. Additionally, EWF loss is proposed, which combines Efficient IOU loss(EIOU), weighted bounding box regression loss, and Focal Loss. We conducted some experiments on the TTT dataset. The experimental results show that YOLO-GCEF can achieve 90.3% mAP.