关键词: YOLOv8, 目标检测, 轻量化, 注意力机制, 道路裂缝

Abstract: Deploying the YOLOv8 l model on the edge device for road crack detection can achieve high accuracy, but it is difficult to guarantee real-time detection. To solve this problem, a target detection algorithm based on the improved YOLOv8 model that can be deployed on the edge computing device Jetson AGX Xavier is proposed. First, the Faster Block structure was designed using partial convolution to replace the Bottleneck structure in the YOLOv8 C2f module, and the improved C2f module was recorded as C2f-Faster; Secondly, an SE (Squeeze-and Excitation) channel attention layer was connected after each C2f-Faster module in the YOLOv8 backbone network to further improve the detection accuracy. The experimental results on the open source road damage dataset RDD20 (Road Damage Detection 20) show that the average F1 score of the proposed method is 0.573, the number of detection frames per second (FPS) is 47, and the model size is 55.5MB. Compared with the SOTA (state-of-the-art) model of GRDDC20 (Global Road Damage Detection Challenge20), the F1 score has increased by 0.8 percentage points, the FPS has increased by 292%, and the model has been reduced by 42%. It realizes real-time and accurate detection of road cracks on edge devices.

Key words: YOLOv8(You Only Look Once version 8), object detection, lightweight, attention mechanism, road crack