Algorithm improvement for traffic sign detection based on YOLOv8

Abstract:

Traffic sign detection, as an important part of automatic driving, occupies a great influence in practical applications, but it is easily affected by many factors such as weather, road conditions, complex background, etc., so an improved algorithm for traffic sign detection based on YOLOv8 is proposed. Firstly, YOLOv8n is used as the base model of the network, the iRMB_EMA attention mechanism is used to improve the model's ability to perceive small targets, which reduces the leakage detection phenomenon of the model, Conv is upgraded to RFAConv, which improves the model's ability to deal with details and complexity in the image, the idea of ASFF adaptive spatial feature fusion is introduced in the detection head, and the small target detection layer, a four-head detection head is designed to improve the model's ability to detect small targets, solves the case of feature loss due to cross-scale fusion, and provides a more accurate loss metric by calculating the distance of keypoints between the predicted and real frames using the Inner-MPDIou loss function. The experimental results of this algorithm on the public dataset CCTSDB show that the improved model accuracy mPA reaches 82.6%, which is 4% higher than the original base model YOLOv8n. This algorithm effectively improves the problem of detail perception and leakage of the model in the detection of small targets, and has a significant detection effect compared to other algorithms.

Key words:traffic sign detection; YOLOv8; RFAConv; FASFFHead; Inner-MPDIou