Research on Helmet Wearing Detection for Dense Cycling Crowd Based on Improved YOLOv8n

An improved YOLOv8n algorithm is proposed to address the issue of low helmet detection accuracy caused by the high density of electric bike riders in complex scenarios. Introducing Channel Prior Convolutional Attention (CPCA) mechanism to enhance the feature extraction capability of the model; Introducing Symmetric Positive Definite Convolution (SPDConv) to enhance ordinary convolutions; Replace the bounding box loss function with a Weighted Intersection over Union (Wise IoU) function to improve the model’s ability to locate small targets. The experimental results showed that the improved model improved the mAP accuracy detected on the self-made dataset by 4.4%, with only a slight increase in the number of model parameters, and has the potential to be used as a solution in practical applications.