To achieve automatic detection of workers wearing safety helmets in industrial settings,a detection algorithm based on improved YOLOv8 network model is proposed.By adopting the Faster-Net approach,we refined the C2f structure in the original YOLOv8 model,significantly reducing the model's parameter count and computational requirements.Additionally,we incorporated an EMA at-tention mechanism into the detection head to enhance the model's feature detection capabilities.Fur-thermore,we introduced SAConv and ASFF algorithms into the detection head to facilitate adaptive fusion of feature information across different scales,thereby improving the feature extraction capabili-ties of the refined model.Through ablation and comparative experiments,we demonstrate that the improved algorithm offers higher detection accuracy and faster detection speed,validating its feasibili-ty and suitability for practical applications in monitoring safety helmets wearing in industrial settings.
关键词
检测算法/YOLOv8/安全帽佩戴检测/FasterNet/注意力机制
Key words
detection algorithm/YOLOv8/detecting wearing behavior of safety helmet/FasterNet/attention mechanism
维普
万方数据