基于改进YOLOv8的电力作业人员安全带检测

Seat Belt Detection for Electrical Workers Based on Improved YOLOv8

正确穿戴安全带是预防电力作业人员高空坠落的重要措施;针对在电力现场中作业人员是否穿戴安全带检测效率低以及实效性差的问题,提出了一种基于YOLOv8n的电力作业人员安全带检测方法;该算法在特征提取网络中添加通道注意力模块,提升了模型的识别能力;引入了加权双向特征金字塔网络结构进行特征融合,提高特征学习能力,并降低模型复杂度;将原先的CIoU损失函数替换为WIoU损失函数,进一步提高了模型的检测效率以及对于小目标的辨识能力;实验结果表明,该算法的mAP平均精度均值达到96.5%,识别效果明显提升,并且优于其他经典目标检测模型,验证了新算法的有效性。

Wearing seat belts correctly is an important measure to prevent electrical workers from falling from heights.In order to solve the low efficiency and poor effectiveness of wearing seat belts for workers in power sites,a safety belt detection method for power workers based on improved YOLOv8 is proposed.The algorithm adds a squeeze and excitation(SE)attention mechanism in the network to improve the recognition ability of the model.Also,a weighted bidirectional feature pyramid network structure is introduced to perform the feature fusion,and improve the feature learning ability,and reduce the complexity of the model.The WIoU loss function is used to replace the original CIoU loss function,which further improves the detection accuracy and adaptability of the model to small targets.Experimental results show that the proposed algorithm reaches the average accuracy by 96.5%,improving the recognition effect significantly,the proposed model is better than other classical object detection ones,which verifies the effectiveness of the new algorithm.