基于轻量级结构重参数化网络的口罩检测算法

Mask Detection Algorithm Based on Lightweight Structure and Re-parameterized Network

常态化疫情防控形势下,火车站、地铁站等公共场所人群密集,容易发生病毒的传播。针对人群密集场所口罩目标较小、模型参数量大、难以部署的问题,提出一种改进的轻量级结构重参数化网络。在Retinaface算法上,使用双重级联金字塔网络替换原有的特征融合网络,增强特征信息,提高对小尺度目标的检测效果;同时使用结构重参数化网络RepVGG替换原有的MobileNet0.25主干网络,在模型训练时,通过残差结构提高模型特征提取能力,在模型推理时,通过模型结构重新参数化减少模型参数,提高推理速度。实验结果表明,本文算法在GPU上帧率达到92.59 fps,在自建数据集的3个不同等级的验证集上的平均准确率(mAP)达到94.17%、93.30%、86.88%,相比原始Retinaface算法分别提高了1.17个百分点、2.89个百分点、5.35个百分点,可以更好地在自然场景中进行口罩佩戴检测。

Under the situation of normalized epidemic prevention and control,there are dense crowds in railway stations,subway stations and other public places,which are prone to the spread of virus.Aiming at the problems of small mask targets,large amount of model parameters and difficult to deploy in crowded places,an improved lightweight structure and re-parameterized network is proposed.On the Retinaface algorithm,the dual cascade pyramid network is used to replace the original feature fusion network to enhance the feature information and to improve the detection effect of small-scale targets.At the same time,the structure re-parameterized network RepVGG is used to replace the original MobileNet0.25 backbone network.During model training,the residual structure is used to improve the feature extraction ability of the model.During model reasoning,the model parameters are reduced and the reasoning speed is improved by re-parameterization of the model structure.The experimental results show that the frame rate is 92.59 fps and the average accuracy rate(mAP)of the proposed algorithm on three different levels of verification sets of self-building data sets is 94.17%,93.30%,86.88%,which is 1.17 percentage points,2.89 percentage points and 5.35 percentage points higher than the original Retinaface algorithm respectively.Mask wearing detection can be better carried out in natural scenes.