Detection of Dress Code of Workers Based on PPYOLO
An improved deep learning detection algorithm PPYOLO-CA has been proposed to solve the problem of low efficiency of traditional manual inspection workers'standardized dressing.First,the deep separable convolution is used to replace the former part of the backbone convolution.The deep separable convolution can reduce the parameters and increase the nonlinearity of the network,thus enhancing its ability to extract features;Secondly,by adding CoordAtt module to the feature extraction section of backbone,to improve the ability of image spatial information extraction and provide rich feature information for target detection;Finally,changing SPP module into SPPF module can effectively reduce the parameters of the model and improve the running speed of the model without affecting the accuracy.In order to verify the algorithm,simulation experiments are carried out on the data set of workers'clothing.The experimental results show that the improved model has better recognition effect than the original model in different scenarios.The improved model on mAP is 1.8%higher than that of the original model and 0.14 MB less than that in model parameters.
NETL
NSTL
万方数据
