The Development Of Workshop Dress Safety Early Warning System Based On SSD

The development of industry promotes the progress of human society,and the realization of this progress is inseparable from the contribution of the production workshop.However,there are some safety hazards in the production workshop,such as irregular dress and illegal operation,which may pose a threat to the physical safety of the operator.In this context,aiming at the phenomenon of irregular dressing in the practical training workshop,this paper develops a set of safety warning system to detect whether the dress of the workshop operators is standardized,and gives a safety warning when finding irregular dressing,so as to ensure the life safety of the operators.The main research contents are:(1)Mobile Net V1 is selected as the feature extraction network to build a lightweight SSD target detection network,and the experiment shows that the model parameters can be reduced,the model size can be compressed,and the development cost can be reduced under the condition of slightly reducing the accuracy.(2)In order to improve the accuracy of the dress code detection model,the Convolutional Block Attention Module(CBAM)is added to the feature extraction network due to the weak feature extraction ability of the SSD algorithm.This mechanism can enhance the channel dimension and space dimension of the feature graph,so as to refine the feature of the feature graph.In addition,the feature fusion mechanism is introduced to further improve the accuracy of the model.By selecting a target prediction layer as the measurement scale,the feature layer above the scale is up-sampled,and the feature layer below the scale is down-sampled.Then the feature map containing more semantic information is obtained by fusing the features of the prediction layer and the adjacent layer.The experiment shows that these two improvement measures can effectively improve the accuracy of dress code detection model.(3)Based on the dress code detection model,the workshop dress safety warning system is designed and implemented.The system is divided into three subsystems:background management system,teacher end and student end.Background management system functions include: basic information management,violation statistics,warning,feedback processing;Teacher side functions include: face registration,face login,basic information management,warning information;Student side functions include: face registration,face login,dress code detection,alarm,warning information.In this study,three improvements were made to the SSD algorithm,and experiments were carried out on the self-made dress code dataset.The experimental results show that the performance of the improved SSD model is better than the original SSD model.In addition,this study also developed a workshop dress safety early warning system aimed at applying the dress code detection model.The system can detect the dress code of the operator,and can issue safety warning,effectively protect the life safety of the operator,and improve the safety awareness of the operator.