Combined with the construction of smart mine, we conduct site staff individual body language collection, analysis, training and eventually form normalized training set summary. Then we use a body language detection and behavior analysis method based on deep learning to analyze and compare the video monitoring of employees' body movements during operation with the training set prepared in the previous stage. Through real-time comparison and analysis, some body movements that may lead to unsafe accidents are captured and analyzed, then audible and visual alarm signals are sent to remind the operating staff in real time. In this way, we effectively prevent safety accidents caused by non-standard body movements during the operation of employees, and reduce the illegal operation of employees scientifically and effectively. Secondly, some advanced exoskeleton equipment will be used to assist employees in positions with high labor intensity and high risk. Meanwhile, the body movements of employees will be collected, analyzed, processed and excavated continuously, so as to find the key factors that affect the improvement of safety and efficiency during operation, by which targeted training and improvement could be carried out. It is of practical significance to enhance the operation safety of individual employees,  promote the cultivation of standard operation behaviors of them and ensure the safety of mines.

 

big data；body language；exoskeleton；safe operation；developing normative behavior