Electric power personnel safety state intelligent identification method based on real-time step counting and medium

Abstract

The invention discloses an electric power personnel safety state intelligent identification method based on real-time step counting. The method comprises the steps of detecting and tracking a boundingbox of each person in a video picture based on a multi-target tracking algorithm, and allocating a personnel ID number, inputting the image in the bounding box of the person into a key point detection algorithm to obtain left and right ankle key points and left and right knee key points of the human body, and associating the left and right ankle key points with the bounding box of the person according to the personal ID number, calculating the step number through the key point data, generating personnel state information by the step number and the personnel ID number, and generating personnelsafety state information based on the personnel state information and a preset safety rule. Key points of all parts of a human body are accurately obtained by analyzing a video, accurate states are obtained, recognition is effectively carried out, violation states are reminded and early warned, non-standard behaviors are prevented in advance, only a camera needs to be used for shooting related pictures, other hardware does not need to be worn, deployment is convenient, the running speed is high, and the method is suitable for different scenes.

Description

technical field [0001] The invention relates to the field of electric power safety, in particular to a method and medium for intelligently identifying the safety state of electric power personnel based on real-time step counting. Background technique [0002] Electric power construction has always emphasized the policy of "safety first, prevention first, and comprehensive management". With the operation of electric power equipment, there will inevitably be live construction or construction near live areas. The existing safety control methods are divided into theoretical training and In recent years, there have been occasional electric shock accidents in the way of on-site safety officer supervision and management or video recording. In order to better ensure the safety of personnel during construction, an intelligent identification terminal based on the safety status of personnel is now needed. wear a safety helmet, wear insulating shoes, whether the dress is standard, etc.)...

AI Technical Summary

Problems solved by technology

[0004] The above methods have the following disadvantages: (1) The sensor-based method can only indirectly judge the walking state of the human body through the collection of body motion state information by specific sensors, which will cause certain errors and cannot be very accurate; (2) In order to achieve detection, The human body needs to wear additional relevant sensors, which is cumbersome to use; (3) The information that can be collected using wearable sensors is very limited

[0005] At present, most of the security control is based on the form of security personnel and video recording. This kind of method has the following disadvantages: (1) It depends on the vision of personnel, and cannot cover all, let alone real-time supervision and viewing; (2) The way of recording video It is impossible to warn and prevent accidents in advance, it can only be analyzed at the right time; (3) The theoretical knowledge and experience requirements of personnel are high, and there may be omissions and insufficient supervision

Images