Artificial intelligence identification system for underground coal mine unsafe behaviors based on deep learning

Abstract

The invention discloses an artificial intelligence identification system for underground coal mine unsafe behaviors based on deep learning. The system comprises mine field instance material collection, a training data set, an AI training platform, an algorithm model, a control platform and the like. According to the AI training platform based on deep learning technology, a training data set generated by collecting mine field instance materials is loaded. The underground coal mine unsafe behaviors are trained. After an algorithm model is acquired, the information is directly imported into an AI camera or an NVR through a control platform. An AI camera or NVR performs artificial intelligence analysis on a real-time video stream, identifies trained unsafe behaviors, pushes an inference result to a control platform, pushes unsafe information to an administrator by the control platform, checks the unsafe information, communicates with the site, confirms and stops the unsafe behaviors, pushes the unsafe behaviors to an administrator and pushes the unsafe behaviors to an unsafe place at the same time, so that acousto-optic warning is carried out on the site. Accidents are reduced, and the application range is wide.

Description

Technical field [0001] The invention relates to an artificial intelligence recognition system for unsafe behavior in underground coal mines based on deep learning, and belongs to the field of artificial intelligence. Background technique [0002] The vast majority of coal mines in my country have the characteristics of complex geological conditions, difficult mining, many types of disasters, and wide distribution. There are many hidden dangers such as unsafe behaviors and unsafe conditions of objects, which lead to accidents in coal mine production and cause personal injury. casualties. Due to the multi-faceted, wide-ranging links in coal mines, it is impossible for on-site safety supervisors to cover everything, and it is impossible to keep an eye on a certain place 24 hours a day. There are many blind spots and risk points for safety supervision, and the cost of coal mine safety management increases. Summary of the invention [0003] The purpose of the present invention is to pr...

AI Technical Summary

Problems solved by technology

Because there are many underground sites in the coal mine, it is impossible for the on-site safety supervisors to cover everything, and it is impossible to keep an eye on a certain location 24 hours a day, resulting in many blind spots and risk points in safety supervision, and the cost of coal mine safety management increases.

Images