Intelligent construction site safety monitoring system and method based on machine vision

Abstract

The invention discloses an intelligent construction site safety monitoring system and method based on machine vision, and the system comprises a construction site data collection system and a safety behavior monitoring system, and the construction site data collection system employs a video capture device to collect construction site video data; the safety behavior monitoring system executes a construction site safety algorithm based on the input video data and identifies potential dangers; the safety behavior monitoring system comprises a target detection module which is used for extracting features of each frame of input image data and determining a detection frame position of a target; the target tracking module is used for tracking the target in the detection frame in real time and determining a track state; and the safety detection module is used for analyzing the behavior of the worker based on target detection and a target tracking algorithm, judging whether one or more conditions of not wearing a safety helmet and falling down exist, and detecting the potential fire hazard. According to the invention, the machine vision technology and the deep learning algorithm are combined, and the construction site risk violation behavior detection capability is improved.

Description

technical field [0001] The invention relates to construction site safety monitoring, in particular to a site safety monitoring system and method based on machine vision for target detection, tracking and safety early warning. Background technique [0002] The construction site environment is complex and changeable, and the current safety management is mainly manual. In the field of intelligent surveillance, traditional video surveillance often has only a few simple cameras, and the delay of the cameras is generally high. It is only used as an auxiliary means of on-site construction security, and cannot perform real-time detection and early warning of on-site security risks. This traditional supervision method not only consumes a lot of human and financial resources, but also has very low work efficiency, which is difficult to meet the safety management of modern intelligent construction. How to ensure the construction safety of workers and the intelligent and unified manage...

AI Technical Summary

Problems solved by technology

[0003] In recent years, more and more companies and scientific research institutions have devoted themselves to improving construction site safety monitoring systems. For example, Chinese Patent Publication No. CN110888950A discloses a smart construction site management system based on smart helmets, and Chinese Patent Application CN202010549716.2 discloses a A smart construction site safety detection method, system, intelligent terminal and storage medium, and some other technologies. However, although the construction site safety detection technology disclosed in the above several technologies is helpful for detecting whether to wear a safety helmet, the following problems still exist: In reality, there are many influencing factors that lead to safety accidents on the construction site. Workers may have fall accidents, and dense smoke and fire accidents may occur on the construction site.

In addition, due to the large number of construction site facilities, the situation of workers is easily blocked, and there is still room for improvement in the detection accuracy of existing systems / methods

Images