Construction site behavior safety detection and recognition method and system based on YOLOv5

Abstract

The invention provides a construction site behavior safety detection and recognition method and system based on YOLOv5, and relates to the field of constructional engineering. The method comprises the following steps: obtaining to-be-recognized construction site image information; adopting a pre-trained detection model to recognize the to-be-recognized construction site image information, detecting the wearing condition of the safety helmet, wherein the detection model is a deep learning model, and a training set used for training the detection model comprises image information marked with the wearing position of the safety helmet; according to the invention, integrated engineering safety supervision application results of a machine vision technology, an intelligent construction site technology and a neural network technology are integrated, real-time monitoring and warning of the safety helmet wearing condition of the worker on the construction site are realized, and the safety helmet wearing condition of the worker on the construction site is monitored and warned. Therefore, the standardization of project management is enhanced, the overall efficiency of the project is improved, the cost consumption is reduced, the occurrence probability of accidents and casualties is reduced, and the construction of intelligent cities and intelligent construction sites is promoted.

Description

technical field [0001] The invention relates to the field of construction engineering, in particular to a YOLOv5-based construction site behavior safety detection and recognition method and system. Background technique [0002] In the process of building construction, there are many potential safety hazards, and the incidence of safety accidents remains high. In the long-term practical demonstration, before starting construction operations, checking the behavioral ability of construction workers and wearing safety equipment can effectively reduce the probability of accidents. Therefore, in daily construction operations, it is particularly important to supervise whether workers wear safety helmets and other safety facilities. However, most of the construction sites currently use manual monitoring, which relies heavily on experienced managers on site. They need to observe and inspect in real time, which is time-consuming and laborious. There is a low level of automation and a...

AI Technical Summary

Problems solved by technology

However, for the construction site, some special scenes such as industrial and mining with complex environments have very high requirements for the real-time performance and accuracy of target detection. For these external interference characteristic information, the machine cannot make accurate judgments in many cases. The real-time performance and accuracy of target detection substances are low, and the versatility is not strong

Images