Safety dressing detection method and system for construction workers

Abstract

The invention discloses a safety dressing detection method and system for construction workers. The method comprises the following steps: collecting the image data of a worker through a camera, and preprocessing the image data; performing index updating on the initial loss function of the convolutional neural network to obtain an improved loss function; carrying out multi-region feature extractionon the preprocessed image data based on the convolutional neural network after the initial loss function is updated so as to acquire positioning information of each region of the operating personnel;and establishing an identification and classification system of standard safety dresses, and carrying out normative judgment on the safety dresses of the operating personnel according to the positioning information of each area. In the embodiment of the invention, mainly aiming at safety dressing management of operating personnel in a complex construction environment, automatic and intelligent detection and early warning functions are realized by introducing a deep learning technology, and the detection efficiency and the detection precision are greatly improved.

Description

technical field [0001] The invention relates to the field of construction safety, in particular to a method and system for detecting safety clothing of construction workers. Background technique [0002] At present, construction safety is a key link in construction management. Among them, the safety management of construction personnel's clothing is particularly important for the safety risks of construction operations, including whether to wear safety helmets and reflective vests. At present, in terms of clothing safety management, it is more dependent on the safety operation training for on-site operators, and after the introduction of video surveillance systems in some construction scenes, the security personnel will conduct inspections in the monitoring room or on-site inspections. The manual detection method has the disadvantage of low efficiency. Until later, a method of using visual algorithms to realize automatic recognition and detection was proposed. When faced wi...

AI Technical Summary

Problems solved by technology

At present, in terms of clothing safety management, it is more dependent on the safety operation training for on-site operators, and after the introduction of video surveillance systems in some construction scenes, the security personnel will conduct inspections in the monitoring room or on-site inspections. Compared with manual detection methods, there are disadvantages of low efficiency.

Images