Workers support unsafe behaviors through visual automatic detection method and system

Abstract

The invention belongs to the related technical field of building engineering informatization, and discloses a visual automatic detection method and system for behaviors of workers to pass through an unsafe support, which comprises the following steps: (1) based on a construction site monitoring video, collecting original data about the workers passing through a support structure under the condition of no safety protection facility; and (2) manually marking the images in the original data to form a data set, the labels are divided into three types, namely a steel support, a concrete support anda worker; and (3) training a Mask RCNN-based segmentation mask prediction model by using the COCO public data set and the labeled data set; and (4) the segmentation mask generated by the prediction model is output to an overlap judgment module, and the overlap judgment module judges the position relation between workers and supports and determines whether unsafe behaviors of the workers passing through the deep foundation pit support exist or not. The behavior of unsafely passing through the foundation pit supporting structure can be detected, the cost is low, popularization is easy, and thehigh automation degree is achieved.

Description

technical field [0001] The invention belongs to the related technical field of construction engineering informatization, and more specifically relates to an automatic detection method and system for unsafe behaviors of construction site workers passing support structures based on (monitoring) video instance segmentation. Background technique [0002] Fall from height (FFH) is an important cause of safety accidents at construction sites. There are many safety policies and procedures for fall from height (FFH). For example, the Occupational Safety and Health Association (OSHA) requires that working on a six-foot-high edge must be protected by fences, safety nets, and worker fall prevention systems (PFAS). [0003] In order to ensure the safety of foundation pit construction, support structures such as concrete supports and steel supports are often set up for deep foundations. On-site experience and monitoring records show that when workers are active at the construction site...

AI Technical Summary

Problems solved by technology

[0004] The current mainstream detection measures are based on on-site inspection. Due to the complete reliance on on-site safety management personnel, there are defects such as time-consuming, manpower-consuming, and lack of timeliness.

Images