Unlock instant, AI-driven research and patent intelligence for your innovation.

Field operation safe wearing identification method, system and device and storage medium

A field operation and recognition method technology, applied in biometric recognition, neural learning methods, character and pattern recognition, etc., can solve the problems of difficulty in extracting wear features, and the recognition method cannot take into account speed and accuracy, so as to enhance the ability of feature extraction, The effect of improving robustness and accuracy, reducing the effects of proportional imbalance

Pending Publication Date: 2022-03-25
BEIJING SGITG ACCENTURE INFORMATION TECH CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the above-mentioned prior art, a method, system, device and storage medium for on-site operation safety wearing identification, and solve the problem that the current identification method cannot take into account both speed and accuracy, and it is difficult to extract wearing features under multiple overlapping targets

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Field operation safe wearing identification method, system and device and storage medium
  • Field operation safe wearing identification method, system and device and storage medium
  • Field operation safe wearing identification method, system and device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0141] The specific implementation process is mainly divided into six steps: the collection and area labeling of on-site operation safety wearable images; image preprocessing; image depth feature extraction through a layered network based on swin-transformer; The interference of the background enhances the attention to the characteristics of safe wear; through the area generation and classification regression network, the category probability and location information are predicted; the overall loss function for unbalanced samples is designed for model training.

[0142] Specifically include the following steps:

[0143] Step 1: Collect images of safety wear for on-site operations. The images are mainly taken from the pictures of the power operation site taken by a provincial power company. Put different category labels on the four types of wearing safety helmets correctly, wearing safety helmets incorrectly, wearing work clothes, and not wearing work clothes, and mark the targ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the field of electric power safety image recognition, and discloses a field operation safety wearing recognition method, system and device and a storage medium, and the method comprises the steps: S1, obtaining safety wearing image data of electric power field historical operation, and carrying out the marking of a target frame region of a safety wearing image; s2, carrying out preprocessing on the marked safe wearing image data; s3, carrying out feature extraction on the preprocessed safe wearing image data to obtain safe wearing depth features; and S4, inputting the safe wearing depth features into a pre-constructed safe wearing identification model to obtain probability prediction of whether the user wears the wearable device correctly and target positioning information. The problems that an existing recognition method cannot give consideration to speed and precision and wearing features are difficult to extract under multiple overlapped targets are solved.

Description

technical field [0001] The invention belongs to the field of electric power safety image recognition, and relates to a method, system, equipment and storage medium for on-site operation safety wear recognition. Background technique [0002] In the actual power production operation, patrol inspection and maintenance process, due to the uneven quality of operators, imperfect safety training and management system, there are a large number of bad operating behaviors such as irregular operating procedures and even habitual violations, which greatly increase Reduced on-site operation risks and buried potential safety hazards. In recent years, more than 90% of accidents in electric power companies are caused by illegal operations, among which failure to wear safety protective equipment correctly is one of the main illegal factors, causing tragedy of personal casualties and huge economic losses. Therefore, real-time detection of the safety wearing conditions of on-site workers is a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06V40/10G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/047G06N3/048G06N3/045G06F18/24
Inventor 许中平孙泽锋毕艳冰赵恩来安丽丽沈睿何小瑜李向阳苏彦龙王高峰晏荣煜余金沄秦亮
Owner BEIJING SGITG ACCENTURE INFORMATION TECH CO LTD