The design and implementation of mask wearing recognition system based on YOLO V5
Sun, Ningyang (2022)
2022
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2022102521579
https://urn.fi/URN:NBN:fi:amk-2022102521579
Tiivistelmä
With the rapid development of deep learning and computer vision, related technologies and equip-ment have been widely used in various fields. And the most basic part of computer vision, as well as the most important step in image understanding, is target detection. Target detection can select the specified object in a large number of captured images. This helps researchers confirm more infor-mation about the target object, and further locate and classify the object.
Based on the nature and characteristics of target detection, combined with the current situation of the epidemic, computer vision should be more applied to epidemic prevention. According to multiple studies, in addition to COVID, there are a number of seasonal diseases that can be reduced by wear-ing masks. Target detection can be used to judge the situation of people wearing masks and help the staff in public places to judge the current environment and further processing in time. At present, such tools have not been widely used in the market, so the main direction of this system is to design a practical, lightweight and universal mask recognition system.
The system is designed to detect the wearing of masks in complex scenarios, which can help employees in crowded public places judge the situation of people wearing masks. he system uses Pytorch as a framework, uses YOLO V5 target detection algorithm to classify the video or image after feature extraction, which can make more accurate judgment, and uses PyQt to make a simple window that presents the results to the user.
Based on the nature and characteristics of target detection, combined with the current situation of the epidemic, computer vision should be more applied to epidemic prevention. According to multiple studies, in addition to COVID, there are a number of seasonal diseases that can be reduced by wear-ing masks. Target detection can be used to judge the situation of people wearing masks and help the staff in public places to judge the current environment and further processing in time. At present, such tools have not been widely used in the market, so the main direction of this system is to design a practical, lightweight and universal mask recognition system.
The system is designed to detect the wearing of masks in complex scenarios, which can help employees in crowded public places judge the situation of people wearing masks. he system uses Pytorch as a framework, uses YOLO V5 target detection algorithm to classify the video or image after feature extraction, which can make more accurate judgment, and uses PyQt to make a simple window that presents the results to the user.