CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume 12058 > Article
Paper
22 December 2021 DMI image information recognition based on improved YOLOv4 with focal loss
He Han, Li Kaicheng, Lei Yuan, Wang Fei
Author Affiliations +
Proceedings Volume 12058, Fifth International Conference on Traffic Engineering and Transportation System (ICTETS 2021); 1205822 (2021) https://doi.org/10.1117/12.2620011
Event: 5th International Conference on Traffic Engineering and Transportation System (ICTETS 2021), 2021, Chongqing, China
Abstract
In the ATP function test project of train control system, the image information recognition of driver machine interface (DMI) is an important part of realizing test automation, because of the reason the real-time detection of DMI image information is very important. In the process of realizing DMI image information recognition based on machine vision, aiming at the problem of low generalization of the model caused by small amount of training data and single background of training image in machine vision system, combined with a variety of data enhancement strategies, a data set construction scheme of the system is proposed. In order to solve the problem of low attention to hard samples when using YOLOv4 model for small target detection, based on the idea of focus loss, a weighted fusion loss function of focus loss + CIoU loss is proposed, and the optimal parameters are obtained through experiments. The results show that the average accuracy of this method for DMI image information recognition is 98.76%, which is 2.6% higher than the traditional YOLOv4 detection model, and the single detection time is 76.9ms, which has a certain engineering application value.
© (2021) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
He Han, Li Kaicheng, Lei Yuan, and Wang Fei "DMI image information recognition based on improved YOLOv4 with focal loss", Proc. SPIE 12058, Fifth International Conference on Traffic Engineering and Transportation System (ICTETS 2021), 1205822 (22 December 2021); https://doi.org/10.1117/12.2620011
ACCESS THE FULL ARTICLE
ORGANIZATIONAL
Sign in with credentials provided by your organization.
INSTITUTIONAL
Select your institution to access the SPIE Digital Library.
SELECT YOUR INSTITUTION
PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
PERSONAL SIGN IN
No SPIE Account? Create one
PURCHASE THIS CONTENT
SUBSCRIBE TO DIGITAL LIBRARY
50 downloads per 1-year subscription
Members: $195
Non-members: $335 ADD TO CART
25 downloads per 1 - year subscription
Members: $145
Non-members: $250 ADD TO CART
PURCHASE SINGLE ARTICLE
Includes PDF, HTML & Video, when available
Members: $17.00
Non-members: $21.00 ADD TO CART
PROCEEDINGS
 12 PAGES
DOWNLOAD PAPER SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Target detection
Data modeling
Image enhancement
Statistical modeling
Control systems
Machine vision
Cameras
RELATED CONTENT
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top