Research on orbital foreign object intrusion detection method based on image segmentation

Youwei Yang; Chao Wang; Jianjun Wang

doi:10.1117/12.3024686

21 February 2024 Research on orbital foreign object intrusion detection method based on image segmentation

Youwei Yang, Chao Wang, Jianjun Wang

Proceedings Volume 12988, Second International Conference on Environmental Remote Sensing and Geographic Information Technology (ERSGIT 2023); 129881D (2024) https://doi.org/10.1117/12.3024686
Event: Second International Conference on Environmental Remote Sensing and Geographic Information Technology (ERSGIT 2023), 2023, Xi’an, China

Abstract

To improve the speed and accuracy of detecting track foreign object intrusions, the OneFormer image segmentation technology is introduced. This technology aids in the identification of track foreign object intrusions, thereby extending the use of image segmentation technology in the field of track foreign object intrusion detection. This paper involves the collection of a substantial volume of images depicting track foreign object intrusions in real-world engineering settings. Following preprocessing, a dataset comprising 10,060 images across five categories of track images was constructed. Subsequently, the experimental environment configuration and training parameters were specified. Finally, a comparative experiment was conducted with other segmentation methods, validating the method's feasibility based on the respective panoramic segmentation evaluation metrics, including PQ, SQ, and RQ. The method presented in this paper exhibits improvements in each evaluation index when compared to other mainstream segmentation methods, such as Panoptic FCN, UPSNet, and OANet. The method effectively addresses the issues of low detection efficiency and limited accuracy encountered in graphical segmentation methods within the domain of track foreign object intrusion detection. Moreover, it successfully overcomes the limitations associated with traditional methods and manual detection processes. As a result, the OneFormer method introduced in this study is better suited for the task of recognizing track foreign object intrusions in real operational environments. It not only offers enhanced performance in terms of detection efficiency and accuracy but also serves as a valuable reference and driver for the application of image segmentation in actual railroad engineering scenarios.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Youwei Yang, Chao Wang, and Jianjun Wang "Research on orbital foreign object intrusion detection method based on image segmentation", Proc. SPIE 12988, Second International Conference on Environmental Remote Sensing and Geographic Information Technology (ERSGIT 2023), 129881D (21 February 2024); https://doi.org/10.1117/12.3024686

ACCESS THE FULL ARTICLE

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
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image segmentation

Panoramic photography

Education and training

Object detection

Computer intrusion detection

Semantics

Binary data

Show All Keywords

Keywords/Phrases

Search In:

Publication Years