Railway clearance intrusion detection method  with binocular stereo vision

Xingfang Zhou; Baoqing Guo; Wei Wei

doi:10.1117/12.2315784

5 March 2018 Railway clearance intrusion detection method with binocular stereo vision

Xingfang Zhou, Baoqing Guo, Wei Wei

Proceedings Volume 10710, Young Scientists Forum 2017; 1071012 (2018) https://doi.org/10.1117/12.2315784
Event: Young Scientists Forum 2017, 2017, Shanghai, China

Abstract

In the stage of railway construction and operation, objects intruding railway clearance greatly threaten the safety of railway operation. Real-time intrusion detection is of great importance. For the shortcomings of depth insensitive and shadow interference of single image method, an intrusion detection method with binocular stereo vision is proposed to reconstruct the 3D scene for locating the objects and judging clearance intrusion. The binocular cameras are calibrated with Zhang Zhengyou’s method. In order to improve the 3D reconstruction speed, a suspicious region is firstly determined by background difference method of a single camera’s image sequences. The image rectification, stereo matching and 3D reconstruction process are only executed when there is a suspicious region. A transformation matrix from Camera Coordinate System(CCS) to Track Coordinate System(TCS) is computed with gauge constant and used to transfer the 3D point clouds into the TCS, then the 3D point clouds are used to calculate the object position and intrusion in TCS. The experiments in railway scene show that the position precision is better than 10mm. It is an effective way for clearance intrusion detection and can satisfy the requirement of railway application.

Citation Download Citation

Xingfang Zhou, Baoqing Guo, and Wei Wei "Railway clearance intrusion detection method with binocular stereo vision", Proc. SPIE 10710, Young Scientists Forum 2017, 1071012 (5 March 2018); https://doi.org/10.1117/12.2315784

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