Chongfan Technology
All Categories >

09

2025

-

04

Optical Precision Engineering Cover | Trackside Intelligent Detection: Stereoscopic Vision "Noise-Resistant Armor"

Author:

Optics and Precision Engineering, 2025, Vol. 33, No. 3

Introduction

In the field of rail transit maintenance, the pantograph, as a core component of the train's power supply system, its accurate and real-time anomaly detection is directly related to operational safety. Traditional anomaly detection of core train components is usually carried out at fixed intervals, which is time-consuming and laborious. Tianjin University Laser and Optoelectronic Testing Team To address this issue, a trackside anomaly detection system was designed. This system can detect abnormal states of key train components and can resist environmental noise interference, providing a new solution for rail transit equipment monitoring in complex environments. This research, entitled " Trackside Anomaly Detection of Pantographs Using Binocular Vision under Strong Environmental Noise ", was published in Optics and Precision Engineering (a journal supported by the second phase of the China Science and Technology Journal Excellence Program, indexed by EI and Scopus, a core Chinese journal, and a "T1" journal in the "High-Quality Scientific and Technological Journal Classification Catalog" for instrumentation and optics and optical engineering), and was selected as the cover article.

Figure 1: Schematic diagram of intelligent pantograph trackside detection

Source: The research team of this article

Technical Challenges in a Strong Noise Environment

Pantograph trackside detection has long been faced with the problem of strong environmental noise interference. Traditional binocular vision systems are easily prone to inaccurate disparity map calculations due to factors such as vibration and structural creep, which in turn affects the anomaly detection results. Especially under harsh conditions such as rain, snow, and strong winds, the false detection rate remains high, and the cost of manual review increases significantly. How to achieve stable and efficient automated detection in high-noise scenarios has become a technical bottleneck that the industry urgently needs to overcome.

Innovative Algorithms to Solve Core Pain Points

The research team started with the optimization of the stereo matching algorithm and proposed three key technological breakthroughs:
1. Anti-interference cost calculation: By using extended neighborhood cost calculation and dynamic weighted aggregation methods, the problem of inaccurate epipolar line matching caused by environmental vibration is effectively offset, ensuring that disparity calculation maintains high robustness even in noisy environments, constructing a "noise-resistant armor" for stereo vision;
2. Scene-adaptive detection: Combining the correlation of local texture information to construct a cost distribution function, customized optimization is performed for the structural characteristics of the pantograph model, significantly improving the accuracy of feature matching;

3. Real-time optimization: Combining the model information of the pantograph component being measured, the disparity search space is limited, reducing redundant calculations, and the algorithm processing speed is increased by more than 30%, meeting the millisecond-level response requirements of trackside equipment.

 

The research team reconstructed the pantograph trackside anomaly detection system framework using innovative algorithms, dividing the entire detection process into data acquisition, stereo vision anti-noise calculation, and anomaly feature calculation parts.


Figure 2: Binocular vision pantograph anomaly detection system framework

Source: Optics and Precision Engineering

3D Reconstruction and Intelligent Diagnosis Closed Loop

Based on the precise acquisition of disparity data, the research further integrates 3D point cloud reconstruction technology. By analyzing the spatial distribution characteristics of the point cloud in key parts of the pantograph (such as carbon slip, hinge), automated identification of abnormal phenomena such as cracks and wear is achieved, effectively improving the stability of the system's anomaly diagnosis.

 

 

Field Application and Industry Value

This technology has been field-tested in the pantograph monitoring system of a certain metro line. Compared with the original equipment, the new system maintains stable output even under extreme conditions such as strong vibration in tunnels and low light at night. This research not only alleviates the pain points of trackside detection equipment's "inability to see clearly and calculate accurately," but also provides key technical support for unmanned operation and maintenance of intelligent rail transit.

 

 

Future Outlook

The research team revealed that the next step will be to promote the deep integration of algorithms with 5G and edge computing technologies to achieve cloud-based collaborative diagnosis of pantograph status. With the increasing demand for intelligent monitoring in the "Outline of Building a Strong Transportation Country," this type of high-precision and high-efficiency detection technology is expected to become a new industry standard, establishing a "China model" for the global rail transit safety assurance system.

 

 

Paper Information

Zhao Jin, Guo Yin, Yin Shibin, et al. Trackside Anomaly Detection of Pantographs Using Binocular Vision under Strong Environmental Noise [J]. Optics and Precision Engineering, 2025, 33(03): 438-451. DOI: 10.37188/OPE.20253303.0438.

https://ope.lightpublishing.cn/thesisDetails#10.37188/OPE.20253303.0438

 

 

Team Introduction

The Tianjin University Laser and Optoelectronic Testing Technology Team is based in the School of Precision Instruments and Optoelectronics Engineering (hereinafter referred to as "Precision Instruments College") of Tianjin University and is one of the core research teams of the national key discipline "Instrument Science and Technology." After years of development, the team has become a research group with significant influence in the field of laser testing technology in China. It mainly engages in research on laser precision measurement technology, optoelectronic detection and imaging, and intelligent optical instruments. It has accumulated a large number of theoretical achievements and engineering experiences in the fields of automobile manufacturing, aerospace, and rail transit maintenance.

Previous

Broadband Perfect Littrow Diffraction Metasurface under Large Angle Incidence

Notice of the 22nd China Young Editors' Academic Seminar on Science and Technology Journals and the First Northeast China Seminar on Innovation and Development of Science and Technology Journals (First Round)

Next

Previous

Broadband Perfect Littrow Diffraction Metasurface under Large Angle Incidence

Next

Notice of the 22nd China Young Editors' Academic Seminar on Science and Technology Journals and the First Northeast China Seminar on Innovation and Development of Science and Technology Journals (First Round)

LATEST NEWS