AI Enhances Railway Security – Optical- and Video-Based Perimeter Detection Safeguards Railway Perimeters

The railway network, vital for connecting cities and driving economic growth, relies heavily on the security of its perimeters. These perimeters face a variety of security challenges, including intrusions by people and foreign objects, as well as complex operational conditions and a lack of safety facilities. Given these challenges, there is an urgent need for a flexible and adaptive security approach to ensure the smooth operation of the railway transportation system. Civil engineers and designers play a crucial role in addressing these challenges, but often tight budgets and a lack of knowledge about available security solutions hinder their efforts. Adopting new technologies that help with improving railway perimeter security is therefore essential to safeguarding the network.

Optical- and Video-Based Perimeter Detection Innovates Railway Protection

To move away from conventional labour and facility assisted protection, the sensing properties of fiber optic cables, combined with intelligent vision algorithms, and Artificial Intelligence (AI) are being harnessed to provide reliable and efficient security solutions that can operate 24 hours a day. These advanced technologies are being seamlessly integrated to establish a perimeter detection system that meets the demands of modern railway security.

Traditional railway protection systems usually adopt vibration cables and optical fibers, cameras, and other technologies, which are mature and have been used for a long time. However, the technologies also lead to certain problems.

Take vibration cables as an example. As the electrical signal attenuates, the amount of received data also decreases. Insufficient cable coverage results in a short detection distance (400 metres in general). In addition, limited by sensor capabilities, electromagnetic induction suffers low transmission frequency. Worse still, dedicated algorithm processors for data processing are unavailable, making it difficult to ensure alarm accuracy.

Although free from trackside power supply and terrain restrictions, the simple vibration optical fiber mode commonly used today faces multiple challenges such as typhoon, rain, and vehicle vibration, resulting in false alarms and weakening protection effects.

Protection methods mentioned above often adopt a single technology coupled with labor-intensive inspections, which fail to support all- whether detection and round-the-clock monitoring.

Thanks to Huawei’s extensive research and deep experience in optical and AI domains, the innovative optical and video-based perimeter detection solution for railways was developed, and tailored specifically for railway protection scenarios.

Using Huawei’s intelligent fiber sensing and cameras, the solution is able to stably detect, review, and report intrusions by integrating algorithm systems, video management platforms, and maps.

Huawei’s Optical- and Video-Based Perimeter Detection Solution eliminates distance and algorithm restrictions and escalates the perimeter protection to whole new level.

It delivers near-zero missed alarms and very few false alarms and provides 24/7 protection in any weather, significantly reducing manual patrol workload.

Near-Zero Missed Alarms

Huawei’s Optical- and Video-Based Perimeter Detection Solution uses the enhanced oDSP module, which improves the effective signal collection rate to 99.9%. Additionally, the dynamic range of vibration monitoring surpasses 50 dB, four times higher than that of industry-standard technologies. Simply put, the system recognises with high degrees of precision, extremely slight as well as violent vibration events. This ensures near-zero missed alarms and effectively safeguards railway lines.

Strong Anti-Interference Capability Ensures Very Few False Alarms

Huawei’s Optical- and Video-Based Perimeter Detection Solution adopts a unique feature identification algorithm to extract key intrusion features through feature reconstruction, and generates alarms based on a judgment mechanism to resist environmental interference, such as strong winds (level-7 wind) and heavy rain. It significantly relieves the on-site disposal workload caused by false alarms. AI and big data technologies continuously enhance railway perimeter protection during actual operations of the platform.

Designed to quickly detect high-risk activities such as climbing or damaged fences, the solution employs intelligent cameras, vibration-sensing fibers, and intelligent sensing devices to identify vibration ripples of intrusions and filters interference events. The devices link with intelligent Pan-Tilt-Zoom (PTZ) dome cameras for real-time analysis, secondary reviews, and prompt alarm generation, ensuring railway perimeter safety.

The solution also provides multi-point intrusion prevention. It can detect and report alarms for at least 10 concurrent substantial intrusions, such as climbing, crossing, and damaging the fence, within the specified coverage area.

Convenient Deployment at a Low Cost

The solution features single-device, single-fiber coverage. Conventionally, devices need to be deployed at an interval of several hundred meters. In contrast, this solution requires only one monitoring device plus one optical fiber to form a 20-km security defence line, simplifying device O&M and slashing costs. In addition, customers can divide and configure protection zones to adapt to dynamic physical surroundings, reducing unnecessary construction costs.

To date, Huawei’s Optical- and Video-Based Perimeter Detection Solution has been successfully deployed in various provinces across South Africa. The solution effectively copes with railway security risks caused by various physical environments such as coastal areas and CBDs, better safeguarding railways across the country. Besides, Huawei has innovatively introduced AI foundation models to improve solution reliability. This points out a new direction for railway perimeter security.