NUST Institutional Repository

Deep-Learning Based Laser Fencing

Show simple item record

dc.contributor.author Khan, Ahmed Umair
dc.contributor.author Khan, Sarosh Ahmad
dc.contributor.author Muhammad, Usman Aksi
dc.contributor.author Malik, Muhammad Awais
dc.contributor.author Supervised by Dr. Fahim Arif
dc.date.accessioned 2025-02-10T08:25:14Z
dc.date.available 2025-02-10T08:25:14Z
dc.date.issued 2023-05
dc.identifier.other PCS-450
dc.identifier.uri http://10.250.8.41:8080/xmlui/handle/123456789/49607
dc.description.abstract As security threats have become more sophisticated and traditional physical barriers have become less effective, there has been a growing demand for laser fencing, especially at the borders. It is a versatile and effective security solution that can be adopted to meet a wide range of security needs. This technology is being used around the world for border security, critical infrastructure protection, military bases, prisons, wildlife conservation, and residential and commercial security. Its advanced sensors and deep-learning algorithms make it a highly reliable solution for detecting and preventing intrusions. Its most common applications involve its deployment at such places at the borders where human presence is practically impossible due to difficult terrain or harsh weather conditions. Features like enhanced security, deterrence to criminal activity, cost-effectiveness, scalability, real-time monitoring, and reduced personnel requirements have made this technology an attractive option for border security agencies around the globe. This paper proposes a solution called Deep-Learning Based Laser Fence that comprises poles fitted with lasers and sensors to detect any intrusion between them. The poles are placed hundreds of meters apart. The system consists of a Transmitter and a Receiver unit, which are constantly talking to each other through data sharing. If an intruder “breaks” the laser beam; an alarm is triggered. Whenever an intrusion is attempted across a particular perimeter, the communication, and the data sharing between the two poles gets disrupted and a pulse is sent to Command Post - the C&C Platform - over the wired communication network. The system constantly evolves using Machine Learning to reduce false positives in case of non-human intrusion and become smarter. Hence, assisting with threat analysis of that perimeter. The build-in algorithm can detect if it is a human or an animal. Even a crawling intruder is also detectable with this software. The solution is designed for use in harsh environments and provides high detection accuracy in all weather conditions. It uses the latest technology which increases detection sensitivity and reduces nuisance alarms. All signals are digitally processed - with proprietary algorithms - which gives maximum detection performance with an extremely low false alarm rate. This ensures a very high security standard is achieved. en_US
dc.language.iso en en_US
dc.publisher MCS en_US
dc.title Deep-Learning Based Laser Fencing en_US
dc.type Project Report en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse

My Account