The Ministry of Information Technology and Telecommunications is working on developing its own facial recognition system that can be used with video surveillance systems. The MoIT mentions that this system will be available soon.
The facial recognition system will be marketed by a startup that focuses on video surveillance. Furthermore, it will be made available and sold as a stand-alone FPGA-based prototype solution.
The project is being developed by PAF-Karachi Institute of Economics and Technology, and will cost around Rs 13.84 million. It is being developed under the name of ‘Design and Development of an FPGA-Based Multi-Scale Face Recognition System.’
According to sources familiar with the project, they mention that for security applications, using facial recognition tech is the best choice.
“With increasing security threats, the problem of invulnerable authentication systems is becoming more acute.”
Currently, to secure a facility, companies have traditionally relied on means and strategies such as smart cards, keys and passwords. In short, these security strategies corresponded to “what you have” or “what you know”.
The problem with these traditional systems of ensuring security was that “these systems could be easily fooled.” For example, you could forget the password or use the same password for multiple facilities, making it easier to get hacked. Even cards or keys can be stolen and then misused.
A safer approach to securing your facility would be to go with the use of biometrics. With such security measures, you have strategies that correspond to “what you are” or “what you exhibit
Facial recognition technology seems a more suitable and natural choice as compared to the other available biometric security measures, such as speech, iris, fingerprints, hand geometry and gait. It is pocket friendly, requires minimal user cooperation and is also non-intrusive. It is also relatively cheaper to implement.
“This project focuses on design and development of a high speed FPGA-based multi-scale face recognition system using Linear Binary Pattern (LBP) features”- sources.
How Will the Facial Recognition Work?
The facial recognition system being developed by MoITT is giving special emphasis to algorithm design that can be efficiently included in the hardware. The technology will rely on the face’s texture information being extracted by LBP features. The texture information would become part of wavelet decomposition, and include sub-bands that could prove useful for classification purposes.
To put it simply, these sub-bands could easily identify ‘discriminating’ features on a person and help anyone from researchers to law-enforcement authorities to find who they are looking for quicker than ever.