Video surveillance: MoIT developing facial recognition system

Ministry of Information Technology and Telecommunication (MoIT) is in process of developing its own facial recognition system that can be used with video surveillance systems and would be available soon. Official sources at Ignite (formerly known as National ICT R& D Fund) on Sunday said it will not only add value to international research scenario but will also carry significant contribution to the society.
The system will also be available as a stand-alone FPGA-based prototype solution that can be marketable through a start up company in domain of video surveillance. The project is being developed by PAF-Karachi Institute of Economics and Technology under the name "Design and Development of an FPGA-Based Multi-Scale Face Recognition System' which would cost Rs 13.84 million.
The sources said for security applications, by far, using facial tech is the best choice. "Video surveillance is an efficient way of securing a facility. It said with increasing security threats; the problem of invulnerable authentication systems is becoming more acute. Traditional means of securing a facility essentially depend on strategies corresponding to "what you have" or "what you know", for example smart cards, keys and passwords.
These systems, however, can easily be fooled, the sources said and added passwords for example, are difficult to remember and therefore people tend to use same password for multiple facilities making it more susceptible to hacking. Similarly cards and keys can easily be stolen or forged. A more inalienable approach is, therefore, to go for strategies corresponding to "what you are" or "what you exhibit" ie biometrics. Among the other available biometrics, such as speech, iris, fingerprints, hand geometry and gait, face seems to be the most natural choice. It is non-intrusive, requires a minimum of user cooperation and is cheap to implement.
The sources said this project focuses on design and development of a high speed FPGA-based multi-scale face recognition system using Linear Binary Pattern (LBP) features. Special emphasis is being given to algorithm design which can be efficiently mapped in Hardware. The LBP features primarily extract texture information of a face image.
Wavelet decomposition of these features result in sub-bands encompassing low and high frequency components which carry useful information for classification, however some sub-bands are more significant than others and an intelligent selection of these discriminant sub-bands is likely to increase overall performance of the face recognition system. The sources said the project is, therefore, aimed to identify discriminant sub-bands for efficient and robust face recognition.