A Hybrid Feature Based Mobile Forensics System

Due to the limited computing ability of current mobile computing platform, many intelligent applications are hard to be implemented, such as some complex pattern recognition algorithms. This paper proposes a kind of fast hybrid feature based progressive person recognition algorithm for forensics application which form the candidate set according to the biometric feature first and then conduct face recognition algorithm in the candidate set. A bionic pattern recognition based face recognition algorithm is also proposed. Experimental results show that the proposed algorithm has a satisfied recognition precision and low time complexity which could be conducted on mobile forensics devices with high efficiency.