Efficient Face Recognition Using FPGA and Semantic Features for Security Controls

Facial features are essential to biometric authentication. Nowadays an effective face recognition system is mostly composed of feature’s transformation, geometric analysis, and recursive training to resist illegal intrusion. However, the design challenge always arises from the dilemma of lower computing resource usages, power consumption, and real-time performance under rigorous operations of construction sites. Here a chip-based face recognition is proposed by using semantic features to achieve a minimal dataset and high processing speed without any complicated formulization processes. Our experimental results on single FPGA chip demonstrate the feasibility to realize a miniature and efficient security control system for construction facilities in the future.

[1]  Li Hui,et al.  An algorithm of skin detection based on texture , 2011, 2011 4th International Congress on Image and Signal Processing.

[2]  Ryan Kastner,et al.  Fpga-based face detection system using Haar classifiers , 2009, FPGA '09.

[3]  Mohammed Bahoura,et al.  FPGA implementation of a feature extraction technique based on Fourier transform , 2012, 2012 24th International Conference on Microelectronics (ICM).

[4]  Ngai Ming Kwok,et al.  On-chip real-time feature extraction using semantic annotations for object recognition , 2014, Journal of Real-Time Image Processing.

[5]  Kemal Hajdarevic,et al.  Face recognition implementation for client server mobile application using PCA , 2012, Proceedings of the ITI 2012 34th International Conference on Information Technology Interfaces.

[6]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Q. P. Ha,et al.  Color tracking for multiple robot control using a system-on-programmable-chip , 2011 .

[8]  Iraj Hosseini,et al.  Low Cost Fpga-Based Highly Accurate Face Recognition System using Combined Wavelets with Subspace Methods , 2006, 2006 International Conference on Image Processing.

[9]  Fei Zuo,et al.  Real-time embedded face recognition for smart home , 2001 .

[10]  David Blaauw,et al.  A 23-mW Face Recognition Processor with Mostly-Read 5T Memory in 40-nm CMOS , 2017, IEEE Journal of Solid-State Circuits.

[11]  Fang Zhong,et al.  Parallel architecture for PCA image feature detection using FPGA , 2008, 2008 Canadian Conference on Electrical and Computer Engineering.