Dynamic Footprint-based Person Recognition Method using Hidden Markov Model

---

• Introduction Images

 

• Research Goal

• Calibration of Force Sensing Resistors (FSR)
• Finding an algorithm which can recognize the person using static footprint and dynamic footprint
• Using the Hidden Markov Model for the dynamic footprint
• Fusion of the results of the static footprint and the dynamic footprint using the Neural Network and Optimizing the network
  
  

• Research Statistics

• Research resource: 2 people
• Research term: 2004.9 ~ 2004.12 (4 months)
• My role responsible of this research: 50%
• Research output: An algorithm for person recognition using static footprint and dynamic footprint using hidden markov model and neural networks, Foot''Rec (A total implementation of the algorithm and interfacing the foot sensor) (10,000 lines of MFC code)
  
  

• Techniques used for this research

• Force Sensing Resistors (FSR): FSRs are also known as "Pressure Sensing", "Force Sensitive Resistors", etc. Force Sensing Resistors (FSRs) are a type of resistor whose resistance changes when a force or pressure is applied. The resistance is inversely proprotional to the force applied, i.e. the resistance decreases as the force gets increases.
• Foot Analyzer: An implementation of the FSR for foot pressure measuring. This is a product from a venture company in Korea, the resolution of the sensor is 80 * 40, and each cell gives us data of the pressure from the value 0 to 255.
• Hidden Markov Model (HMM): A statistical model where the system being modeled is assumed to be a Markov process with unknown parameters, and the challenge is to determine the hidden parameters from the observable parameters. The extracted model parameters can then be used to perform further analysis, for example for pattern recognition applications. A HMM can be considered as the simplest dynamic Bayesian network.
• Neural Network: A computing paradigm that is loosely modeled after cortical structures of the brain. It consists of interconnected processing elements called neurons that work together to produce an output function. The output of a neural network relies on the cooperation of the individual neurons within the network to operate. Processing of information by neural networks is often done in parallel rather than in series (or sequentially). Since it relies on its member neurons collectively to perform its function, a unique property of a neural network is that it can still perform its overall function even if some of the neurons are not functioning. That is, they are very robust to error or failure (i.e., fault tolerant).
  
  

• My contribution to the research

• Calibration of Force Sensing Resistors (FSR): For high performance in footprint recognizion, FSR should give precise data which is a numerical value implying pressure. As FSR is a sensitive device, I considered the relations between location and weight and designed an efficient method of calibration.
• Hidden Markov Model (HMM) Application: To recognize who he is by footprints, we need pattern matching algorithm using input footprints and the data base. I designed a numerical learning system and implemented the efficient algorithm using Hidden Markov Model.
• Algorithm for static and dynamic footprint recognition: The static and dynamic footprints has distinctive characteristics, such as foot-shape and movement of weight center. Although each has a good recognizion on footprints, I could get better results When I mixed static and dynamic footprints properly.
  
  

• Results

• Calibration of Foot Analyzer (using FSR)
• FootRec (controls the FSR, and learns data from the sensor, and stores the data, and finally matches the data to identify a person)
• A static footprint recognition algorithm
• A dynamic footprint recognition algorithm using the Hidden Markov Model
• A neural network which combines the result of the static footprint recognition results and the dynamic footprint recognition results
  
  

• References

• A. Jain, R. Bolle, and S. Pankanti, Biometrics: Personal Identification in Networked Society, KAP press, 1999.
• A. J. Thurston, “Giovanni borelli and the study of human movement: an historical review, “ The Australian and New Zealand journal of surgery, Vol. 69, No. 4, pp. 276-299, 1999.
• D. D. Zhang, Automated Biometrics – Technologies and Systems, KAP press, 2000.
• D. J. Kim and Z. Bien, “A Personalized Facial Expression Recognition System using Fuzzy Neural Networks with a Novel Feature Selection Scheme,” Proceedings of FUZZ-IEEE, pp. 908-913, 2003.
• J. W. Jung and Z. Bien, “Dynamic Footprint-based User Identification system: A Novel Biometric Technique for Ubiquitous Computing Environment,” Doctoral Thesis of KAIST, 2004.
• K. B. Kennedy, “Uniqueness of bare feet and its use as a possible means of identification, “ Forensic Science International, Vol. 82, Issue 1, pp. 81-87, 1996.
• K. Nakajima, Y. Mizukami, K. Tanaka, and T. Tamura, “A New Biometrics Using Footprint,” Transactions Institute of Electrical Engineers of Japan D, Vol. 121, No. 7, pp. 770-776, 2001.
• K. Nakajima, Y. Mizukami, K. Tanaka, and T. Tamura, “Foot-Based Personal Recognition,” IEEE Tr. On Biomedical Engineering, Vol. 47, No. 11, 2000.
• M. Razeghi and M. E. Batt, “Foot type classification: a critical review of current methods,” Gait and Posture, Vol. 15, pp. 282-291, 2002.
• M. Weiser, “The Computer for the 21st Century,” Scientific Am., Vol. 265, No. 3, pp. 66-76, Sept. 1991.
• P. R. Cavanagh, E. Morga, A. J. M. Boulton, M. J. Young, K. T. Deffner and S. E. Pammer, “The relationship of static foot structure to dynamic foot function,” J.Biomechanics, Vol. 30, No. 3, p. 243-250, 1997.
• R. B. Kennedy, I. S. Pressman, S. Chen. P. H. Petersen, A. E. Pressman, “Statistical Analysis of Barefoot Impressions,” Journal of Forensic Science, Vol. 48, No. 1, pp. 55-63, 2003.
• S. Liu and M. Silverman, “A practical guide to biometric security technology,” IEEE IT Pro, pp.27-32, Jan./Feb., 2001.
• S. Nanavati, M.Thieme, and R. Nanavati, Bioimetrics: Identify Verification in a Networked World, WILEY press, 2002.
• T. Y. Shiang, S. H. Lee, S. J. Lee, and W. C. Chu, “Evaluating different footprint parameters as a predictor of arch height, “ IEEE Engineering in Medicine and Biology Magazine, Vol. 17, Issue 6, 1998.