Secure Grip

The main research question addressed in this project is whether an image of the pressure pattern exerted while holding an object can be used to reliably authenticate or identify a person. This type of biometric recognition is called grip-pattern recognition. It is a useful instrument to secure (1) the legal use of devices and (2) the legal access to locations. The research focuses on the design, implementation and evaluation of prototype grip-pattern recognition systems for two applications: (1) A smart gun, intended for use by the police, in which grip-pattern recognition ensures that the weapon can only be fired by its rightful user. (2) An intelligent door handle, equipped with grip-pattern recognition, capable of recognizing who enters or leaves a room. The prototypes will be evaluated under realistic conditions of use.

The project involves three distinct disciplinary areas: (1) Sensors, concerning the design of pressure sensor grids that can be applied to more or less arbitrary three-dimensional structures of various materials. (2) Algorithms, concerning the design and implementation of biometric recognition algorithms for authentication or identification based on two-dimensional pressure patterns. (3) Security architectures, concerning aspects of, e.g., security, template management, scalability and maintenance. The sensors will be developed, using state-of-the art technology.

The research and engineering challenges are: (1) Development of suitable sensor arrays that can be mounted on the three-dimensional structures of the grip of the police gun and the door handle. (2) Design of recognition algorithms achieving an extremely low probability of not recognizing the rightful user from a limited set of training data. (3) Development and verification of a security architecture.

Full project proposal pdf

Project management Team

• Dr. ir. Raymond Veldhuis (UT, SAS)
• Prof. Dr. Pieter Hartel (UT, DIES)
• Dr. ir. Askar Bazen (UT, SAS)
• Dr. Sandro Etalle (UT, DIES)

PhD students

• MSc. Xiaoxing Shang (UT, SAS)
• MSc. Ileana Balazs (UT, DIES)

Technician

• Wessel Ganzevoort (UT, SAS)

Publications of project SecureGrip members

• 2004

  [ Vel03 ] R. N. J. Veldhuis and A. M. Bazen and J. Kauffman and P. H. Hartel
  "Biometric verification based on grip-pattern recognition (invited paper)" ,
  IS&T/SPIE 16th Annual Symp. on Electronic Imaging - Security, Steganography, and Watermarking of Multimedia Contents, vol. 5306, E. J. Delp III and P. W. Wong (eds.) , published by SPIE -- The Int. Society for Optical Engineering, Washington, held in San Jose, California, Jan. , 2004, pp. 634-641
  Download Univ. Twente Technical Report
  Download html
  ISBN 0-8194-5209-2
  Abstract
  This paper describes the design, implementation and evaluation of a user-verification system for a smart gun, which is based on grip-pattern recognition. An existing pressure sensor consisting of an array of 44 x 44 piezoresistive elements is used to measure the grip pattern. An interface has been developed to acquire pressure images from the sensor. The values of the pixels in the pressure-pattern images are used as inputs for a verification algorithm, which is currently implemented in software on a PC. The verification algorithm is based on a likelihoodratio classifier for Gaussian probability densities. First results indicate that it is feasible to use grip-pattern recognition for biometric verification.
  
  
  

• 2002

  [ Hen02 ] N. J. Henderson and N. M. White and R. N. J. Veldhuis and P. H. Hartel and C. H. Slump
  "Sensing pressure for authentication" ,
  3rd IEEE Benelux Signal Processing Symp. (SPS), published by Unknown Publisher, held in Leuven, Belgium, Mar. , 2002, pp. 241-244
  Download Univ. Twente Technical Report
  ISBN not assigned
  Abstract
  The use of signals resulting from tapping a rhythm on a pressure sensor is explored for authentication. The features used for authentication can be divided into rhythm and waveform features. This paper studies the use of waveform features. A verification scheme based on prototype waveforms is presented. The scheme is tested on experimental data. Based on waveform-only information, an Equal Error Rate of 7.7% is achieved with an indication of further room for improvement. Suggestions for data collection and training are presented.