Hand Geometry at Purdue University Rec Center

This research was led by Eric Kukula.

Our group prepares students who have practical experience in applied biometrics - from system design through to implementation. One such example was the hand geometry study in the Recreation Center. Partnering with Recognition Systems, the Dean of Students Office, and the Division of Rec Sports, students have had exposure in implementing a data collection system in a live operational environment providing them with the experience of monitoring and testing a biometric system in a live environment. Following the evaluation of Recognition Systems' hand geometry readers at the Recreational Sports Center (RSC) during the spring 2005 semester, students' reactions and performance were promising. During the 2005-06 school year, we continued the study to gather more feedback from students on how they like using the device and evaluate the performance of the hand geometry system.

The purpose of the study was to determine:

  1. How students & faculty interact with the device
  2. Perceptions of students & faculty of hand geometry vs. the current ID system
  3. Performance of the hand geometry system

 

Location

The study took place in the front entrance area of the RSC. The image above shows setup of the hand geometry system before the renovations at the facility. The image to the right shows a user placing his hand on the device.

Hand Geometry Overview

In order to “recognize” your hand, the hand geometry system must first acquire an image. The first time you use the device, you enroll in the device. On each subsequent visit, you verify that you are in the system. During enrollment, each person is assigned a unique identification number or given an ID card. The enrollment procedure requires the new user to place their hand three separate times on a platen, to capture three images. The three enrollment images are “averaged” and used to create the template for the new user. Every time a person returns to use the device their hand is compared to their template.

During verification, a user will type their identification number or swipe an ID card and then place their hand on the reflective platen around the 5 guide pins. The most common hand geometry systems consist of a reflective platen, guide pins, a charged couple device (CCD) camera, an infrared light source, and an input mechanism (keypad or magnetic stripe reader). The hand is placed on the platen and aligned by five pressure sensitive guide pins ( Figure on the left). These guide pins minimize the variability of subsequent hand placements, by properly aligning the hand, and they serve as a liveliness detection function by requiring enough dexterity of the hand to squeeze the pins with a certain amount of pressure.

Once enough pressure is applied to each pin, the light source and CCD camera are triggered, and an image of the hand is captured. The resulting image is a silhouette of the hand ( Figure on the right). The image appears as a silhouette due to the design of the image capture process. The process consists of emitting an infrared light toward the hand in order for the CCD camera to capture the resulting light reflections. Portions of the reflective platen that remain uncovered by the hand appear as white areas in the image (reflected light), whereas dark areas represent the hand and appear as a black area (absorbed light).

Previous Work / Publications

Kukula*, E. P., & Elliott, S. J. (2005). Implementation of hand geometry at Purdue University's recreational center: an analysis of users perspectives and system performance, Proceedings of the 39th Annual International Carnahan Conference on Security Technology (ICCST) (pp. 83-89). Las Palmas de G. C., Spain.
Hernandez, C. A., Elliott, S. J., & Huddard, M. (2004). Implementation of Hand Geometry in Recreational Sports Center , Center for Education and Research in Information Assurance and Security conference, West Lafayette, IN.