An Application of Biometric Technology: Hand Vien/ Geometry Recognition
Overview of Previous Article
Our last article reviewed fingerprint biometrics, and the steps that are involved in the fingerprint recognition process: Image acquisition; the location and determination of unique features in the fingerprint; template creation and template matching.
Also, the different types of fingerprint capture technologies were also reviewed:
(1) Optical technology; (2) Silicon technology; and (3) Ultrasound technology.
This article focuses upon yet another widely used biometric technology—hand geometry recognition, and is divided into the following sections: (1) Key advantages of hand geometry recognition; (2) How hand geometry recognition works; and (3) Applications of hand geometry recognition.
Key Advantages of Hand Geometry Recognition
This technology possesses a number of key advantages. First, hand geometry recognition, along with fingerprint recognition, has been around the longest, and as a result, it has certainly proved itself to be a viable technology. Second, it is deemed to be one of the easiest to use and administer of all of the biometric technologies that are available today. Third, hand geometry recognition can work in the harshest of environments, both internal and external. In terms of an internal environment, the technology can tolerate a fair amount of rough usage from end users, especially in large factory, warehouse, and retail settings. With respect to an external environment, the technology has proven to work in the most extreme of atmospheric climates, ranging from very hot to very cold. Fourth, hand geometry recognition is the least susceptible to privacy rights issues-primarily because of its simple enrollment and verification procedures. Fifth, the hand is a stable biometric whose physical characteristics are not susceptible to major biological changes (except for conditions of arthritis, swelling, or deep cuts), thus making hand geometry recognition a very reliable technology.
There is only one primary vendor for hand geometry recognition technology, and that is Recognition Systems, Inc. They produce hand geometry scanners for two broad market segments: (1) Physical access entry; and (2) Time and attendance. With respect to physical access entry, a widely used hand geometry scanner is called the HandKey II.
How Hand Geometry Recognition Works
The hand geometry scanner looks for unique features in the structure of the hand. These unique features include the finger thickness, length, and width, the distances between finger joints, the hand’s overall bone structure, etc. It should be noted here that with iris and fingerprint recognition, the primary goal is to look for extremely distinctive features. However, this is not the case with hand geometry recognition, as it is looking for moderately unique features. Thus, hand geometry recognition would not be the biometric tool of choice for high security applications or identification purposes where iris recognition or fingerprint recognition would be, respectively.
The user first places his or her hand onto a platen. This platen consists of 5 pegs which help the user position their fingers properly in order to insure quality enrollment and verification templates. The hand geometry scanner consists of a charged couple device camera (CCD), as well as various reflectors and mirrors in order to capture various black and white pictures of the hand. Two basic types of pictures of the hand are captured: (1) An image of the top of the hand; and (2) An image of the side of the hand.
In the enrollment phase, the user is prompted by the hand geometry scanner to place their hand on the platen three different times, so that three images can be captured and then averaged. The resulting image forms the basis for the enrollment template, which is then stored in the database of the hand geometry scanner. The enrollment phase can be accomplished in just five seconds.
In the verification phase, the user is prompted to place their hand only once on the platen. An image is captured, and forms the basis for the verification template. The verification template is compared against the enrollment template, in the exact same fashion as fingerprint recognition. The verification phase can be accomplished in just under one second.
In the enrollment and verification phases, the hand geometry scanner takes 96 measurements of the hand. The enrollment and verification templates are only 9 bytes.
Applications of Hand Geometry Recognition
There are numerous types of applications for which hand geometry recognition is utilized. The most recognized use for this technology is in physical access entry applications, because the system is user-friendly to configure. In fact, this was the first application that this technology was used for when it first came out onto the market. Another application gaining popularity for hand geometry recognition is time and attendance. There are hand geometry scanners today that are just designed for this, as discussed previously. The advantage is that the use of timecards, identification badges, and Social Security numbers is eliminated. Also, the costly problem of “buddy punching” (associated with time clocks) is non-existent. Hand geometry recognition is also used for point of sale applications. Examples of this include subsidized school lunch programs and luxury hotels. All of these involve the use of the hand geometry scanner to deduct or debit money from a user’s fund account when a purchase is made. Hand geometry recognition is also utilized in the Immigration and Naturalization Service Passenger Accelerated Service System (INSPASS). With this system, frequent international business travelers can simply use their hand geometry to enter the United States, rather than waiting in long immigration lines at the airport. Finally, hand geometry recognition is making a presence in the financial sector; a number of banks (in particular the Bank of America and the Nevada State Bank) are planning to adopt this technology to give customers easier and timely access to their safe deposit vaults.