Image analysis – Applications – Personnel identification
Reexamination Certificate
2000-05-19
2002-12-17
Mancuso, Joseph (Department: 2623)
Image analysis
Applications
Personnel identification
C382S118000, C382S116000, C340S005800
Reexamination Certificate
active
06496595
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to automated systems for permitting authorized persons to access secured buildings or other areas while preventing such access by unauthorized persons. More particularly, the invention relates to an access control system which uses biometric parameters to identify authorized persons and to permit access by the authorized persons.
2. Description of the Related Art
The invention relates to restricting access into a restricted area and thus the term “access”, as used herein, refers to physical entry into a building, or other restricted area. However, the restricted area can be the exterior of a building or the like and thus the invention can be applied to controlling entry or exit out of a building or other area. There have always been situations in which it was desirable to restrict access to certain physical areas to a select person or group of people. Such restricted access has been accomplished by fences, walls, locks and other barriers. However, even the use of barriers has not prevented unauthorized access. Accordingly, it has been necessary in many instances to provide human surveillance in the form of a security guard at an entrance to an area or through the use of video cameras or the like to transmit images to a security guard at a remote or centralized location. Of course, the use of security guards and video cameras can become expensive and is only as reliable as the particular guard and his state of alertness at any particular time.
The complexities of modern society have only served to increase the need for access control. For example, many government agencies and contractors work on matters that are of a confidential or even “top secret” nature. In fact, most workplaces, such as offices, warehouses, and even retail stores in some instances, have a need to implement access control to prevent the theft of intellectual property and/or goods.
In response to the need for access control, many government agencies and businesses have issued identification cards to their employees and other authorized personnel. Often, the identification card includes a picture of the authorized person. However, such an identification card still requires a security guard or other personnel for verification. Also, such cards are easily forged by replacing the picture with that of an unauthorized person. To overcome these limitations, it is known to provide the employee with a personal identification number (PIN) or other identifying code. The identifying code can be encoded in a magnetic strip or the like in a security card and read by an automated reader at an entrance to grant access only to persons having the card. Alternatively, the code can be entered by the user on a keypad to gain access. However, the use of identifying codes also has drawbacks in access control applications. In particular, the identification card can be stolen or the user can be forced under duress to reveal their code. In such cases, unauthorized possessors of the card or code can gain access to a restricted area.
The use of biometrics has been proposed as a solution to the limitations noted above. Generally, the term “biometrics” refers to the study of measurable biological characteristics, i.e. biometric parameters, of a living being. In the context of security, “biometrics” refers to techniques that rely on a unique, measurable characteristic of a living being for automatically recognizing or verifying identity. Examples of biometric parameters are facial data, retinal data, fingerprint data, speech data, and the like.
Generally, biometric systems operate in the following manner. First, a system captures a sample of at least one biometric parameter during an “enrollment” process. The parameters are then converted by the system into a mathematical code, i.e., data, that is stored as the biometric template representing measured biometric parameters for that person. There may also be a way to correlate the template to the person. For example, a personal identification number (PIN) can be given to an enrolled user, which is entered to access the template. A new biometric sample is then taken and compared to the template or to a group of templates. If one of the templates and the new sample match, the person is recognized as authorized. Central to a biometric system is the “engine” which processes the biometric data in accordance with various algorithms or artificial neural networks.
Some biometric systems use “identification” methods and some use “verification” methods. In identification systems, a sample is presented to the biometric system and the system then attempts to find out who the sample belongs to by comparing the sample with a plurality of templates obtained through enrollment. Verification systems on the other hand perform a one-to-one process where the biometric system is seeking to verify identity. A single biometric sample is matched against a single template obtained during enrollment. If the two match, the system effectively confirms that the person actually is who he presents himself to be. The key difference between these two approaches centers on the logic addressed by the biometric system and how these fit within a given application. Identification systems decide who the person is and can check whether more than one matching biometric template exists. Accordingly, identification systems can deny access to an individual who is attempting to pass himself off with more than one identity. Verification, on the other hand only decides if the person is who he says he is. Accordingly, identification systems are more versatile and powerful. However, verification systems generally require less processing horsepower and thus are commonly used.
In typical biometric access control systems, biometric sensors are placed proximate entrances and are linked to a central computer having biometric “enrollment” data, i.e., templates, representing biometric parameters of authorized users collected from a central enrollment station. If the biometric parameter collected at the entrance matches a template stored in the central computer, access is granted. However, conventional biometric systems have several limitations. In particular, the sensing accuracy of biometric parameters, such as facial parameters, retinal parameters, and the like is highly dependent on the environment in which the parameters are sensed. For example, the lighting intensity, angle and color will affect sensing of biometric parameters. Accordingly, the ability to reliably match data representing parameters collected at an entrance with data of templates collected during enrollment is limited, especially when the entrance is an external entrance where the weather, season, and time of day will affect lighting significantly. Also, sensing biometric parameters, converting the parameters to data, communicating the data to a central computer, and comparing the data parameters with templates of enrollment data in the central computer is relatively time consuming, even with modern high speed computers and communication links. Accordingly, such systems present significant inconveniences to the authorized persons through improperly rejected access and time delays prior to granting access.
The use of less environmentally affected biometric parameters, such as fingerprint parameters can alleviate some of the problems noted above. However, sensing such parameters requires that the person desiring access make physical contact with a sensor, such as a fingerprint scanner, upon each entry request. Also, such parameters are easily affected by skin debris, and skin blemishes. Therefore, the person desiring access may have to wipe off their finger, not to mention remove any gloves, prior to submitting to the physical contact with the sensor. The match tolerance between a template and data corresponding to a measured parameter can be increased to minimize lockout of authorized users. However, such an approach inherently reduces the accuracy and effectiveness of the system by increasing the li
Puchek Daniel R.
Tumey David M.
Xu Tianning
Bali Vikkram
Kaufman Marc S.
Mancuso Joseph
Nextgenid, Ltd.
Nixon & Peabody LLP
LandOfFree
Distributed biometric access control apparatus and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Distributed biometric access control apparatus and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Distributed biometric access control apparatus and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2986916