Image analysis – Applications – Personnel identification
Reexamination Certificate
1998-05-15
2002-07-16
Ahmed, Samir (Department: 2623)
Image analysis
Applications
Personnel identification
C340S005200
Reexamination Certificate
active
06421453
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
The present invention relates to methods and apparatus for user recognition (classification/identification/verification) to grant access or service to authorized users and, more particularly, to methods and apparatus for providing same employing gesture recognition, speaker recognition, and additional biometric and/or non-biometric features.
2. Discussion of Related Prior Art
Natural computing is a relatively new field which is becoming increasingly popular. In a natural computing system, a display screen, preferably a giant screen covering a wall, is located in front of a user. Conventional input devices (i.e., user interfaces), such as a mouse and keyboard, may be completely eliminated through the use of voice commands and/or gestures. That is, gestures such as pointing fulfill the pointing and clicking roles of the mouse while speech provides the command and control inputs.
Consequently, these relatively new user interfaces dramatically impact the transactions between users and computers. Remote transactions between a user and computer, similar to that between a user and television set via a remote control, may be realized. Accordingly, it would be desirable and highly advantageous to utilize gestures to not only provide command and control inputs to the computer, but also to recognize individuals attempting to utilize the computer in the first place in order to restrict access to authorized users.
In a natural computing environment, user recognition (classification, identification, and verification) is a paramount concern. To illustrate this, consider the following example. Three individuals are attempting to simultaneously interface with a computer that controls a truck manufacturing facility through either voice commands or gestures. The first individual is an unauthorized user, the second individual is a data entry person desiring to simply enter data of a non-urgent nature, and the third individual is a supervisor desiring to immediately initiate a power shutdown of a particular assembly line before a catastrophic event occurs (e.g., he has observed a severe stress crack in an automatic lift which is about to raise a truck off the ground for further assembly).
In such a situation, it would be desirable for the computer to classify, identify, and verify the individuals. Classification involves the differentiation of multiple individuals simultaneously attempting to interact with the system. Individuals must be differentiated so that each command provided to the computer is associated to a particular individual. This is because the same command (e.g., word, phrase, gesture) may have different meanings from one individual to the next. Next, the computer must be able to identify the individuals from their respective commands (or through an imaging system, etc.) without reception of an identity claim (i.e., indicia supplied by the individual to initially identify the individual). Then, the computer must verify that the individuals are indeed authorized to access the computer (prior to executing any offered commands). Further, in this example, it would be desirable to implement an extension of the identification task where the individuals attempting to interface with the computer are ranked so that a higher ranking individual (i.e., the supervisor) is allowed access over a lower ranked individual (i.e., the data entry person). If all these steps are properly performed, the computer will process the command from the supervisor to shutdown power before any commands from the data entry person are processed. Further, the computer will ignore any commands from the first individual (i.e., the unauthorized user). Thus, as the example illustrates, user recognition (classification, identification, and verification) is a significant factor in a natural computing environment.
Currently, there are several techniques and apparatus for recognizing an individual. They have been significantly implemented in systems which verify the identity of an individual requesting access to a service or facility in order to determine if in fact the individual is authorized to access the service or facility. In such situations, users typically have to write down, type or key in (e.g., on a keyboard) certain information in order to send an order, make a request, obtain a service, perform a transaction or transmit a message.
Verification or authentication of a customer prior to obtaining access to such services or facilities typically relies essentially on the customer's knowledge of passwords or personal identification numbers (PINs). However, such conventional user verification techniques present many drawbacks. First, passwords and pins selected by the user are usually based on some personal aspect of the user's life, such as, for example, their mother's maiden name or child's birthday. A seasoned perpetrator intent on committing fraud can usually decipher/determine user selected passwords and pins fairly easily. In the case where the password or PIN is provided to the user without his input, such measures are generally not reliable mainly because the user is usually unable to remember this information or because many users write the information down thus making the fraudulent perpetrator's job even easier. For instance, it is known that the many unwitting users actually write their PINs on the back of their ATM or smart card. Additionally, advances in technology have made it easier for a perpetrator to fraudulently obtain a password or PIN. For example, a perpetrator may view a transaction between a user and an ATM via binoculars or other enhanced viewing device (e.g., night vision goggles) in order to obtain the user's PIN. Similarly, an enhanced audio obtaining device (e.g., miniaturized audio amplifiers that resemble hearing aids) may used to fraudulently overhear and thus obtain a password.
Similarly, user verification techniques employing items such as keys, ID cards, and ID cards with embedded PINs also present many drawbacks. For example, such items may be stolen or lost and subsequently fraudulently used. This is especially true when a perpetrator obtains such items by stealing a wallet or pocketbook where other information may be contained therein informing the perpetrator where the key or card can be used (e.g., the location of the victim's bank, ATM, etc.). Additionally, such items are not easily utilized by children. Also, they are difficult to change periodically which may be desirable from a security point of view.
The shortcomings inherent with the above-discussed security measures have prompted an increasing interest in biometric security technology, i.e., verifying a person's identity by personal biological characteristics. Several biometric approaches are known, such as, for example, the recognition of voice print, facial bone structure, signature, face temperature infrared pattern, hand geometry, writing instrument velocity, writing instrument pressure, fingerprint, and retinal print, to name a few.
However, conventional biometric approaches also present drawbacks. For example, they are not 100% foolproof as several people (e.g., twins) can have the same voice print or same face. Additionally, conventional biometrics are not transferable and thus, for example, cannot be transferred to a friend to allow her access to a home. Also, they cannot be changed if a higher level of security is required. Furthermore, conventional biometrics are not always conveniently furnished such as in the case where weather conditions make it prohibitive to remove one's gloves in order to provide a fingerprint to gain access to a facility. Moreover, they are not constant, as a person can, for example, cut a finger (affecting their fingerprint) and age (affecting, e.g., their bone structure scan). These drawbacks not only impact recognition systems for accessing a service or facility but also affect the recognition functions required in a natural computing environment.
Accordingly, it would be desirable and highly a
Kanevsky Dimitri
Maes Stephane H.
Ahmed Samir
F. Chau & Associates LLP
International Business Machines - Corporation
LandOfFree
Apparatus and methods for user recognition employing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and methods for user recognition employing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and methods for user recognition employing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2867005