Systems and algorithms for stateless biometric recognition

• US 8,549,319 B2
• Filed: 01/18/2012
• Issued: 10/01/2013
• Est. Priority Date: 09/01/2005
• Status: Active Grant

• Current Assignee: Memphis Technologies Incorporated
• Original Assignee: Memphis Technologies Incorporated
• Inventors: Shachar, Yair, Brikman, Josef, Menczel, Yaron
• Primary Examiner(s): Arani, Taghi
• Assistant Examiner(s): Jeudy, Josnel
• Application Number:
  US13/352,443
• Publication Number:
  US 20130015952A1
• Time in Patent Office: 622 Days
• Field of Search: 713182-186, 382115-119, 382/124, 340 581- 584, 340/5.52
• US Class Current: 713/186
• CPC Class Codes: G05B 1/03 for comparing digital signals
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

A system and algorithms to authenticate a person where a system only has some standard personal text data about the person, and cannot have a real biometric template obtained using an enrollment procedure. The authentication allows access to restricted resources by the person. This method is especially useful when it is used as an auxiliary authentication service with other methods such as password or Callback that dramatically lower the chances for an imposter.

Claims

1. A method for classifying a user comprising:

receiving a biometric input from said user;

extracting at least one biometric-based parameter from said biometric input,

wherein said biometric-based parameter comprises at least one of age, gender, ethnical or geographical origin, pronunciation, emotional state of the user, credibility level, or level of alcohol or other materials in blood;

comparing said at least one biometric-based parameter to a known user parameter to provide a comparison,

provided that said known user parameter does not contain a biometric signature of said user;

classifying said user based on said comparison to provide a classification; and

selecting an action based upon said classification.

2. The method of claim 1, wherein the extracting at least one biometric-based parameter results in a probability function that defines one or more estimated probabilities of said biometric-based parameter to match a set of ranges of values.

3. A method for authenticate a user in credit card accounting system comprising:

receiving a biometric input from said user;

extracting at least one biometric-based parameter from said biometric input,

wherein said biometric-based parameter comprises at least one of age, gender, ethnical or geographical origin, pronunciation, emotional state of the user, credibility level, or level of alcohol or other materials in blood;

comparing said at least one biometric-based parameter to a known user parameter to provide a comparison,

provided that said known user parameter does not contain a biometric signature of said user; and

authenticating said user based on said comparison.

4. The method of claim 3, wherein the extracting at least one biometric-based parameter results in a probability function that defines one or more estimated probabilities of said biometric-based parameter to match a set of ranges of values.

5. A system for classifying a user comprising;

at least one data terminal to receive biometric input of said user;

at least one extraction unit to extract at least one biometric-based parameter from said biometric input,

wherein said biometric-based parameter comprises at least one of age, gender, ethnical or geographical origin, pronunciation, emotional state of the user, credibility level, or level of alcohol or other materials in blood;

comparing said at least one biometric-based parameter to a known user parameter to provide a comparison,

provided that said known user parameter does not contain a biometric signature of said user;

classifying said user based on said comparison to provide a classification; and

selecting an action based upon said classification.