Personal identification system using multiple parameters having low cross-correlation

• US 6,205,233 B1
• Filed: 09/16/1998
• Issued: 03/20/2001
• Est. Priority Date: 09/16/1997
• Status: Expired due to Term

• Current Assignee: Invisitech Co., Ltd.
• Original Assignee: Invisitech Co., Ltd.
• Inventors: Hill, Lawrence W., Morley, Richard E.
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): NAKHJAVAN, SHERVIN K
• Application Number:
  US09/154,325
• Time in Patent Office: 916 Days
• Field of Search: 382/100, 382/103, 382/115, 382/116, 382/286, 382/276, 382/312, 382/291, 340/825.31, 348/152, 348/153, 348/154, 348/155
• US Class Current: 382/103
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16

Abstract

A personal identification system for use in security and various other types of pattern matching applications provides accurate recognition of a known person. This system uses a large quantity of course but inexpensive sensors to create a profile of a person seeking to enter the controlled space or to take other action. These sensors measure different statistically independent characteristics of the person, such as height, weight, stride, voice timbre, and time to perform various functions. The measurements so obtained may individually be of very low resolution. They are used to search a data base of patterns related to admissible individuals. Active agent methods are used to support learning and adaptation of the sensor set to address sensor drift and individual characteristic changes over time. A combination of many parameters will identify a person with very low probability of both false negative and false positive errors. The resulting low sensor resolution required allows such a system may be implemented at very low cost in comparison to other biometric systems.

Claims

1. A detection system for analyzing personal behavioral traits and matching such traits with known trait profiles, comprising:

a plurality of sensors, each monitoring a respective condition, having sensor output signals, each said output signal indicative of a subject criteria;

a sensor reading repository connected to said plurality of sensors for storing a plurality of said sensor output signals;

a profile generator, connected to said sensor reading repository, receiving said sensor output signals and generating a subject profile from said plurality of sensor output signals;

an accepted profile repository for storing a plurality of accepted subject profiles;

a profile verifier for receiving said subject profiles for comparing said subject profile to said plurality of accepted profiles, and for generating profile output signals in response to said comparing.

2. The system as in claim 1 further comprising:

an output device interface for receiving said output signals;

at least one output device connected to said profile verifier and adapted to respond to said output signals.

3. The system as in claim 1 further comprising:

a graphical user interface connected to said profile verifier and adapted to respond to said profile verifier by generating output display signals.

4. The system as in claim 1 wherein said sensor reading repository further comprises a time stamp associated with each of said plurality of sensor output signals.

5. The system as in claim 1 wherein said profile verifier is operative to update said accepted profile repository in response to said comparison of said subject profile with said plurality of accepted subject profiles.

6. The system as in claim 5 wherein said update of said accepted profile repository is determined as a function of partial matching of said subject profile and said plurality of accepted profiles.

7. The system as in claim 1 further comprising a plurality of bins, wherein said sensor output signals indicative of said subject criteria are classified in said plurality of bins and said bins denote a predetermined range of said sensor output signals.

8. The system as in claim 7 wherein said each subject criteria is directed towards a separate quantifiable trait of said subject.

9. The system as in claim 8 wherein said sensor is an infrared beam.

10. The system as in claim 8 wherein said sensor has varied electric resistance in response to pressure.

11. The system as in claim 8 wherein said sensor reads magnetic pulses.

12. A method of matching patterns indicative of personal characteristic traits, comprising the steps of:

reading a plurality of sensors each monitoring a respective condition;

determining at least one criteria value from each said reading by a processor in communication with said plurality of sensors;

storing said at least one criteria value in a sensor reading database associated with said processor;

building a subject profile by said processor from said at least one criteria value;

storing a plurality of accepted subject profiles in an accepted profile database; and

comparing said subject profile to said plurality of accepted subject profiles in said accepted profile database.

13. The method as in claim 12 wherein said reading step further comprises calculating a time stamp and said storing step further comprises storing said time stamp in association with said reading.

14. The method as in claim 13 wherein said step of determining said at least one criteria value further comprises calculating a bin value, wherein said bin value denotes a subrange of values for said criteria.

15. The method as in claim 14 further comprising calculating a match value from said comparing of said subject profile and said accepted profile database.

16. The method as in claim 15 further comprising generating an output signal from said match value.

17. The method as in claim 16 further comprising an output device interface, said output device interface receiving said output signal and selectively manipulating an output device according to said output signal.

18. The method as in claim 15 wherein said subject profile is comprised of a series of said bin values for said at least one criteria value.

19. The method as in claim 18 wherein said step of determining said match value further comprises the steps of comparing said series of bin values to said series of accepted subject profiles in said accepted profile database.

20. The method as in claim 19 wherein said comparing of bin values occurs for each of said at least one criteria.

21. The method as in claim 20 wherein said comparing of bin values further comprises a tolerance threshold wherein said determining of said match value is computed as a function of said bin values differing by said tolerance threshold.

22. The method of claim 21 wherein said storing of said subject profile further includes updating a previously stored subject profile as a function of said readings stored in said sensor reading database, said at least one criteria value, and said previously stored subject profile.

23. The method as in claim 22 wherein said at least one criteria include height, weight, stride, pace, and speed.

24. The method as in claim 22 wherein said at least one criteria include handwriting height, width, pressure, duration, and saturation.

25. A method of matching patterns indicative of personal characteristic traits, comprising the steps of:

reading at least one sensor and calculating a time stamp for said reading;

determining at least one criteria value including handwriting height, width, pressure, duration and saturation from said reading by a processor in communication with said at least one sensor by calculating a bin value, wherein said bin value denotes a subrange of values for said criteria;

storing said at least one criteria value and said time stamp in a sensor reading database associated with said processor;

building a subject profile comprised of a series of said bin values for said at least one criteria value by said processor from said at least one criteria value;

storing an accepted subject profile by updating a previously stored accepted subject profile as a function of said readings stored in said sensor reading database, said at least one criteria value, and a previously stored accepted subject profile in a plurality of accepted subject profiles in an accepted profile database;

comparing said subject profile to said plurality of accepted subject profiles in said accepted profile database; and

calculating a match value from said subject profile and said accepted profile database, comprising comparing said series of bin values for each of said at least one criteria to said series of accepted subject profiles in said accepted profile database, said calculating said match value is computed as a function of said bin values differing from said accepted profile database by a tolerance threshold.