Methods and apparatus for comparison

• US 8,983,153 B2
• Filed: 10/19/2009
• Issued: 03/17/2015
• Est. Priority Date: 10/17/2008
• Status: Active Grant

• Current Assignee: Forensic Science Service Limited (Government of United Kingdom)
• Original Assignee: Forensic Science Service Limited (Government of United Kingdom)
• Inventors: Neumann, Cedric, Evett, Ian
• Primary Examiner(s): CHAWAN, SHEELA C
• Application Number:
  US13/123,879
• Publication Number:
  US 20120013439A1
• Time in Patent Office: 1,975 Days
• Field of Search: 382/124, 382/100, 382/115, 382/125, 382/181, 382/224, 382/228, 716/100, 716/101, 716/102, 716/103, 716/104, 716/132, 345/418, 345/419, 345/420, 345/473, 702/127, 702/179, 702/181, 706/45, 714/100, 714/E11.001, 714/E11.021, 714/E11.03, 714/E11.031, 714/1, 714/25, 714/37, 714/38.1
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1371 Matching features related to minutiae or pores

Abstract

An apparatus and a computer implemented method compare a first representation of an identifier with a second representation of an identifier to establish a likelihood ratio considering the probability the first representation and second representation originate from the same identifier and the probability of the first representation and second representation originate from different identifiers. The approach generates one or more variant expressions from the first representation, second representation and other representations. A boundary around the expression of the second representation is used to establish the number of the variant expressions of the expression of the first representation within it and the number of variant expressions of the expressions of the other representations within it and so provide the measure of comparison between the first representation of the identifier and the second representation of the identifier from the first value and second value.

Claims

1. A computer implemented method of comparing a first representation of an identifier with a second representation of an identifier, wherein the identifier is a biometric identifier or a form of identifier specific to a person or item, and in which the first representation and second representation is a fingerprint, palm print, ear print, retina image, footprint, shoe print, tool mark, or a part of any of these, the method comprising:

converting the first representation to an expression of the first representation, wherein the expression of the first representation is a sub-set of the data contained within the first representation;

generating one or more variant expressions from the expression of the first representation, wherein the one or more variant expressions are generated from the expression of the first representation to simulate the effect of a source of variation upon the expression of the first representation;

converting the second representation to an expression of the second representation, wherein the expression of the second representation is a sub-set of the data contained within the second representation;

providing a plurality of other representations and converting each of the other representations to an expression of the other representations, wherein the expressions of the other representations are a sub-set of the data contained within those other representations;

generating one or more variant expressions from each expression of the plurality of other expressions, in which the expressions of the other representations are a sub-set of the data contained within those other representations;

defining a boundary with respect to the expression of the second representation, the boundary defining a space inside the boundary;

determining a first value, the first value being an expression of the number of the variant expressions of the expression of the first representation within the space defined by the boundary;

determining a second value, the second value being an expression of the number of variant expressions of the expressions of the other representations within the space defined by the boundary;

determining a measure of comparison between the first representation of the identifier and the second representation of the identifier from the first value and the second value, in which the measure of comparison is the ratio of the first value to the second value.

2. A method according to claim 1 wherein the first representation is obtained from a known person and/or the second representation is obtained from an unknown person.

3. A method according to claim 1 wherein the other representations are obtained from a sample of a population.

4. A method according to claim 1 wherein the source of variation is distortion of the identifier during the generation of the representation.

5. A method according to claim 1 wherein the variant expressions are generated experimentally and/or the expression is used to automatically generate the variant expressions via a simulation method.

6. A method according to claim 1 wherein the variant expressions for the other representations are pre-generated and the variant expressions are stored in a memory.

7. A method according to claim 1 wherein the boundary is provided around the expression of the second representation.

8. A method according to claim 1 wherein the boundary has the same extent about the expression of the second representation in two or more dimensions.

9. A method according to claim 1 wherein the boundary has a different extent about the expression of the second representation in one or more dimensions.

10. A method according to claim 1 wherein the method includes determining a relationship between one or more of:

the expression of the second representation and the expression of the first representation; and

/or the expression of the second representation and one or more of the variant expressions of the expression of the first representation; and

/or the expression of the second representation and the expression of one or more of the other representations; and

/or the expression of the second representation and one or more of the variant expressions of one or more of the expressions of the other representations;

potentially, for one or more of these pairings.

11. A method according to claim 10 wherein the relationship is a distance between the expressions.

12. A method according to claim 1 wherein the method includes establishing a first probability distribution for one or more of:

the expression of the second representation and the expression of the first representation; and

/or the expression of the second representation and one or more of the variant expressions of the expression of the first representation.

13. A method according to claim 1 wherein the method includes establishing a second probability distribution for one or more of:

the expression of the second representation and the expression of one or more of the other representations; and

/or the expression of the second representation and one or more of the variant expressions of one or more of the expressions of the other representations;

potentially, for one or more of these pairings.

14. A method according to claim 1 wherein the first probability distribution relates to the variation between expressions and/or variant expressions from the same identifier.

15. A method according to claim 1 wherein the second probability distribution relates to the variation between expressions and/or variant expressions from different identifiers.

16. A method according to claim 1 wherein the first value is an expression of the proximity of the expression and/or variant expressions of the first representation to the expression of the second representation.

17. A method according to claim 1 wherein the first value is an expression of the proportion of the expression and/or variant expression of the first representation within the space defined by the boundary.

18. A method according to claim 17 wherein the expression of the proportion is compared with the total number of expression and/or variant expressions for the first representation.

19. A method according to claim 1 wherein the first value is the number of expression and/or variant expressions of the first representation within the space defined by the boundary divided by the total number of expression and/or variant expressions for the first representation.

20. A method according to claim 1 wherein the second value is an expression of the proximity of the expression and/or variant expressions of the other representation to the expression of the second representation.

21. A method according to claim 1 wherein the second value is an expression of the proportion of the expressions and/or variant expressions of the other representations within the space defined by the boundary.

22. A method according to claim 21 wherein the expression of the proportion is compared with the total number of expressions and/or variant expressions for the other representations.

23. A method according to claim 1 wherein the second value is the number of expressions and/or variant expressions of the other representations within the space defined by the boundary divided by the total number of expressions and/or variant expressions for the other representations.

24. A method according to claim 1 wherein the measure of comparison is a likelihood ratio and the likelihood ratio is the quotient of two probabilities, the numerator being the probability the first representation and second representation originate from the same identifier, the denominator being the probability of the first representation and second representation originate from different identifiers.

25. A method according to claim 1 wherein the measure of comparison provides an indication as to whether the first representation is likely to have the same source as the second representation.

26. A method according to claim 25 wherein the indication as to whether the first representation is likely to have the same source as the second representation is a yes or no indication and/or a quantified indication.

27. A method according to claim 1 wherein the measure of comparison is used to provide a decision for an access control apparatus and/or the measure of comparison is used to provide evidence for legal proceedings and/or the measure of comparison is used to provide evidence or information linking one item and/or location and/or person to another item and/or location and/or person.

28. An apparatus for comparing a first representation of an identifier with a second representation of an identifier, wherein the identifier is a biometric identifier or a form of identifier specific to a person or item, and in which the first representation and second representation is a fingerprint, palm print, ear print, retina image, footprint, shoe print, tool mark, or a part of any of these, the apparatus comprising:

an input channel configured to receive an expression of the first representation, wherein the expression of the first representation is a sub-set of the data contained within the first representation, and optionally a unit adapted to convert the first representation into an expression of the first representation;

a data processor configured to generate one or more variant expressions from the expression of the first representation, wherein the one or more variant expressions are generated from the expression of the first representation to simulate the effect of a source of variation upon the expression of the first representation;

wherein the input channel is configured for receiving an expression of the second representation, wherein the expression of the second representation is a sub-set of the data contained within the second representation, and optionally a unit adapted to convert the second representation to an expression of the second representation;

wherein the input channel is configured for receiving and/or data storage unit for storing an expression of a plurality of other representations, wherein the expressions of the other representations are a sub-set of the data contained within those other representations, and optionally a unit adapted to convert the plurality of other representations into expressions of the other representations;

wherein the data processor is configured to generate one or more variant expressions from each expression of the plurality of other expressions, wherein the one or more variant expressions are generated from each expression of the other representations to simulate the effect of a source of variation upon each expression of the other representations;

wherein the data processor is configured to calculate and/or to obtain from a data storage unit a definition of a boundary with respect to the expression of the second representation, the boundary defining a space inside the boundary;

wherein the data processor is configured to determine a first value, the first value being an expression of the number of the variant expressions of the expression of the first representation within the space defined by the boundary;

wherein the data processor is configured to determine a second value, the second value being an expression of the number of variant expressions of the expressions of the other representations within the space defined by the boundary;

wherein the data processor is configured to determine a measure of comparison between the first representation of the identifier and the second representation of the identifier from the first value and the second value, wherein the measure of comparison is the ratio of the first value to the second value, and optionally an output channel for expressing the measure of comparison.

29. Apparatus according to claim 28, wherein the data processor has a stored algorithm to generate the variant expressions from the first representation and/or plurality of other representations.

30. Apparatus according to claim 28, wherein the unit adapted to convert the first representation into an expression of the first representation and/or the unit adapted to convert the second representation into an expression of the second representation and/or the unit adapted to convert the plurality of other representations into expressions of the plurality of other representations includes an image scanner and/or the unit and/or device include a user interface and/or the unit and/or device include a data storage unit to provide a record of the selected features and/or the device includes an output unit for producing a record of the selected features.