Biometric matching technology

• US 9,582,723 B2
• Filed: 10/14/2016
• Issued: 02/28/2017
• Est. Priority Date: 04/09/2012
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services Limited (Accenture PLC)
• Inventors: Tiwari, Abhinav, Paul, Sanjoy
• Primary Examiner(s): Rahmjoo, Manuchehr
• Application Number:
  US15/293,496
• Publication Number:
  US 20170032204A1
• Time in Patent Office: 137 Days
• Field of Search: 382/116
• US Class Current: 1/1
• CPC Class Codes: G06F 16/532 Query formulation, e.g. graphical querying
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 18/23 Clustering techniques
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 30/242 Division of the character sequences into groups prior to recognition; Selection of dictionaries
• CPC Class Codes: G06V 30/244 using graphical properties, e.g. alphabet type or font
• 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 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: G06V 40/172 Classification, e.g. identification
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities

Abstract

Biometric matching technology, in which a watch list is managed, multiple images of a potential suspect are accessed, and parallel pre-processing of the multiple images is controlled. Based on the pre-processing, an image of the potential suspect to use in matching against the watch list is determined and the determined image is used to search sorted biometric data included in the watch list. A subset of persons from the watch list is identified based on the search and parallel analysis of the determined image of the potential suspect against detailed biometric data associated with the subset of persons in the watch list is controlled. Based on the parallel analysis, it is determined whether the potential suspect matches a person in the watch list and a result is outputted based on the determination.

Claims

1. A system comprising:

a database storage configured to maintain, for each person in a group of multiple, different people, data for the person; and

an enrollment system that includes at least one computer, the enrollment system configured to;

for each of a plurality of potential reference images;

compare, for each person in the group of multiple, different people, the potential reference image with an image of at least a portion of the person; and

determine, for each person in the group of multiple, different people, a similarity score that represents a similarity between the potential reference image and the image of at least the portion of the person using the comparison of the potential reference image with the image of at least the portion of the person;

determine which of the potential reference images have similarity scores that are most evenly distributed for the images of the persons in the group of multiple, different people;

select, as a reference image, a single potential reference image that has similarity scores that are most evenly distributed for the images of the persons in the group of multiple, different people;

determine, for each person in the group of multiple, different people, a location in the database storage at which to store an identifier for the person using the respective similarity score for the person as an index in the database storage; and

store, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage.

2. The system of claim 1, wherein the enrollment system is configured to:

receive, for each person of one or more new persons to add to the group of multiple, different people, an image of at least a portion of the person;

compare, for each person of the one or more new persons, the selected reference image with the image of at least the portion of the person;

determine, for each person of the one or more new persons, a similarity score that represents a similarity between the selected reference image and the image of at least the portion of the person;

evaluate a distribution of similarity scores for the selected reference image;

determine that the distribution of the similarity scores for the selected reference image are not evenly distributed using the evaluation of the distribution of similarity scores for the selected reference image;

select, as a second reference image that is a different reference image than the reference image, a single potential reference image from among multiple reference images in response to determining that the distribution of the similarity scores for the selected reference image are not evenly distributed;

determine, for each person in the group of multiple, different people, a second location in the database storage at which to store the data for the person using a second similarity score for the person as an index in the database storage, wherein the second similarity score represents a similarity between the second reference image and the image of the person; and

store, in the database storage for each person in the group of multiple, different people, the data for the person at the respective second location in the database storage.

3. The system of claim 2, wherein determining that the distribution of the similarity scores for the selected reference image are not evenly distributed comprises determining, for a first cluster of the similarity scores for the selected reference image, that a first distribution of similarity scores in the cluster is more dense than a second distribution of similarity scores in a second cluster of similarity scores for the selected reference image.

4. The system of claim 2, wherein:

storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people and at the respective location in the database storage, an association to biometric data for the person, wherein the respective location references the biometric data for the person using the association; and

storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective second location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people and at the respective second location in the database storage, the association to the biometric data for the person.

5. The system of claim 4, wherein, for each person in the group of multiple, different people, the association comprises a link, for each person in the group of multiple, different people, that identifies a particular location at which the biometric data for the person is stored.

6. The system of claim 1, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people and at the respective location in the database storage, an association to biometric data for the person, wherein the respective location references the biometric data for the person using the association.

7. The system of claim 1, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location comprises storing, in the database storage for each person in the group of multiple, different people at the respective location, biometric data for the person.

8. The system of claim 1, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises:

creating, in a relational database storage included in the database storage, a table;

storing, for each person in the group of multiple, different people, the data for the person at the respective location in the table;

determining, for each person in the group of multiple, different people, whether blob data for biometric data of the person meets a threshold storage size; and

based on determining whether the blob data for the biometric data of the person meets the threshold storage size, performing one of;

for each person in the group of multiple, different people for which the blob data for the biometric data of the person does not meet the threshold storage size, storing the blob data in the table with the data for the person, or

for each person in the group of multiple, different people for which the blob data for the biometric data of the person meets the threshold storage size, storing the blob data in a non-relational storage and storing a link to the blob data in the table with the data.

9. The system of claim 1, wherein the enrollment system is configured to:

for two or more of the plurality of potential reference images;

determine at least one criterion relevant to the potential reference image based on at least one of sensor data and camera or camera data;

associate the potential reference image with the at least one criterion;

determine, for each person in the group of multiple, different people, a particular location in the database storage at which to store an data for the person using a similarity score, for the person computed using the potential reference image, as an index in a portion of the database storage associated with the potential reference image; and

store, in the portion of the database storage associated with the potential reference image and for each person in the group of multiple, different people, the data for the person at the respective particular location in the database storage.

10. The system of claim 9, comprising an authentication system that includes at least one computer, the authentication system configured to:

receive a particular image of at least a portion of a potential suspect;

determine at least one particular criterion relevant to the particular image based on at least one of sensor data or camera data;

select, as the reference image, a particular reference image from the two or more of the plurality of reference images that is associated with the particular criterion;

compute a particular similarity score that represents a similarity between the particular image and the particular reference image;

search the portion of the database storage associated with the particular reference image using the particular similarity score; and

output a result based on the search of the portion of the database storage associated with the particular reference image using the particular similarity score.

11. The system of claim 1, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people, an identifier for the person at the respective location in the database storage.

12. A method comprising:

for each of a plurality of potential reference images;

comparing, for each person in a group of multiple, different people, the potential reference image with an image of at least a portion of the person; and

determining, for each person in the group of multiple, different people, a similarity score that represents a similarity between the potential reference image and the image of at least the portion of the person using a result of comparing the potential reference image with the image of at least the portion of the person;

determining which of the potential reference images have similarity scores that are most evenly distributed for the images of the persons in the group of multiple, different people;

selecting, as a reference image, a single potential reference image that has similarity scores that are most evenly distributed for the images of the persons in the group of multiple, different people;

determining, for each person in the group of multiple, different people, a location in a database storage at which to store an identifier for the person using the respective similarity score for the person as an index in the database storage; and

storing, in the database storage for each person in the group of multiple, different people, data for the person at the respective location in the database storage.

13. The method of claim 12, comprising:

receiving, for each person of one or more new persons to add to the group of multiple, different people, an image of at least a portion of the person;

comparing, for each person of the one or more new persons, the selected reference image with the image of at least the portion of the person;

determining, for each person of the one or more new persons, a similarity score that represents a similarity between the selected reference image and the image of at least the portion of the person;

evaluating a distribution of similarity scores for the selected reference image;

determining that the distribution of the similarity scores for the selected reference image are not evenly distributed using a result of evaluating the distribution of similarity scores for the selected reference image;

selecting, as a second reference image that is a different reference image than the reference image, a single potential reference image from among multiple reference images in response to determining that the distribution of the similarity scores for the selected reference image are not evenly distributed;

determining, for each person in the group of multiple, different people, a second location in the database storage at which to store the data for the person using a second similarity score for the person as an index in the database storage, wherein the second similarity score represents a similarity between the second reference image and the image of the person; and

storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective second location in the database storage.

14. The method of claim 13, wherein determining that the distribution of the similarity scores for the selected reference image are not evenly distributed comprises determining, for a first cluster of the similarity scores for the selected reference image, that a first distribution of similarity scores in the cluster is more dense than a second distribution of similarity scores in a second cluster of similarity scores for the selected reference image.

15. The method of claim 13, wherein:

storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people and at the respective location in the database storage, an association to biometric data for the person, wherein the respective location references the biometric data for the person using the association; and

storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective second location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people and at the respective second location in the database storage, the association to the biometric data for the person.

16. The method of claim 15, wherein storing, for each person in the group of multiple, different people, the association to the biometric data for the person comprises storing, for each one person in the group of multiple, different people, a link that identifies a particular location at which the biometric data for the person is stored.

17. The method of claim 12, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises storing, in the database storage for each person in the group of multiple, different people and at the respective location in the database storage, an association to biometric data for the person, wherein the respective location references the biometric data for the person using the association.

18. The method of claim 12, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location comprises storing, in the database storage for each person in the group of multiple, different people at the respective location, biometric data for the person.

19. The method of claim 12, wherein storing, in the database storage for each person in the group of multiple, different people, the data for the person at the respective location in the database storage comprises:

creating, in a relational database storage included in the database storage, a table;

storing, for each person in the group of multiple, different people, the data for the person at the respective location in the table;

determining, for each person in the group of multiple, different people, whether blob data for biometric data of the person meets a threshold storage size; and

based on determining whether the blob data for the biometric data of the person meets the threshold storage size, performing one of;

for each person in the group of multiple, different people for which the blob data for the biometric data of the person does not meet the threshold storage size, storing the blob data in the table with the data for the person, or

for each person in the group of multiple, different people for which the blob data for the biometric data of the person meets the threshold storage size, storing the blob data in a non-relational storage and storing a link to the blob data in the table with the data.

20. At least one non-transitory computer-readable storage medium encoded with executable instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:

for each of a plurality of potential reference images;

comparing, for each person in a group of multiple, different people, the potential reference image with an image of at least a portion of the person; and

determining, for each person in the group of multiple, different people, a similarity score that represents a similarity between the potential reference image and the image of at least the portion of the person using a result of comparing the potential reference image with the image of at least the portion of the person;

determining which of the potential reference images have similarity scores that are most evenly distributed for the images of the persons in the group of multiple, different people;

selecting, as a reference image, a single potential reference image that has similarity scores that are most evenly distributed for the images of the persons in the group of multiple, different people;

determining, for each person in the group of multiple, different people, a location in a database storage at which to store an identifier for the person using the respective similarity score for the person as an index in the database storage; and

storing, in the database storage for each person in the group of multiple, different people, data for the person at the respective location in the database storage.