Transaction Verification System

• US 20140067679A1
• Filed: 08/27/2013
• Published: 03/06/2014
• Est. Priority Date: 08/28/2012
• Status: Abandoned Application

• Current Assignee: Solink Corporation
• Original Assignee: Solink Corporation
• Inventors: O'Reilly, John, Youmaran, Richard, Paliga, Andrzej, Matta, Michael, Waddington, Christopher
• Application Number:
  US14/010,757
• Publication Number:
  US 20140067679A1
• Time in Patent Office: Days
• US Class Current: 705/44
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06Q 20/40145 Biometric identity checks
• CPC Class Codes: G06V 40/168 Feature extraction; Face representation

Abstract

An analytics server for use in a transaction system with a terminal for performing authenticated user-initiated transactions and generating transaction data including a user identity associated with each transaction and a camera for capturing image data of a user performing an authenticated transaction at the terminal is configured to extract user characteristic features from the image data associated with authenticated transactions and iteratively update a user database of the user characteristic features over multiple authenticated transactions. The analytics server is further configured to compute a match score, based on preset rules, of the user characteristic features for a current transaction with the user characteristic features associated with a current user stored in the user database, and raise an alarm when the match score fails to meet a threshold value.

Claims

1. A system comprising:

a terminal for performing an authenticated user-initiated transaction by a current user and generating transaction data including a user identity generated in association with the authenticated user-initiated transaction;

a camera for capturing image data of a user performing an authenticated transaction at the terminal;

an analytics server connected to the terminal via network, the analytics server comprising at least a processing circuit for processing image data from the camera, the analytics server being configured to;

extract user characteristic features from the image data associated with authenticated transactions;

iteratively update a user database of the user characteristic features over multiple authenticated transactions;

computing with the processing circuit a match score based upon preset rules for the user characteristic features relating to a current user performing a current transaction with the user characteristic features associated with an authenticated user, the authenticated user being established in dependence upon the user identity; and

determining whether to raise an alarm when the match score fails to meet a threshold value.

2. The system according to claim 1, wherein characteristic feature vectors are at least one of created and updated after every N transactions, where N is a positive integer.

3. The system according to claim 1, wherein

the analytics server is further configured to at least one of:

process the user characteristic features in real time and update the user database when the match score meets the threshold value;

process the user characteristic features as a background process by matching the time of authenticated transactions with the time of the captured image data; and

send an alarm to an alarm reporter module in the event of the match score failing to reach meet the threshold value.

4. The system according to claim 1, wherein

the camera is one camera of a plurality of cameras, each camera connected to the analytics server via the network and at least one of associated with the terminal and capturing images of an area within which the terminal is located.

5. The system according to claim 1, wherein

the analytics server is configured to apply a plurality of image processing algorithms to the image data, each image processing algorithm selected from the group comprising background extraction, region of interest determination, region of interest extraction;

mask extraction, mask generation, a morphological operator, Gaussian segmentation, watershed segmentation, a filter, a Gabor filter, Gabor extraction, colour filtering, and texture filtering.

6. The system according to claim 1, wherein

the analytics server is configured to perform a recognition process upon at least one of the image data and the image data after image processing, the recognition process selected from the group comprising mask extraction using combined Gaussian and watershed-based segmentation techniques, identification of a human body, identification of a human face, identification of a human body segment, detection of occlusion of a human face, global human face feature extraction, colour information extraction, texture extraction, generating a region of interest, facial information extraction, and human ear information extraction.

7. The system according to claim 1, wherein

the analytics server further comprises;

a colour module for generating from the image data a color index table using N main colors, where N is a positive integer;

a block extraction module for extracting a plurality of blocks of pixels from the image data; and

a feature vector block for organizing in vector format the image pixel values within block of pixels of the plurality of blocks of pixels; and

a colour vector module for computing a color vector for each block of pixels of the plurality of blocks of pixels for dominant colors;

and

the match score is also generated in dependence upon the output from a matching algorithm using a color vector of the image data and color data stored in association with the user characteristic features in the user database.

8. The system according to claim 1, wherein

computing the match score comprises selecting a user match score of a plurality of user match scores, each user match score being generated by applying a predetermined sequence of image processing algorithms on the image data to establish current user characteristic features and applying a predetermined set of rules to the current user characteristic features and a set of user characteristic features associated with an individual associated with a financial instrument employed within the authenticated user-initiated transaction.

9. The system according to claim 1, wherein

an initial set of user characteristic features are established within the user database for a user based upon the system being provided a training set of images selected by at least one of the user themselves and the system itself based upon accessing one or more social networks relating to at least one of the user and a friend of the user.

10. The system according to claim 1, where

the terminal comprises at least an electronic device and a software application installed upon a the electronic device;

and

the camera is part of the electronic device and allows either acquisition of the image data automatically during the authenticated user-initiated transaction or as a discrete step within the authenticated user-initiated transaction.

11. A system comprising:

an analytics server for connecting to a network comprising at least a processing circuit, the analytics server being configured to;

receive image data from a camera connected to the network for capturing image data of a user performing an authenticated transaction at a terminal also connected to the network;

receive transaction data from the terminal, the transaction data including a user identity generated in association with the authenticated user-initiated transaction performed by a current user;

extract user characteristic features from the image data associated with authenticated transactions;

iteratively update a user database of the user characteristic features over multiple authenticated transactions;

compute with the processing circuit a match score based upon preset rules for the user characteristic features relating to a current user performing a current transaction with the user characteristic features associated with an authenticated user, the authenticated user being established in dependence upon the user identity; and

determining whether to raise an alarm when the match score fails to meet a threshold value.

12. The system according to claim 11, wherein characteristic feature vectors are at least one of created and updated after every N transactions, where N is a positive integer.

13. The system according to claim 11, wherein

the analytics server is further configured to at least one of:

process the user characteristic features in real time and update the user database when the match score meets the threshold value;

process the user characteristic features as a background process by matching the time of authenticated transactions with the time of the captured image data; and

send an alarm to an alarm reporter module in the event of the match score failing to reach meet the threshold value.

14. The system according to claim 11, wherein

the analytics server is configured to apply a plurality of image processing algorithms to the image data, each image processing algorithm selected from the group comprising background extraction, region of interest determination, region of interest extraction;

mask extraction, mask generation, a morphological operator, Gaussian segmentation, watershed segmentation, a filter, a Gabor filter, Gabor extraction, colour filtering, and texture filtering.

15. The system according to claim 11, wherein

the analytics server is configured to perform a recognition process upon at least one of the image data and the image data after image processing, the recognition process selected from the group comprising mask extraction using combined Gaussian and watershed-based segmentation techniques, identification of a human body, identification of a human face, identification of a human body segment, detection of occlusion of a human face, global human face feature extraction, colour information extraction, texture extraction, generating a region of interest, facial information extraction, and human ear information extraction.

16. The system according to claim 11, wherein

the analytics server further comprises;

a colour module for generating from the image data a color index table using N main colors, where N is a positive integer;

a block extraction module for extracting a plurality of blocks of pixels from the image data; and

a feature vector block for organizing in vector format the image pixel values within block of pixels of the plurality of blocks of pixels; and

a colour vector module for computing a color vector for each block of pixels of the plurality of blocks of pixels for dominant colors;

and

the match score is also generated in dependence upon the output from a matching algorithm using a color vector of the image data and color data stored in association with the user characteristic features in the user database.

17. The system according to claim 11, wherein

computing the match score comprises selecting a user match score of a plurality of user match scores, each user match score being generated by applying a predetermined sequence of image processing algorithms on the image data to establish current user characteristic features and applying a predetermined set of rules to the current user characteristic features and a set of user characteristic features associated with an individual associated with a financial instrument employed within the authenticated user-initiated transaction.

18. The system according to claim 11, wherein

an initial set of user characteristic features are established within the user database for a user based upon the system being provided a training set of images selected by at least one of the user themselves and the system itself based upon accessing one or more social networks relating to at least one of the user and a friend of the user.

19. A method comprising:

receiving at a server comprising at least a processing circuit image data from a camera connected to a network, the captured image data relating to a current user performing an authenticated transaction at a terminal also connected to the network;

receiving at the server transaction data from the terminal, the transaction data including a user identity generated in association with the authenticated user-initiated transaction performed by the current user;

extracting with the processing circuit user characteristic features from the image data associated with authenticated transaction;

iteratively updating with the processing circuit a user database of the user characteristic features over multiple authenticated transactions;

computing with the processing circuit a match score based upon preset rules for the user characteristic features relating to a current user performing a current transaction with the user characteristic features associated with an authenticated user, the authenticated user being established in dependence upon the user identity; and

determining with the processing circuit whether to raise an alarm when the match score fails to meet a threshold value.

20. The method according to claim 19, wherein

the camera is a camera of a plurality of cameras, each camera connected to the network and at least one of associated with the terminal and capturing images of an area within which the terminal is located.

21. The method according to claim 19, wherein

extracting with the processing circuit user characteristic features comprises applying a plurality of image processing algorithms to the image data and performing a plurality of recognition processing algorithms on the processed image data.

22. The system according to claim 19, wherein

computing the match score comprises selecting a user match score of a plurality of user match scores, each user match score being generated by applying a predetermined sequence of image processing algorithms on the image data to establish current user characteristic features and applying a predetermined set of rules to the current user characteristic features and a set of user characteristic features associated with an individual associated with a financial instrument employed within the authenticated user-initiated transaction.

23. The method according to claim 19, wherein

an initial set of user characteristic features are established within the user database for a user based upon the system being provided a training set of images selected by at least one of the user themselves and the system itself based upon accessing one or more social networks relating to at least one of the user and a friend of the user.

24. The method according to claim 19, where

the terminal comprises at least an electronic device and a software application installed upon a the electronic device;

and

the camera is part of the electronic device and allows either acquisition of the image data automatically during the authenticated user-initiated transaction or as a discrete step within the authenticated user-initiated transaction.