Validation template for valuable media of multiple classes

• US 20080159614A1
• Filed: 08/02/2007
• Published: 07/03/2008
• Est. Priority Date: 12/29/2006
• Status: Active Grant

• Current Assignee: NCR Corporation
• Original Assignee: NCR Corporation
• Inventors: Ross, Gary A., He, Chao
• Granted Patent: US 8,503,796 B2
• Time in Patent Office: Days
• US Class Current: 382/135
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06F 18/2433 Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection

Abstract

A technique for use in automatic validation of a media item involves accessing a template that comprises multiple one-class classifiers, each corresponding to one of multiple classes to which the media item might belong, and then applying each of the one-class classifiers to an image of the media item to generate a result set for each of the multiple classes. The result set for each media class is then analyzed to assess whether the media item belongs to that class.

Claims

1. A method for use in automatic validation of a media item, the method comprising:

accessing a template that comprises multiple one-class classifiers, each corresponding to one of multiple classes to which the media item might belong;

applying each of the one-class classifiers to an image of the media item to generate a result set for each of the multiple classes; and

for each of the multiple classes, analyzing the result set and assessing whether the media item belongs to that class.

2. The method of claim 1, further comprising concluding that the media item belongs to none of the multiple classes and, in response, concluding that the media item is not valid.

3. The method of claim 1, further comprising concluding that the media item belongs to one of the multiple classes and, in response, concluding that the media item is valid.

4. The method of claim 1, where at least one of the multiple classes is associated with each of multiple orientations in which the media item can be inserted into an automated validation machine.

5. The method of claim 1, where at least one of the multiple classes is associated with each of multiple production series in which the media item might be produced.

6. The method of claim 1, where the media item is a bank note, and where accessing the template includes accessing a template that corresponds to a particular currency and a particular denomination to which the bank note belongs.

7. The method of claim 1, where at least one of the multiple classes is associated with each of multiple orientations in which a bank note of the particular currency and particular denomination can be inserted into an automatic validation machine.

8. The method of claim 1, where at least one of the multiple classes is associated with each of multiple production series in which a bank note of the particular currency and particular denomination might be produced.

9. The method of claim 1, further comprising:

retrieving from the template a single segmentation map that applies to media items belonging to all of the multiple classes; and

applying the segmentation map to the image of the media item to extract a feature set for the image.

10. The method of claim 9, where applying each of the one-class classifiers to the image of the media item includes applying each of the one-class classifiers to the feature set for the image to generate the result set for each of the multiple classes.

11. The method of claim 1, where accessing the template includes accessing a template that comprises multiple one-class classifiers that each represents one of multiple feature clusters, where each of the multiple feature clusters is associated with one of the multiple classes to which the media item might belong.

12. The method of claim 1, where the media item is a bank note.

13. A method for use in creating a template for automated validation of media items, the method comprising:

receiving images of multiple media items of similar type, each of the images belonging to one of multiple classes associated with media items of that type;

using the images to creates a single segmentation map that represents media items belonging to all of the multiple classes;

using the images and the segmentation map to create multiple one-class classifiers, where each of the one-class classifiers is associated with one of the multiple classes; and

defining the template to include the segmentation map and the multiple one-class classifiers.

14. The method of claim 13, further comprising using the images to create multiple feature clusters that each is associated with one of the multiple classes.

15. The method of claim 14, where using the images and the segmentation map to create multiple one-class classifiers includes applying the segmentation map to each of the multiple feature clusters to extract multiple feature sets, where each of the feature sets is associated with one of the multiple classes.

16. The method of claim 13, where at least one of the multiple classes is associated with each of multiple orientations in which the media item can be inserted into an automated validation machine.

17. The method of claim 13, where at least one of the multiple classes is associated with each of multiple production series in which media items of that type are produced.

18. The method of claim 13, where the media item is a bank note that belongs to a particular currency and a particular denomination.

19. The method of claim 18, where at least one of the multiple classes is associated with each of multiple orientations in which a bank note of the particular currency and particular denomination can be inserted into an automatic validation machine.

20. The method of claim 18, where at least one of the multiple classes is associated with each of multiple production series in which bank notes of the particular currency and particular denomination are produced.

21. The method of claim 13, where the media items include bank notes.