Granted Patent B2

US 9,824,268 · App. 14/264,860 · Granted Nov 21, 2017

Abstract

Claims (31)

1. A method of validating a media item, the method comprising:

receiving image data corresponding to a plurality of digital images of a media item to be validated, wherein each image is associated with a respective one of a plurality of different channels of an illumination source for illuminating the media item, each channel representing a different frequency of light emitted from the illumination source;

selecting image data corresponding to two of the plurality of images, wherein the selecting the image data further includes converting the image data to pixel intensity values;

dividing a common parameter of the selected image data associated with corresponding locations of the two selected images and creating ratio image data as a ratio image for a first channel of two different channels of the illumination source, wherein a first channel is represented as a first frequency of light that is reflected off the media item by a reflective first color of visible light that is produced by illuminating the media item with a green color of visible light with the first frequency of light captured for a first image of the media item, and a second channel of the two different channels is represented as a second frequency of light that is reflected off the media item by a second color of non-visible light that is produced by illuminating the media item with the second color of non-visible light with the second frequency of light captured for a second image of the media item, wherein the second color is infrared light, and wherein the ratio image data represents the pixel intensity values for the first and second images of the media item, and wherein the ratio image is the pixel intensity values present in the first image divided by the pixel intensity values present in the second image;

extracting feature information associated with the ratio image: and

classifying the extracted feature information with at least one classifier of a validation template to determine whether or not the media item is valid.

2. The method as claimed in claim 1 , further comprising:

segmenting the ratio image using a segmentation map of the validation template;

extracting feature information from each segment; and

classifying the extracted feature information from each segment with a corresponding segment classifier of the validation template.

3. The method as claimed in claim 1 , further comprising:

combining the extracted feature information from all segments; and

classifying the combined extracted feature information from all segments with a classifier of the validation template.

4. The method as claimed in claim 3 , further comprising:

inputting the extracted feature information into one or more functions of the at least one classifier to produce an output; and

comparing the output with at least one threshold parameter of the at least one classifier to determine whether or not the media item is valid.

5. The method as claimed in claim 1 , wherein the at least one classifier is a one class classifier.

6. The method as claimed in claim 1 , wherein the validation template is selected from a plurality of validation templates responsive to a class of media item to be validated.

7. The method as claimed in claim 6 , wherein the media item is a banknote and the class comprises currency, denomination, orientation and design series.

8. The method as claimed in claim 6 , wherein each validation template is created on the basis of information about a set of training ratio images of a plurality of media items of the same class.

9. The method as claimed in claim 6 , wherein the segmentation map of a corresponding validation template is created by using a clustering algorithm to cluster pixel locations of each corresponding ratio image of a selected channel using a plurality of media items of the same class.

10. The method as claimed in claim 9 , wherein the segmentation map of a corresponding validation template is created by using a clustering algorithm to cluster pixel locations of each corresponding ratio image of all ratio channels together using a plurality of media items of the same class.

11. The method as claimed in claim 1 , wherein the common parameter comprises luminous intensity.

12. The method as claimed in claim 1 , further comprising:

adjusting at least one property of the ratio image.

13. A method of validating a media item, the method comprising:

receiving image data corresponding to a plurality of digital images of a media item to be validated, wherein each image is associated with a respective one of a plurality of different channels of an illumination source for illuminating the media item;

selecting image data corresponding to two of the plurality of images, wherein the selecting the image data further includes converting the image data to pixel intensity values;

dividing a common parameter of the selected image data associated with corresponding locations of the two selected images and creating ratio image data as a ratio image for a first channel of two different channels of the illumination source represented by a reflective first color of visible light for a first image of the media item and a second channel of the two different channels of the illumination source represented by a reflective second color of non-visible light for a reflective infrared light of a second image of the media item, wherein the ratio image data represents the pixel intensity values present in the images, and the ratio image is the pixel intensity values present in the first image divided by the pixel intensity values present in the second image, and wherein the first channel is a first frequency captured as the reflective first color of light when the media item is illuminated with a green color of light and captured as the first image, and wherein the second channel is a second frequency captured as the reflective second color of light when the media item is illuminated with infrared light and captured as the second image;

extracting feature information associated with the ratio image: and

classifying the extracted feature information with at least one classifier of a validation template to determine whether or not the media item is valid.

Assignments (9)