System and method for identification of printed matter in an image

• US 9,336,452 B2
• Filed: 01/16/2012
• Issued: 05/10/2016
• Est. Priority Date: 01/16/2011
• Status: Active Grant

• Current Assignee: EyeCue Vision Technologies LTD
• Original Assignee: EyeCue Vision Technologies LTD
• Inventors: Kaftory, Ran
• Primary Examiner(s): Carter, Aaron W
• Application Number:
  US13/979,671
• Publication Number:
  US 20130294700A1
• Time in Patent Office: 1,576 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 10/243 by compensating for image skew or non-uniform image deformations
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/40 Extraction of image or video features

Abstract

A system and method of identifying a printed matter in an image that includes the printed matter by rectifying the image, identifying a series of pre-defined areas in the image relative to a frame or other form in the printed matter in the image, and comparing a characteristic of the pre-defined area in the printed matter in the image, to a characteristic of a corresponding pre-defined area from a sample of printed matters.

Claims

1. A method of identifying a printed matter in an image, comprising:

capturing a first image of a first printed matter, said first printed matter including a first reference form;

storing characteristics of a first plurality of pre-defined areas in said first printed matter in said first image, wherein each area of said first plurality of predefined areas is at a known location relative to said first reference form,

capturing a second image, said second image including a second printed matter;

rectifying a portion of said second image, said portion of said second image including at least a first portion of said second printed matter;

identifying in said first portion of said second printed matter in said second image a second reference form and a second plurality of pre-defined areas, each area of said second plurality of pre-defined areas at a known location in said second printed matter relative to said second reference form;

comparing

a characteristic of a first pre-defined area of said second plurality of pre-defined areas of said second printed matter in said second image

to

the stored characteristics of a first pre-defined area of said first plurality of pre-defined areas in said first printed matter in said first image,

where the location of said first pre-defined area of said second plurality of pre-defined areas in said second printed matter relative to said second reference form in said second printed matter,

corresponds to

said location of said first pre-defined area of said first plurality of pre-defined areas in said first printed matter in said first image relative to said first reference form in said first image.

2. The method as in claim 1, wherein said rectifying said portion of said second image comprises rectifying a portion of said second printed matter in said second image that includes said second reference form and said second plurality of pre-defined areas having known locations in said second printed matter relative to said second reference form.

3. The method as in claim 1, comprising calculating a probability that said characteristic of said first pre-defined area of said second plurality of pre-defined areas of said second printed matter in said second image corresponds to said characteristic of said first pre-defined area of said first plurality of pre-defined areas in said first printed matter.

4. The method as in claim 1, wherein said comparing a characteristic of said first pre-defined area of said second plurality of pre-defined areas of said second printed matter in said second image comprises

comparing

said characteristic of said first pre-defined area of said second plurality of pre-defined areas, said plurality of areas constituting a pre-defined shape in said second printed matter in said second image,

to

stored data about a characteristic of said first pre-defined area of said first plurality of pre-defined areas constituting said shape in said first printed matter.

5. The method as in claim 1, wherein said comparing said characteristic comprises comparing a color of said first pre-defined area of said second plurality of pre-defined areas of said second printed matter in said second image, to stored data of a color of said first pre-defined area in said first printed matter.

6. The method as in claim 1, wherein said comparing comprises comparing a detected pattern of said second plurality of pre-defined areas of said second printed matter in said second image, to stored data about a pattern of said first plurality of pre-defined areas of said first printed matter, wherein said pattern comprises background spaces and non-background spaces.

7. The method as in claim 1, wherein said identifying in said first portion of said second printed matter appearing in said second image a second plurality of pre-defined areas of said second printed matter appearing in said second image, comprises calculating a distance of said first pre-defined area of said second plurality of pre-defined areas in said second image from said second reference form appearing in said second printed matter in said second image.

8. The method as in claim 1, wherein said comparing comprises comparing a characteristic of said first pre-defined area of said second plurality of pre-defined areas of said second printed matter in said second image to stored data about said characteristic of a pre-defined area in a plurality of printed matters.

12. The method as in claim 1, wherein said second image is captured later than said first image.

9. A system comprising:

an imager to capture

a first image, said first image including a first printed matter; and

,

a second image, said second image including a second printed matter,

a memory to store

a characteristic of a first point of a first plurality of points in said first printed matter and

a characteristic of a first point of a second plurality of points in said second printed matter,

and a processor,

said processor configured to

rectify a portion of said second image,

said portion including said second printed matter;

calculate a location of said first point of said first plurality of points in said first printed matter relative to a printed form in said first image; and

to

calculate a location of said first point of said second plurality of points in said second printed matter relative to a printed form in said second image;

identify said location of said first point of said first plurality of points with said location of said first point of said second plurality of points; and

compare said characteristic of said first point of said second plurality of points in said second image to said characteristic of said first point of said first plurality of points in said first image.

10. The system as in claim 9, wherein said processor is configured to calculate a probability that said second printed matter in said second image is an image of said first printed matter.

11. The system as in claim 9, wherein said processor is configured to compare a characteristic of said first point of said second plurality of points in said second printed matter in said second image to stored data about said characteristic of a point in a plurality of printed matters.