Image recognition apparatus, image recognition method and image recognition program for specific pattern

• US 7,155,051 B2
• Filed: 05/21/2002
• Issued: 12/26/2006
• Est. Priority Date: 05/24/2001
• Status: Expired due to Fees

• Current Assignee: Minolta Corporation Limited (Konica Minolta Inc.)
• Original Assignee: Minolta Corporation Limited (Konica Minolta Inc.)
• Inventors: Murakawa, Akira
• Primary Examiner(s): Mehta, Bhavesh M.
• Assistant Examiner(s): Shah, Utpal
• Application Number:
  US10/151,165
• Publication Number:
  US 20020176616A1
• Time in Patent Office: 1,680 Days
• Field of Search: 382/287, 382/278, 382/291, 382/296, 382/131, 382/132, 382/135, 382144-145
• US Class Current: 382/135
• CPC Class Codes: G06V 10/40 Extraction of image or video features
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors

Abstract

In the image recognition processing, input color image data is analyzed, and a specific pattern contained in the graphics of the bill, security or the like is detected. The color image data is binarized with regard to specific colors. In order to detect the specific pattern from the binary image, a filter is sequentially moved to scan the entirety of the binarized image data. The examination target image within the filter is subjected to linear conversion (without changing the size thereof) at each examination position, and the degree of correlation between the pre-linear conversion image and the post-linear conversion image are calculated. It is then determined based on the calculated degree of similarity (i.e., the degree of matching) whether or not the specific pattern is included in the examination target image.

Claims

1. A specific pattern image recognition apparatus comprising:

a binary image generator which binarizes input multi-level color image with regard to specific colors and generates a binary image;

a linear conversion processor which performs linear conversion of an examination target image contained in the binary image without changing the size thereof;

a degree of correlation calculator which compares the examination target image with the linear-converted image and calculates a degree of correlation between the examination target image and the linear-converted image obtained by the linear conversion processor; and

a specific pattern determination unit that determines, based on the degree of correlation calculated from the comparison of the examination target image with the linear-converted image, whether the examination target image is a specific pattern.

2. A specific pattern image recognition apparatus as claimed in claim 1, further comprising:

an examination target image setter that specifies the examination target image in the binary image by moving a filter of a prescribed size.

3. A specific pattern image recognition apparatus as claimed in claim 1, further comprising:

a resolution converter that converts the input multi-level color image into a low-resolution multi-level color image,

wherein said binary image generator binarizes the low-resolution multi-level color image obtained via the conversion by the resolution converter.

4. A specific pattern image recognition apparatus as claimed in claim 1,

wherein said linear conversion performed by the linear conversion processor comprises 180°

rotation.

5. A specific pattern image recognition apparatus as claimed in claim 1,

wherein said linear conversion performed by the linear conversion processor comprises line-symmetry conversion.

6. A specific pattern image recognition method comprising the steps of:

inputting a multi-level color image;

binarizing the input multi-level color image with regard to specific colors and generating a binary image;

performing linear conversion of an examination target image in the binary image without changing the size thereof;

comparing the examination target with the linear-converted image of the examination target;

calculating a degree of correlation between the examination target image and the linear-converted image obtained after the linear conversion based on said comparison; and

determining whether the examination target image is the specific pattern based on the calculated degree of correlation calculated from the comparison of the the examination target image with the linear-converted image.

7. A specific pattern image recognition method as claimed in claim 6, further comprising the step of:

setting the examination target image in the binary image by moving a filter of a prescribed size.

8. A specific pattern image recognition method as claimed in claim 6, further comprising the step of:

converting the input multi-level color image into a low-resolution multi-level color image,

wherein the step of generating a binary image binarizes the low-resolution multi-level color image obtained via the resolution conversion.

9. A specific pattern image recognition method as claimed in claim 6,

wherein said linear conversion comprises 180°

image rotation.

10. A specific pattern image recognition method as claimed in claim 6,

wherein said linear conversion comprises line-symmetry conversion.

11. A specific pattern image recognition method comprising the steps of:

inputting a multi-level color image;

binarizing the input multi-level color image with regard to specific colors without changing the size thereof;

performing linear conversion of an examination target image in the binary image without changing the size thereof;

comparing the examination target image and the post-linear conversion image; and

determining whether or not the examination target image is the specific pattern based on the comparison result.

12. A computer program recorded on a computer recordable medium that causes the computer to carry out specific pattern image recognition, the program comprising the steps of:

inputting a multi-level color image;

binarizing the input multi-level color image with regard to specific colors and generating a binary image;

performing linear conversion of an examination target image in the binary image without changing the size thereof;

comparing the examination target image and the linear-converted image obtained after the linear conversion;

calculating a degree of correlation between the examination target image and the linear-converted image obtained after the linear conversion; and

determining based on the calculated degree of correlation whether or not the examination target image is the specific pattern calculated from the comparison of the examination target image with the linear-converted image.

13. A computer program recorded on a computer recordable medium as claimed in claim 12, further comprising the step of:

setting the examination target image in the binary image by moving a filter of a prescribed size.

14. A computer program recorded on a computer recordable medium as claimed in claim 12, further comprising the step of:

converting the input multi-level color image into a low-resolution multi-level color image,

wherein the step of generating a binary image binarizes the low-resolution multi-level color image obtained via the resolution conversion.

15. A computer-readable recording medium recording the program as claimed in claim 12.