Image Recognition Device, Copy Apparatus and Image Recognition Method

• US 20080075369A1
• Filed: 09/20/2007
• Published: 03/27/2008
• Est. Priority Date: 09/27/2006
• Status: Active Grant

• Current Assignee: Brother Kogyo Kabushiki Kaisha
• Original Assignee: Brother Kogyo Kabushiki Kaisha
• Inventors: SATO, Yuji
• Granted Patent: US 8,358,848 B2
• Time in Patent Office: Days
• US Class Current: 382/190
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 20/95 Pattern authentication; Markers therefor; Forgery detection

Abstract

An image recognition device of the invention includes an image reading means for reading image information from a manuscript, discriminating a specific image in which a surrounded image is surrounded by a surrounding image from the image information. The image recognition device includes a first determination means for determining whether the surrounding image is included in the image information or not and a second determination means for determining whether there is the specific image in the image information or not by extracting n×n images to be processed by dividing an extraction region positioned at the center of the surrounding image into n×n regions (“n” is an odd-number of three or more) when the first determination means determines that the surrounding image is included in the image information, then, by checking respective images to be processed with a specific image template.

Claims

1. An image recognition device including an image reading means for reading image information from a manuscript and discriminating a specific image in which a surrounded image is surrounded by a surrounding image from the image information, comprising:

a first determination means for determining whether the surrounding image is included in the image information or not; and

a second determination means for determining whether there is the specific image in the image information or not by extracting n×

n images to be processed by dividing an extraction region positioned at the center of the surrounding image into n×

n regions (“

n”

is an odd-number of three or more) when the first determination means determines that the surrounding image is included in the image information, then, by checking respective images to be processed with a specific image template.

2. The image recognition device according to claim 1,

wherein the first determination means extracts images to have pre-processing which have larger areas than the images to be processed from the image information, and determines whether there is the surrounding image in the image information based on the respective images to have pre-processing.

3. The image recognition device according to claim 2,

wherein the first determination means divides a pre-extraction region at an optional position in the image information into 2×

2 regions and extracts four images to have pre-processing, and

wherein the second determination means divides the extraction region into 3×

3 regions and extracts nine images to be processed.

4. The image recognition device according to claim 2,

wherein the first determination means determines whether there is the surrounding image in the image information or not by using a peak value of similarity obtained by checking respective images to have pre-processing with a surrounding image template.

5. The image recognition device according to claim 1,

wherein the second determination means determines whether there is the specific image in the image information or not by using a peak value of similarity obtained by checking respective images to have pre-processing with the specific image template.

6. The image recognition device according to claim 1, further comprising:

a third determination means for determining a color of the specific image when the first determination means determines that the surrounding image is included in the image information and/or the second determination means determines that there is the specific image in the image information.

7. The image recognition device according to claim 1,

wherein the second determination means checks up one image to be processed positioned at the center of the extraction region first.

8. The image recognition device according to claim 7,

wherein the second determination means checks respective images to be processed while relatively rotating respective images to be processed and the specific image template, and

wherein the second determination means, after checking up one image to be processed positioned at the center of the extraction region, checks up other images to be processed based on a rotation angle determined to correspond.

9. The image recognition device according to claim 1,

wherein the first determination means and/or the second determination means performs smoothing processing with respect to the image information in advance.

10. The image recognition device according to claim 1, further comprising:

a first image processing means provided at the downstream of the reading means, including, in order from the upstream, an A/D converter, a D-range correction unit, a reading gamma correction unit, a variable magnification processing unit, a filtering unit, a masking unit, a GCR unit, a recording gamma correction unit, a binarization unit, and a recording unit; and

a second image processing means provided in the downstream of the position of the D-range correction unit of the first image processing means and parallel to the position from the reading gamma correction unit until the recording unit in the first image processing means including, in order from the upstream, a gamma correction unit, a variable magnification processing unit, a filtering unit, GCR unit and a monochrome unit.

11. The image recognition device according to claim 10,

wherein the binarization unit of the second image processing means performs binarization according to luminance of input RGB values, and conversion to luminance is performed by weighting attaching importance to a color which is often used for a specific mark corresponding to the specific image is performed.

12. A copy apparatus, comprising:

an image recognition device including an image reading means for reading image information from a manuscript and discriminating a specific image in which a surrounded image is surrounded by a surrounding image from the image information; and

a printing means for performing printing based on the image information read by the image reading means of the image recognition device, and

wherein the image recognition device includes

a first determination means for determining whether the surrounding image is included in the image information or not; and

a second determination means for determining whether there is the specific image in the image information or not by extracting n×

n images to be processed by dividing an extraction region positioned at the center of the surrounding image into n×

n regions (“

n”

is an odd-number of three or more) when the first determination means determines that the surrounding image is included in the image information, then, by checking respective images to be processed with a specific image template.

13. An image recognition method reading image information from a manuscript and discriminating a specific image in which a surrounded image is surrounded by a surrounding image from the image information, comprising the steps of:

determining whether the surrounding image is included in the image information;

extracting n×

n images to be processed by dividing an extraction region positioned at the center of the surrounding image into n×

n regions (“

n”

is an odd-number of three or more) when determined that the surrounding image is included in the image information, and

determining whether there is the specific image in the image information or not by checking respective images to be processed with a specific image template.