Image recognition method and image recognition apparatus

• US 7,751,610 B2
• Filed: 03/11/2005
• Issued: 07/06/2010
• Est. Priority Date: 03/12/2004
• Status: Expired due to Fees

• Current Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Original Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Inventors: Takarada, Shinichi
• Primary Examiner(s): Bali; Vikkram
• Assistant Examiner(s): ENTEZARI, MICHELLE M
• Application Number:
  US11/076,860
• Publication Number:
  US 20050201622A1
• Time in Patent Office: 1,943 Days
• Field of Search: None
• US Class Current: 382/144
• CPC Class Codes: G06V 10/758 Involving statistics of pixels or of feature values, e.g. histogram matching

Abstract

Provided is an image recognition method in which a first pattern area which is set inside a target to be recognized and a second pattern area which is set outside the target are used as a mask pattern. When a comparison circuit judges that a difference obtained by subtracting, by a differential circuit, the largest luminance value in the target image contained in the first pattern area, which is detected by a largest value detection circuit, from the smallest luminance value in the target image contained in the second pattern area, which is detected by the smallest value detection circuit, is larger than a certain offset amount, the matching judgement that a predetermined pattern is present is made. This recognition judgment is performed with the whole target image being scanned.

Claims

1. An image recognition method in which a mask pattern indicating characteristics of an image pattern to be recognized is used to recognize an isolated image isolated from target image data having a plurality of images to be recognized, the target image data having digitized gradations of two or more values and a same configuration and size, said method comprising:

generating, by a mask pattern generation unit, the mask pattern to have a first pattern area which is smaller than the image to be recognized and is not positioned in a center portion, and a second pattern area which is positioned outside the first pattern area, is larger than the image to be recognized, and is smaller than a predetermined distance to adjacent other images to be recognized;

recognizing, by a mask pattern generation unit, an area surrounded by the first pattern area as a fixed area, and an area sandwiched between the first pattern area and the second pattern area as a binarized area;

obtaining, by a largest value detection circuit, a first representative value, which is a largest value of a luminance of a target image data, from the first pattern area;

obtaining, by a smallest value detection circuit, a second representative value, which is a smallest value of a luminance of the target image data, from the second patter area;

judging, by a comparison circuit, that an image to be recognized is present in the target image when a difference between the second representative value and the first representative value is equal to or larger than a predetermined threshold value; and

comparing, by a binarization circuit, the target image data in the binarized area with a predetermined threshold value to binarize them when an image to be recognized is judged as being present in the target image.

2. The image recognition method as defined in claim 1, wherein the predetermined threshold value is an intermediate value between the representative value of the luminance of the target image data in the first pattern area and the representative value of the luminance of the target image data in the second pattern area.

3. The image recognition method as defined in claim 1, wherein the predetermined threshold value is an intermediate value between a largest value of luminances of the target image data, which are contained in the first pattern area, and a smallest value of luminances of the target image data, which are contained in the second pattern area.

4. The image recognition method as defined in claim 1, comprising:

binarizing the target image data in the fixed value area so that the target image data have a predetermined one of two values when the image to be recognized is judged as being present in the target image.

5. The image recognition method as defined in claim 1, wherein a portion of the target image data which is not binarized is set to a predetermined one of two values to obtain a binarization result of the whole image.

6. An image recognition apparatus which uses the image recognition method as defined in claim 1 to process an image, comprising:

a display unit for displaying the first pattern area and the second pattern area on a screen; and

a position specification unit for determining positions of respective pixels in the first pattern area and the second pattern area on the screen.

7. The image recognition apparatus as defined in claim 6, comprising:

an edition unit for overlay-displaying the first pattern area and the second pattern area on a displayed image of the target image data to edit the first pattern area and the second pattern area when the first pattern area and the second pattern area are determined.

8. The image recognition apparatus as defined in claim 7, wherein the target image data are displayed as luminance components, and the first pattern area and the second pattern area are displayed as color difference components different from each other.

9. The image recognition apparatus as defined in claim 6, comprising:

an edition unit for overlay-displaying the first pattern area, the second pattern area, and a binarized area on a displayed image of the target image data to edit the first pattern area, the second pattern area, and the binarized area, when the first pattern area, the second pattern area, and the binarized area are determined.

10. The image recognition apparatus as defined in claim 9, wherein the target image data are displayed as luminance components, and the first pattern area, the second pattern area, and the binarized area are displayed as color difference components different from each other.

11. The image recognition apparatus as defined in claim 6, comprising:

an edition unit for overlay-displaying the first pattern area, the second pattern area, a binarized area, and a fixed value area on a displayed image of the target image data to edit the first pattern area, the second pattern area, the binarized area, and the fixed value area, when the first pattern area, the second pattern area, the binarized area, and the fixed value area are determined.

12. The image recognition apparatus as defined in claim 11, wherein the target image data are displayed as luminance components, and the first pattern area, the second pattern area, the binarized area, and the fixed value area are displayed as color difference components different from each other.

13. The image recognition apparatus as defined in claim 6, wherein the first pattern area and the second pattern area are recorded as relative positions with respect to a pixel to be noted.

14. The image recognition apparatus as defined in claim 6, wherein the first pattern area, the second pattern area, and a binarized area are recorded as relative positions with respect to a pixel to be noted.

15. The image recognition apparatus as defined in claim 6, wherein the first pattern area, the second pattern area, a binarized area, and a fixed value area are recorded as relative positions with respect to a pixel to be noted.