Detection of a pattern in a digital image

• US 7,116,824 B2
• Filed: 10/23/2002
• Issued: 10/03/2006
• Est. Priority Date: 10/23/2001
• Status: Active Grant

• Current Assignee: Stmicroelectronics France (STMicroelectronics NV)
• Original Assignee: STMicroelectronics SA (STMicroelectronics NV)
• Inventors: Plaza, Laurent
• Primary Examiner(s): MARIAM, DANIEL G
• Application Number:
  US10/278,626
• Publication Number:
  US 20030091238A1
• Time in Patent Office: 1,441 Days
• Field of Search: 382/103, 382/173, 382/181, 382/195, 382/209, 382/218, 382/224, 382/232, 382/250
• US Class Current: 382/209
• CPC Class Codes: G06V 10/507 Summing image-intensity values; Histogram projection analysis
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors
• CPC Class Codes: G06V 10/758 Involving statistics of pixels or of feature values, e.g. histogram matching
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

A method for detecting a pattern in a digital image by successive comparison of the content of current windows with the content of at least one reference window representative of the pattern, the reference window containing values of variance or standard deviation of sub-blocks of pixels of the pattern and the current windows containing values of variance or standard deviation of sub-blocks of pixels of the image, the sub-blocks being of the same size in the current and reference windows.

Claims

1. A method for detecting a pattern in a digital image by successive comparison of the content of current windows with the content of at least one reference window representative of the pattern, the reference window containing values of variance or standard deviation of sub-blocks of pixels of the pattern and the current windows containing values of variance or standard deviation of sub-blocks of pixels of the image, the sub-blocks being of the same size in the current and reference windows.

2. The detection method of claim 1, including the steps of:

dividing the image and the pattern into sub-blocks of same dimensions;

forming a reference window containing the variances or standard deviations of the levels of grey of the pixels of each sub-block of the pattern; and

successively comparing the values of the reference window with values of at least one current window, of same size as the reference window, and containing the variances or standard deviations of the levels of grey of pixels of sub-blocks of the image.

3. The method of claim 2, wherein the values contained in the current windows are calculated along a scanning of the image.

4. The method of claim 2, wherein the values of the current windows are all calculated before performing the comparisons.

5. The method of claim 2, wherein the steps of division, formation, and comparison are performed for several sizes of sub-blocks.

6. A system for detecting a pattern in a digital image, according to the method of claim 1.

7. A method for detecting a pattern in a digital image, comprising:

defining a reference window containing variance or standard deviation values representative of sub-blocks of the pattern;

defining a current window of equal size as the reference window, the current window containing variance or standard deviation values representative of sub-blocks of the image, the sub-blocks of the pattern and the sub-blocks of the image having equal dimensions; and

successively comparing the values of the sub-blocks of the reference window with the values of the sub-blocks of the current window.

8. A method as defined in claim 7, wherein comparing the values of the sub-blocks of the reference window with the values of the sub-blocks of the current window is performed for multiple current windows of the image, further comprising judging detection of the pattern in the digital image based on the comparison.

9. A method as defined in claim 7, further comprising dividing the image into sub-blocks to form a reduced image and scanning the reduced image with the current window.

10. A method as defined in claim 7, further comprising scanning the image with the current window and calculating the variance or standard deviation values of sub-blocks of the current window at each location of the current window during scanning.

11. A method as defined in claim 7, wherein the variance or standard deviation values are representative of levels of gray in the sub-blocks.

12. A method as defined in claim 7, wherein defining a reference window, defining a current window, and comparing the values of the sub-blocks of the reference window with values of the sub-blocks of the current window are repeated for sub-blocks of different sizes.