Contour tracing and boundary detection for object identification in a digital image

• US 6,674,904 B1
• Filed: 12/14/1999
• Issued: 01/06/2004
• Est. Priority Date: 12/14/1999
• Status: Expired due to Term

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: McQueen, Ward A.
• Primary Examiner(s): WERNER, BRIAN P
• Application Number:
  US09/460,650
• Time in Patent Office: 1,484 Days
• Field of Search: 382/103, 382/118, 382/104, 382/165, 382/170-173, 382/199, 382/200, 382/205, 382/257, 382/259, 382/263, 382/272, 382/291
• US Class Current: 382/199
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06V 10/449 Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters

Abstract

An edge detection algorithm may use contour tracing followed by boundary detection to improve the accuracy of the detected boundaries and to enable the detection of boundaries in cases where the image is either distorted or subject to the effects of noise. The contour tracing and boundary detection may result in an object identification, which if unsatisfactory, may be retried with different threshold values. A high pass gaussian filter may be applied before tracing contours, resulting in the selection of a band limited set of pixel values which are centered around a value of one half the bit depth.

Claims

1. A method comprising:

receiving digital image information;

tracing the contour of an object depicted in said image information with a threshold value;

detecting the boundary of said contour traced object;

evaluating results of tracing the contour and detecting the boundary; and

adjusting the threshold value based on the results.

2. The method of claim 1 further including setting said threshold value initially as a function of a bit depth of the digital image information.

3. The method of claim 1 further including subjecting the digital image information to high pass gaussian filtering.

4. The method of claim 3 including median filtering said digital image information before high pass gaussian filtering.

5. The method of claim 3 including threshold filtering after high pass gaussian filtering.

6. The method of claim 1 wherein contour tracing includes comparing a value of a target pixel to a corresponding value of a plurality of neighboring pixels.

7. The method of claim 6 including determining whether the target pixel and any of the neighboring pixels have a value above a predetermined threshold value.

8. The method of claim 7 including analyzing said pixels based on color.

9. The method of claim 1 wherein detecting the boundary includes analyzing a set of eight pixels surrounding a target pixel and locating a surrounding pixel which was exposed by the contour trace filter.

10. The method of claim 9 including scanning the values of said surrounding pixels one after another in a predetermined sequence.

11. The method of claim 10 including scanning said pixels in a clockwise arrangement.

12. An article comprising a medium storing instructions that cause a processor-based system to:

trace the contour of an object depicted in image information with a threshold value;

detect the boundary of said contour traced object; and

evaluate results of contour tracing and boundary detection, and adjust the threshold value based on said results.

13. The article of claim 12 further storing instructions that cause a processor-based system to set said threshold value initially as a function of a bit depth of said image information.

14. The article of claim 12 further storing instructions that cause a processor-based system to subject the image information to high pass gaussian filtering.

15. The article of claim 14 further storing instructions that cause a processor-based system to median filter said image information before high pass gaussian filtering.

16. The article of claim 14 further storing instructions that cause a processor-based system to threshold filter said image information after high pass gaussian filtering.

17. The article of claim 12 further storing instructions that cause a processor-based system to compare a value of a target pixel to a corresponding value of a plurality of neighboring pixels.

18. The article of claim 17 further storing instructions that cause a processor-based system to determine whether the target pixel is below a predetermined threshold value and any of the neighboring pixel values are above the predetermined threshold value.

19. The article of claim 18 further storing instructions that cause a processor-based system to analyze said pixels based on color.

20. The article of claim 18 further storing instructions that cause a processor-based system to analyze a set of eight pixels surrounding a target pixel and locate a surrounding pixel which exceeds a predetermined value.

21. The article of claim 20 further storing instructions that cause a processor-based system to scan the values of said surrounding pixels one after another in a predetermined sequence.

22. The article of claim 21 further storing instructions that cause a processor-based system to scan said pixels in a clockwise order.

23. The article of claim 22 further storing instructions that cause a processor-based system to scan said pixels beginning from a point counterclockwise from the direction in which a previous pixel exceeded the predetermined threshold value.

24. A digital image processing system comprising:

a processor; and

a storage coupled to said processor, said storage storing instructions that cause a processor-based system to trace a contour of an object depicted in digital image information with a threshold value;

detect the boundary of said contour traced object;

evaluate the results of contour tracing and boundary detection; and

adjust the threshold value based on said results.

25. The digital image processing system of claim 24, wherein the storage further stores instructions to cause the processor-based system to subject the digital image information to high pass gaussian filtering.