Device and method for estimating if an image is blurred

• US 8,068,668 B2
• Filed: 07/17/2008
• Issued: 11/29/2011
• Est. Priority Date: 07/19/2007
• Status: Expired due to Fees

• Current Assignee: Nikon Corporation
• Original Assignee: Nikon Corporation
• Inventors: Hong, Li
• Primary Examiner(s): KOZIOL, STEPHEN R
• Application Number:
  US12/527,675
• Publication Number:
  US 20100086206A1
• Time in Patent Office: 1,230 Days
• Field of Search: 382/100, 382/168, 382/276
• US Class Current: 382/168
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/44 using image operators, e.g. filters, edge density metrics or local histograms
• CPC Class Codes: H04N 23/64 Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
• CPC Class Codes: H04N 23/80 Camera processing pipelines; Components thereof

Abstract

A method for estimating if an image (14) is either blurred or sharp includes the steps of (i) determining an image gradient histogram distribution (410) of at least a portion of the image (14), and (ii) comparing at least a portion of the image gradient histogram distribution (410) to a Gaussian model gradient histogram distribution (414) of at least a portion of the image (14) to estimate if the image (14) is either blurred or sharp. In most cases, a sharp image (14) has a relatively large tailed distribution when compared to the Gaussian model, while a blurred image has a relatively small tailed distribution when compared to the Gaussian model.

Claims

1. A method for estimating if an image is blurred or sharp, the method comprising the steps of:

determining an image gradient histogram distribution of at least a portion of the image; and

comparing at least a portion of the image gradient histogram distribution to a Gaussian model of at least a portion of the image to estimate if the image is either blurred or sharp.

2. The method of claim 1 wherein the step of determining includes the steps of dividing the image into a plurality of image regions and determining an image gradient histogram distribution of at least one of the image regions, and wherein the step of comparing includes the step of comparing the image gradient histogram distributions for at least one of the image regions to a Gaussian model for that image region to estimate if the image is either blurred or sharp.

3. The method of claim 2 wherein the step of comparing includes the step of comparing a tail section of the Gaussian model to a tail section of the image gradient histogram distribution.

4. The method of claim 1 wherein the step of determining includes the steps of dividing the image into a plurality of image regions and determining an image gradient histogram distribution for each of the image regions, and wherein the step of comparing includes the step of comparing the image gradient histogram distributions for each of the image regions to its respective Gaussian model to estimate if the image is either blurred or sharp.

5. The method of claim 4 wherein the step of comparing includes the step of comparing a tail section of the Gaussian model to a tail section of the image gradient histogram distribution for each image region.

6. The method of claim 1 further comprising the step of evaluating the image to estimate if there is a possible motion blur direction, and wherein the step of determining an image gradient histogram distribution of at least a portion of the image includes comparing pixels along the possible motion blur direction.

7. The method of claim 1 further comprising the step of evaluating at least a portion of the image to determine if the contrast is above a predetermined contrast threshold.

8. The method of claim 7 further comprising the step of adjusting the contrast for at least a portion of the image if the contrast is determined to be below the predetermined contrast threshold.

9. The method of claim 1 wherein the step of determining an image gradient histogram distribution of at least a portion of the image includes the step of excluding pixels having an intensity above a predetermined intensity threshold.

10. A method for estimating if an image is blurred or sharp, the method comprising the steps of:

blurring the image to create an artificially blurred image;

determining an image gradient histogram distribution for at least a portion of the image;

determining an image gradient histogram distribution for at least a portion of the artificially blurred image; and

comparing at least a portion of the image gradient histogram distribution of the image to at least a portion of the image gradient histogram distribution for the artificially blurred image to compute a distribution probability difference to estimate if the image is either blurred or sharp.

11. A device for estimating if an image is either blurred or sharp, the device comprising:

a control system including a processor that (i) determines an image gradient histogram distribution of at least a portion of the image; and

(ii) compares at least a portion of the image gradient histogram distribution to a Gaussian model of at least a portion of the image to estimate if the image is either blurred or sharp.

12. The device of claim 11 wherein the control system divides the image into a plurality of image regions and determines an image gradient histogram distribution of at least one of the image regions, and wherein the control system compares the image gradient histogram distributions for at least one of the image regions to a Gaussian model for that image region to estimate if the image is either blurred or sharp.

13. The device of claim 12 wherein the control system compares a tail section of the Gaussian model to a tail section of the image gradient histogram distribution.

14. The device claim 11 wherein the control system divides the image into a plurality of image regions and determines an image gradient histogram distribution for each of the image regions, and wherein the control system compares the image gradient histogram distributions for each of the image regions to its respective Gaussian model to estimate if the image is either blurred or sharp.

15. The device of claim 14 wherein the control system compares a tail section of the Gaussian model to a tail section of the image gradient histogram distribution for each image region.

16. The device of claim 11 wherein the control system evaluates the image to estimate if there is a possible motion blur direction, and wherein the control system compares pixels along the possible motion blur direction to determine the image gradient histogram distribution.

17. The device of claim 11 wherein the control system evaluates at least a portion of the image to determine if the contrast is above a predetermined contrast threshold.

18. The device of claim 17 wherein the control system adjusts the contrast for at least a portion of the image if the contrast is determined to be below the contrast predetermined threshold.

19. The device of claim 11 wherein the control system deconvolutes the blurred image to provide an adjusted image.

20. A method for estimating a blur degree of an image, the image having a Gaussian model, the method comprising the steps of:

determining an image gradient histogram distribution of at least a portion of the image; and

comparing at least a portion of the image gradient histogram distribution to at least a portion of the Gaussian model to estimate the blur degree of the image.

21. The method of claim 20 wherein the step of determining includes the steps of dividing the image into a plurality of image regions and determining an image gradient histogram distribution of at least one of the image regions, and wherein the step of comparing includes the step of comparing the image gradient histogram distributions for at least one of the image regions to a Gaussian model for that image region.

22. The method of claim 20 wherein the step of comparing includes the step of comparing a tail section of the Gaussian model to a tail section of the image gradient histogram distribution.

23. The method of claim 20 wherein the step of determining includes the steps of dividing the image into a plurality of image regions and determining an image gradient histogram distribution for each of the image regions, and wherein the step of comparing includes the step of comparing the image gradient histogram distributions for each of the image regions to its respective Gaussian model.

24. The method of claim 23 wherein the step of comparing includes the step of comparing a tail section of the Gaussian model to a tail section of the image gradient histogram distribution for each image region.

25. The method of claim 20 further comprising the step of evaluating the image to estimate if there is a possible motion blur direction, and wherein the step of determining an image gradient histogram distribution of at least a portion of the image includes comparing pixels along the possible motion blur direction.

26. The method of claim 20 further comprising the step of evaluating at least a portion of the image to determine if the contrast is above a predetermined contrast threshold.

27. The method of claim 26 further comprising the step of adjusting the contrast for at least a portion of the image if the contrast is determined to be below the predetermined contrast threshold.

28. The method of claim 20 further comprising the step of determining an image gradient histogram distribution of at least a portion of the image includes the step of excluding pixels having an intensity above a predetermined intensity threshold.