Method and apparatus for adaptively sharpening an image

• US 6,640,017 B1
• Filed: 05/26/1999
• Issued: 10/28/2003
• Est. Priority Date: 05/26/1999
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Tsai, Ping-Sing, Tan, Yap-Peng, Acharya, Tinku
• Primary Examiner(s): Lee, Thomas D.
• Assistant Examiner(s): Brinich, Stephen
• Application Number:
  US09/320,192
• Time in Patent Office: 1,616 Days
• Field of Search: 382/260-269, 382/199-200
• US Class Current: 382/266
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening

Abstract

Briefly, in accordance with one embodiment of the invention, a method of sharpening an image includes the following. A crispening parameter is adaptively computed for a captured image based, at least in part, on a measure of the edges and brightness of the captured image. A kernel is applied to the captured image using the adaptively computed crispening parameter.

Claims

1. A method of sharpening an image comprising:

adaptively computing a crispening parameter for at least a portion of a captured image based, at least in part, on a measure of the edges and the brightness of the at least a portion of the captured image, comprising;

computing a luminance value for the at least a portion of the captured image;

computing a gradient magnitude value for the at least a portion of the captured image; and

computing the crispening parameter based, at least in part, on, approximately, the relative values of the computed luminance value and the computed gradient

magnitude value; and

applying a kernel to the at least a portion of the captured image using the adaptively computed crispening parameter.

2. The method of claim 1, wherein adaptively computing a crispening parameter for at least a portion of a captured image comprises adaptively computing a crispening parameter for at least a portion of a captured image based, at least in part, on a measure of the strength of the edges and the degree of brightness of the at least a portion of the captured image.

3. The method of claim 2, wherein the strength of the edges are measured using the gradient magnitude value for the pixels of the at least a portion of the captured image and the degree of brightness is measured using the luminance value for the pixels of the at least a portion of the captured image.

4. The method of claim 3, wherein computing the crispening parameter is based, at least in part, on, approximately, the relative values of the computed luminance value and the computed gradient magnitude value comprises computing the crispening parameter substantially in accordance with the following relationship:

5. The method of claim 3, wherein the computed luminance value comprises a normalized, offset reduced, weighted measure of the luminance values of at least substantially all the image pixels in the at least a portion of the captured image.

6. The method of claim 5, wherein the measure is weighted by computing the mean luminance value of at least substantially all of the image pixels in the at least a portion of the captured image.

7. The method of claim 5, wherein the measure is offset-reduced, from the mean luminance value, subtracting a luminance value that is greater than the 5 percent lowest luminance values of at least substantially all of the image pixels in the at least a portion of the captured image.

8. The method of claim 5, wherein the measure is normalized by dividing by a computed range of luminance values of at least substantially all of the image pixels of the at least a portion in the captured image.

9. The method of claim 8, wherein the computed range of luminance values is substantially in accordance with the following relationship:

10. The method of claim 5, wherein the luminance value is computed substantially in accordance with the following relationship:

11. The method of claim 3, wherein the computed gradient magnitude value for the at least a portion of the captured image comprises a normalized, offset reduced, weighted measure of the gradient magnitude values of at least substantially all the image pixels in the at least a portion of the captured image.

12. The method of claim 11, wherein the measure is weighted by computing the mean gradient magnitude value of at least substantially all of the image pixels in the at least a portion of the captured image.

13. The method of claim 12, wherein the measure is offset reduced by, from the mean gradient magnitude value, subtracting a gradient magnitude value greater than 5 percent of at least substantially of the image pixels in the at least a portion of the captured image.

14. The method of claim 11, wherein the measure is normalized by dividing by a computed range of gradient magnitude values of at least substantially all of the image pixels in the at least a portion of the captured image.

15. The method of claim 8, wherein the computed range of gradient magnitude values is substantially in accordance with the following relationship:

range=G₉₅−

G₅, where G₉₅ is the gradient magnitude value smaller than 95 percent of the gradient magnitude values, and G₅ is the gradient magnitude value greater than 5 percent of the gradient magnitude values, of at least substantially all of the image pixels in the at least a portion of the captured image.

16. The method of claim 11, wherein the gradient magnitude value is computed substantially in a accordance with the following relationship:

17. The method of claim 1, wherein the at least a portion of the captured image comprises an entire captured image.

18. An article comprising:

a storage medium;

said storage medium having stored thereon, instructions, that, when executed, result in the following;

adaptively computing a crispening parameter for at least a portion of a stored image based, at least in part, on a measure of the edges and the brightness of the at least a portion of the stored image, including;

computing a luminance value for the at least of portion of the stored image;

computing a gradient magnitude value for the at least a portion of the stored image; and

computing the crispening parameter based, at least in part, on, approximately, the relative values of the computed luminance value and the computed gradient magnitude value; and

applying a kernel to the at least a portion of the stored image using the adaptively computed crispening parameter.

19. The article of claim 18, wherein the instructions, when executed, further result in adaptively computing a crispening parameter for at least a portion of a stored image based, at least in part, on a measure of the strength of the edges and the degree of brightness of the at least a portion of the stored image.

20. The article of claim 19, wherein the instructions, when executed, result in the strength of the edges being measured using the gradient magnitude value for the pixels of the at least a portion of the stored image and the degree of brightness being measured using the luminance value for the pixels of the at least a portion of the stored image.

21. The article of claim 20, wherein the instructions, when executed, result in computing the crispening parameter substantially in accordance with the following relationship:

22. The article of claim 20, wherein the computed luminance value comprises a normalized, offset reduced, weighted measure of the luminance values of at least substantially all the image pixels in the at least a portion of the stored image.

23. The article of claim 20, wherein the measure is weighted by computing the mean luminance value of at least substantially all of the image pixels in the at least a portion of the stored image.

24. The article of claim 22, wherein the measure is offset reduced by, from the mean luminance value, subtracting a luminance value that is greater than the 5 percent lowest luminance values of at least substantially all of the image pixels in the at least a portion of the stored image.

25. The article of claim 22, wherein the measure is normalized by dividing by a computed range of luminance values of at least substantially all of the image pixels of the at least a portion of the stored image.

26. The article of claim 18, wherein the at least a portion of the stored image comprises an entire stored image.

27. A system comprising:

a digital imager;

said digital imager including the capability to adaptively compute a crispening parameter for at least a portion of a captured image based, at least in part, on a measure of the edges, the brightness of at least a portion of the captured image, including a relative magnitude of a pixel luminance and a pixel gradient magnitude.

28. The system of claim 27, wherein said digital imager further includes the capability to apply a kernel to the at least a portion of the captured image using the adaptively computed crispening parameter.

29. The system of claim 27, and further comprising a computing platform coupled to the digital imager.

30. The system of claim 29, wherein the computing platform comprises a personal computer.

31. The system of claim 30, wherein the personal computer comprises one of a laptop and a desktop computer.

32. The system of claim 27, wherein the at least a portion of the captured image comprises an entire captured image.