Technique for performing error diffusion

• US 7,440,632 B2
• Filed: 05/06/2005
• Issued: 10/21/2008
• Est. Priority Date: 05/06/2005
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Ranganathan, Sridharan
• Primary Examiner(s): Do; Anh Hong
• Application Number:
  US11/123,593
• Publication Number:
  US 20060250651A1
• Time in Patent Office: 1,264 Days
• Field of Search: 382/252, 382/237, 382/239, 382/272, 358/3.03, 358/1.9, 358/3.04, 358/3.05, 358/466, 347/2, 347/67, 347/16, 347/14
• US Class Current: 382/252
• CPC Class Codes: H04N 1/4052 by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions

Abstract

A technique to perform error distribution in a graphics system. More particularly, embodiments of the invention include at least one technique to distribute the error resulting from converting computer generated graphics pixels from one pixel depth to another.

Claims

1. An apparatus comprising:

error diffusion acceleration logic to perform at least one function associated with distributing pixel state error among at least one neighboring pixel within a digitally-represented image;

a processing element coupled to the error diffusion logic to perform user-prescribed pixel error diffusion acceleration functions in conjunction with the at least one function.

2. The apparatus of claim 1 wherein the pixel state error is to be incurred as a result of converting between pixel component depths.

3. The apparatus of claim 1 wherein the at least one function includes pixel error calculation and distribution to the at least one neighboring pixel and either determining a pixel error threshold value or performing a thresholding function.

4. The apparatus of claim 1 wherein the user-prescribed functions include producing an output pixel to be applied to the digitally-represented image.

5. The apparatus of claim 3, wherein the at least one function includes producing an intermediate pixel value, the intermediate pixel value being a function of an initial pixel value and the error distributed to the neighboring pixels.

6. The apparatus of claim 5 wherein the user-prescribed functions include generating a modified intermediate pixel value.

7. A processor comprising:

error diffusion acceleration logic to determine an intermediate pixel state value;

a processing element to perform user-generated instructions, which if executed by the processing element, are to perform any of a group of functions consisting of;

determining a threshold value, a thresholding operation, and determining a modified intermediate pixel state value using the intermediate pixel state value.

8. The processor of claim 7 wherein the error diffusion acceleration logic is to determine the threshold value or perform the thresholding operation, respectively, only if the user-generated instructions do not perform either of these functions.

9. The processor of claim 7 wherein the error diffusion acceleration logic comprises an error buffer to store an input pixel and neighboring error values associated with a plurality of pixels neighboring the input pixel.

10. The processor of claim 9 wherein each of the neighboring error values are to be derived from a cumulative weighted error value of preceding pixel error values.

11. The processor of claim 10 further comprising a circular error buffer to temporarily store the preceding pixel error values until they can be restored to the error buffer.

12. The processor of claim 9 wherein the error buffer is to store pixel error values according to a rasterization pattern that is to traverse an image, in which the pixels are displayed, in a top-to-bottom and left-to-right path.

13. The processor of claim 12 wherein the circular error buffer is stored within a memory controller hub.

14. The processor of claim 7 wherein the processor is an image signal processor.

15. A system comprising:

error diffusion acceleration logic to generate an intermediate pixel value;

a memory unit coupled to the error diffusion acceleration logic to store a set of third-party vendor instructions, which if executed by a processor perform a method comprising;

determining a pixel error threshold value if the pixel error threshold value is not to be determined by the error diffusion acceleration logic;

comparing the intermediate pixel value to the threshold value if the error diffusion acceleration logic is not to compare the intermediate pixel value to the threshold value;

generating a modified intermediate pixel value;

generating an output pixel.

16. The system of claim 15 wherein the error diffusion accelerator is to generate a pixel error value by comparing the threshold value with either the intermediate pixel value or the modified intermediate pixel value.

17. The system of claim 15 further comprising an audio device coupled to the memory unit.

18. The system of claim 16 wherein the memory unit is within a processing element, the processing element further comprising an arithmetic logic unit to execute the third-party vendor instructions.

19. The system of claim 16 wherein the pixel error value is due to a conversion of components of the pixel from one pixel depth to another.

20. A method comprising:

converting pixel components of a digitally-represented image from a first depth to a second depth;

traversing the digitally-represented image;

storing pixel errors temporarily into a first-in-first-out buffer until pixels to which the pixel errors correspond are traversed;

performing a first set of error diffusion operations within an error diffusion accelerator and performing a second set of error diffusion operation within a processing element performing user-generated error diffusion instructions.

21. The method of claim 20 further comprising choosing between a set of results from the first and second set of operations.

22. The method of claim 21 wherein the first and second set of operations include any of a group consisting of:

determining an error threshold value, a thresholding operation, generating an intermediate value.

23. The method of claim 22 wherein the set of results include any of a group consisting of:

the error threshold value, the result of thresholding operation, the intermediate value.

24. The method of claim 23 wherein the choosing is to result in the same choice of results among the set of results for all pixels traversed within the digitally-represented image.

25. The method of claim 24 further comprising generating an error to be distributed among a plurality of cumulative error values, the cumulative error values corresponding to neighboring pixels of the converted pixels.