Image processing system and method with dynamically controlled pixel processing

• US 20050157943A1
• Filed: 01/13/2005
• Published: 07/21/2005
• Est. Priority Date: 01/16/2004
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: Enuclia Semiconductor, Inc.
• Inventors: Ruggiero, Carl J.
• Granted Patent: US 7,308,159 B2
• Time in Patent Office: Days
• US Class Current: 382/276
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining
• CPC Class Codes: G06T 3/4007 based on interpolation, e.g. bilinear interpolation image demosaicing G06T3/4015; edge-driven or edge-based scaling G06T3/403
• CPC Class Codes: H04N 5/205 for correcting amplitude versus frequency characteristic
• CPC Class Codes: H04N 7/012 Conversion between an interlaced and a progressive signal for graphics images G09G2310/0229
• CPC Class Codes: H04N 7/0125 one of the standards being a high definition standard
• CPC Class Codes: H04N 7/0137 dependent on presence/absence of motion, e.g. of motion zones H04N7/014 takes precedence; movement detection in television signals H04N5/144
• CPC Class Codes: H04N 7/0145 the interpolation being class adaptive, i.e. it uses the information of class which is determined for a pixel based upon certain characteristics of the neighbouring pixels

Abstract

A system for processing digital images. The system may included a controller that includes a processor and a memory. The system may also include a plurality of image processing blocks operatively coupled with the controller. Each image processing block can be configured to perform a different image processing operation. The image processing blocks and controller can be interconnected and configured to provide sequential pixel processing, in which each image processing block processes input pixels so as to produce output pixels, with the output pixels of an upstream one of the image processing blocks can be fed forward as the input pixels to a downstream one of the image processing blocks. The system may also include a classification block configured to obtain, for each of the image processing blocks, updated classification data for the input pixels to be applied to the image processing block. Processing at each image processing block can be dynamically controlled based on the updated classification data for the input pixels applied to the image processing block.

Claims

1. A system for processing digital images, comprising:

a controller including a processor and a memory;

a plurality of image processing blocks operatively coupled with the controller, each image processing block being configured to perform a different image processing operation, where the image processing blocks and controller are interconnected and configured to provide sequential pixel processing, in which each image processing block processes input pixels so as to produce output pixels, with the output pixels of an upstream one of the image processing blocks being fed forward as the input pixels to a downstream one of the image processing blocks; and

a classification block configured to obtain, for each of the image processing blocks, updated classification data for the input pixels to be applied to the image processing block, and where processing at each image processing block is dynamically controlled based on the updated classification data for the input pixels applied to the image processing block.

2. The system of claim 1, where for at least one of the image processing blocks, the controller is configured to dynamically control processing by selectively combining multiple different processing techniques associated with such image processing block, where such selective combination is performed based on updated classification data of the input pixels applied to such image processing block.

3. The system of claim 2, where the controller is configured to dynamically control processing occurring at another of the image processing blocks based on the selective combination of processing techniques at the at least one of the image processing blocks.

4. The system of claim 1, where the controller is configured to dynamically vary sharpness controls applied at one of the image processing blocks based on processing techniques applied at another of the image processing blocks.

5. The system of claim 1, where one of the image processing blocks is an image interpolator block configured to change the resolution of the input pixels that are applied to the image interpolator block, and where dynamically controlling the processing at the image interpolator block includes dynamically varying image scaling coefficients used to derive the output pixels of the image interpolator block.

6. The system of claim 1, where the image processing blocks include a de-interlacing block, an image interpolator block configured to effect resolution changes, and a color processing block.

7. The system of claim 6, where processing at the image interpolator block is dynamically controlled in response to processing at the de-interlacing block.

8. The system of claim 6, where processing at the color processing block is dynamically controlled in response to processing at the image interpolator block.

9. The system of claim 1, where the updated classification data includes motion data associated with the input pixels.

10. The system of claim 1, where the updated classification data includes frequency data associated with the input pixels.

11. The system of claim 1, where the updated classification data includes color data associated With the input pixels.

12. A system for processing digital images, comprising:

a controller including a processor and a memory; and

a plurality of image processing blocks operatively coupled with the controller, each image processing block being configured to perform a different image processing operation, where the image processing blocks and controller are interconnected and configured to provide sequential pixel processing, in which each image processing block processes input pixels so as to produce output pixels, with the output pixels of an upstream one of the image processing blocks being fed forward as the input pixels to a downstream one of the image processing blocks,

where the controller is configured to dynamically control the image processing operation performed at one of the image processing blocks based on the image processing operation performed at another of the image processing blocks.

13. The system of claim 12, where the controller is configured to dynamically control the image processing blocks in response to classification data associated with pixels being processed by the image processing blocks, and where the classification data varies as the pixels move from one image processing block to another, and where the controller is configured to be responsive to such variation in the classification data when dynamically controlling the image processing blocks.

14. The system of claim 13, where for at least one of the image processing blocks, the controller is configured to dynamically control processing by selectively combining multiple different processing techniques associated with such image processing block, where such selective combination is performed based on updated classification data of the input pixels applied to such image processing block.

15. The system of claim 12, where one of the image processing blocks is an image interpolator block configured to change the resolution of the input pixels that are applied to the image interpolator block, and where the controller is configured to dynamically vary image scaling coefficients used to derive the output pixels of the image interpolator block, based on the image processing operation performed at another of the image processing blocks.

16. The system of claim 12, where the controller is configured to dynamically vary sharpness controls applied at one of the image processing blocks based on the image processing operation performed at another of the image processing blocks.

17. A method of processing digital images, comprising:

receiving input pixels;

obtaining classification data associated with the input pixels;

performing a first image processing operation on the input pixels, where the first image processing operation is dynamically controlled based on the classification data obtained for the input pixels;

analyzing output pixels output from the first image processing operation to obtain updated classification data associated with the pixels output from the first image processing operation; and

performing a second image processing operation on the pixels output from the first image processing operation, where the second image processing operation is dynamically controlled based on the updated classification data.

18. The method of claim 17, where the first image processing operation and the second image processing operation are dynamically controlled based on motion data associated with pixels being processed by the first image processing operation and the second image processing operation.

19. The method of claim 17, where the first image processing operation and the second image processing operation are dynamically controlled based on frequency data associated with pixels being processed by the first image processing operation and the second image processing operation.

20. The method of claim 17, where the first image processing operation and the second image processing operation are dynamically controlled based on color data associated with pixels being processed by the first image processing operation and the second image processing operation.

21. A method of processing digital images, comprising:

receiving input pixels;

performing a first image processing operation on the input pixels; and

performing a second image processing operation, where the second image processing operation is dynamically controlled based on processing data associated with the first image processing operation, where the processing data includes information about processing methods used to perform the first image processing operation.

22. The method of claim 21, further comprising:

outputting output pixels from the first image processing operation;

obtaining updated classification data associated with the output pixels; and

dynamically controlling the second image processing operation based on the updated classification data.

23. The method of claim 22, where obtaining updated classification data includes obtaining motion data associated with the output pixels.

24. The method of claim 23, where dynamically controlling the second image processing operation includes dynamically varying image scaling coefficients associated with the second image processing operation based on the motion data.

25. The method of claim 23, where dynamically controlling the second image processing operation includes dynamically and selectively combining multiple alternate processing methods associated with the second image processing operation based on the motion data.

26. The method of claim 22, where obtaining updated classification data includes obtaining frequency data associated with the output pixels.

27. The method of claim 22, where obtaining updated classification data includes obtaining color data associated with the output pixels.

28. The method of claim 22, where the updated classification data includes historical data associated with the output pixels.