Visually comfortable content-adaptive dithering for super-resolution

• US 9,600,858 B1
• Filed: 10/17/2013
• Issued: 03/21/2017
• Est. Priority Date: 10/17/2013
• Status: Active Grant

• Current Assignee: Pixelworks Incorporated
• Original Assignee: Pixelworks Incorporated
• Inventors: Liu, Guodong, Zhang, Bob, Ma, Yue, Chen, Junhua
• Primary Examiner(s): Sherali, Ishrat I
• Application Number:
  US14/056,768
• Time in Patent Office: 1,251 Days
• Field of Search: 382/254, 382/274, 382/299
• US Class Current: 1/1
• CPC Class Codes: G06T 3/4069 by subpixel displacements

Abstract

A method of producing a high resolution image includes generating high resolution image data for a high resolution image using a processor, wherein the high resolution image has a resolution higher than a resolution of an original image, determining image content of the high resolution image data, and selectively adding dither noise to the high resolution image data based upon the image content of the high resolution image data to produce final image data.

Claims

1. A method of producing a high resolution image, comprising:

generating high resolution image data for a high resolution image using a processor, wherein the high resolution image has a resolution higher than a resolution of an original image;

determining image content of the high resolution image data on a block basis, where the image is divided into blocks of multiple pixels;

weighting each block of image data based upon the image content;

decomposing the weights for each block to the pixels of the block;

selectively adding dither noise to the pixels of the high resolution image data based upon the weight for each block to produce final image data.

2. The method of claim 1, wherein generating the high resolution image data comprises applying one of one-dimensional upscaling.

3. The method of claim 1, wherein generating the high resolution image data comprises generating super resolution image data by one of single frame super resolution, multi-frame super resolution, or self-similarity super resolution to original image data.

4. The method of claim 1, wherein determining image content comprises analyzing the image data to determine at least one of image texture and brightness.

5. The method of claim 1, wherein weighting each block of for image data comprises assigning a higher weight to image data having more texture than image data with less texture.

6. The method of claim 1, wherein weighting each block of image data comprises assigning a lower weight to image data having high or low brightness and assigning a higher weight to image data having a moderate brightness.

7. The method of claim 1, further comprising generating dither noise.

8. The method of claim 7, wherein generating dither noise comprises modifying a pre-existing pattern of dither noise.

9. The method of claim 7, wherein generating dither noise comprises applying a high pass filter to random noise.

10. The method of claim 7, wherein generating dither noise comprises producing a halftoned image and overlaying random noise on the half-toned image.

11. The method of claim 7, wherein generating dither noise further comprises enhancing dither noise and/or resampling dither noise.

12. An apparatus, comprising:

an image processor configured to receive original image data at a first resolution and generate super resolution image data at a second resolution higher than the first resolution;

a post processor configured to receive the super resolution image data, divide the super resolution image data into block, weight each block depending upon image data, decompose the weights for the blocks to pixels, and apply dither noise according to the weights to the super resolution data to produce improved super resolution image data.

13. The apparatus of claim 12, wherein the image processor and the post processor are portions of the same processor.

14. The apparatus of claim 12, wherein the image processor and the post processor are separate processors.