Automatically suggesting regions for blur kernel estimation

Patent metadata

Field	Value
Publication number	US-9779489-B2
Application number	US-201615161073-A
Country	US
Kind code	B2
Filing date	May 20, 2016
Priority date	Oct 23, 2013
Publication date	Oct 3, 2017
Grant date	Oct 3, 2017

Abstract

Embodiments of the present invention provide systems, methods, and computer storage media directed towards automatic selection of regions for blur kernel estimation. In one embodiment, a process divides a blurred image into a regions. From these regions a first region and a second region can be selected based on a number of edge orientations within the selected regions. A first blur kernel can then be estimated based on the first region and a second blur kernel can be estimated for the second region. The first and second blur kernel can then be utilized to respectively deblur a first and second portion of the image to produce a deblurred image. Other embodiments may be described and/or claimed.

First claim

The invention claimed is:

1. One or more computer-readable media having instructions stored thereon, which, when executed by one or more processors of a computing system cause the computing system to:
receive a blurred image having varying levels of blur across the blurred image;
divide the blurred image into a plurality of regions;
select a first region and a second region from the plurality of regions based on a number of edge orientations within the first region and the second region;
estimate a first blur kernel to deblur a first portion of the blurred image based on the first region;
estimate a second blur kernel deblur a second portion of the blurred image based on the second region; and
deblur the first portion of the blurred image using the first blur kernel and the second portion of the blurred image using the second blur kernel to produce a deblurred image.

2. The one or more computer-readable media of claim 1 , wherein to produce the deblurred image, the instructions further cause the computing system to blend results of the deblur in an area located between the first portion and the second portion.

3. The one or more computer-readable media of claim 1 , wherein to produce the deblurred image the instructions further cause the computing system to:
assign a first weight to a pixel based on a distance of the pixel from a center of the first portion;
assign a second weight to the pixel based on a distance of the pixel from a center of the second portion;
determine a blended color for the pixel based on the first weight and the second weight.

4. The one or more computer-readable media of claim 3 , wherein to produce the deblurred image the instructions further cause the computing system to:
determine a first color of the pixel that results from application of the first blur kernel to the pixel; and
determine a second color of the pixel that results from application of the second blur kernel to the pixel, and wherein to determine a blended color of the pixel is further based on the first color and the second color.

5. The one or more computer-readable media of claim 4 , wherein to determine the blended color of the pixel is based on a weighted average of the first color and the second color in accordance with the first weight and the second weight, respectively.

6. The one or more computer-readable media of claim 3 , wherein the pixel is located between the first portion and the second portion.

7. The one or more computer-readable media of claim 3 , wherein the pixel is located in an overlapping area that falls within the first portion and the second portion.

8. The computer-implemented method of claim 1 , wherein to produce the deblurred image, the instructions further cause the computing system to: apply a tiled based blending technique to pixels located between the first portion and the second portion.

9. A computer-implemented method, comprising:
receiving a blurred image having varying levels of blur across the blurred image;
dividing the blurred image into a plurality of regions;
selecting a first region and a second region from the plurality of regions based on a number of edge orientations within the first region and the second region;
estimating a first blur kernel for deblurring a first portion of the blurred image based on the first region;
estimating a second blur kernel for deblurring a second portion of the blurred image based on the second region; and
producing a deblurred image by deblurring the first portion of the blurred image using the first blur kernel and the second portion of the blurred image using the second blur kernel.

10. The computer-implemented method of claim 9 , wherein producing the deblurred image further comprises:
positioning the first blur kernel at a first location with respect to the first portion of the blurred image; and
positioning the second blur kernel at a second location with respect to the second portion of the blurred image and the first location.

11. The computer-implemented method of claim 9 , wherein producing the deblurred image further comprises:
selecting the first blur kernel as a base kernel based on a distance from a center of the first portion to a point within the blurred image;
in response to the selecting of the first blur kernel, positioning the first blur kernel at a first location within the blurred image; and
positioning the second blur kernel at a second location within the blurred image, the second location selected based on the first location.

12. The computer implemented method of claim 11 , wherein the point within the blurred image is a center of the blurred image.

13. The computer implemented method of claim 11 , wherein the point within the blurred image is a focal point of the blurred image.

14. The computer implemented method of claim 13 , wherein the focal point of the blurred image is identified from metadata associated with the blurred image.

15. The computer-implemented method of claim 9 , wherein selecting the first region and the second region from the plurality of regions is further based on a number of over-exposed or under-exposed pixels occurring within the first region and the second region.

16. The computer-implemented method of claim 9 , wherein selecting a first and second region from the plurality of regions is further based on a gradient magnitude of pixels within the the first region and the second region.

17. The computer-implemented method of claim 9 , wherein selecting a first and second region from the plurality of regions is further based on one or more weights associated with each pixel within the first region and the second region.

18. A computing system comprising:
one or more processors; and
memory, coupled with the one or more processors, having instructions stored thereon, which, when executed by the one or more processors provide the computing system with an image deblurring engine to:
receive a blurred image having varying levels of blur across the blurred image;
divide the blurred image into a plurality of regions;
select a first region and a second region from the plurality of regions based on a number of edge orientations within the first region and the second region;
estimate a first blur kernel to deblur a first portion of the blurred image based on the first region;
estimate a second blur kernel deblur a second portion of the blurred image based on the second region; and
deblur the first portion of the blurred image using the first blur kernel and the second portion of the blurred image using the second blur kernel to produce a deblurred image.

19. The one or more computer-readable media of claim 1 , wherein to produce the deblurred image the instructions further cause the computing system to:
assign a first weight to a pixel based on a distance of the pixel from a center of the first portion;
assign a second weight to the pixel based on a distance of the pixel from a center of the second portion;
determine a first color of the pixel that results from application of the first blur kernel to the pixel;
determine a second color of the pixel that results from application of the second blur kernel to the pixel, and wherein to determine a blended color of the pixel is further based on the first color and the second color;
determine a blended color for the pixel based on based on a weighted average of the first color and the second color in accordance with the first weight and the second weight, respectively.

20. The one or more computer-readable media of cla

Assignees

• Adobe Systems Inc

Inventors

• Cho Sunghyun
• Wang Jue
• Chien Jen-Chan
• Feng Dong

Classifications

• G06T11/10

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T2207/20201

  Motion blur correction · CPC title

• G06T2207/20192

  Edge enhancement; Edge preservation · CPC title

• G06T2207/20221

  Image fusion; Image merging · CPC title

• G06T2207/10024

  Color image · CPC title