Local contrast enhancement of images

• US 8,238,687 B1
• Filed: 09/23/2008
• Issued: 08/07/2012
• Est. Priority Date: 01/09/2008
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Helwett-Packard Development Company, L.P.
• Inventors: Lim, Suk Hwan, Kisilev, Pavel
• Primary Examiner(s): Kassa, Yosef
• Application Number:
  US12/236,261
• Time in Patent Office: 1,414 Days
• Field of Search: 382/260, 382/274, 382/275, 358/3.26, 358/3.27, 358/463
• US Class Current: 382/274
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement

Abstract

A set of local noise variation values characterizing local noise levels in the input image is calculated. Local contrast enhancement is performed on image forming element values derived from the input image to produce a contrast enhanced image. Values of local noise amplification caused by the local contrast enhancement are ascertained. Noise in the contrast enhanced image is reduced based on the set of local noise variation values and the local noise amplification values.

Claims

1. A machine-implemented method of processing an input image, comprising:

calculating a set of local noise variation values characterizing local noise levels in the input image;

performing local contrast enhancement on image forming element values derived from the input image to produce a contrast enhanced image;

ascertaining values of local noise amplification caused by the local contrast enhancement; and

reducing noise in the contrast enhanced image based on the set of local noise variation values and the local noise amplification values.

2. The method of claim 1, wherein the local noise variation values correspond to values of variance of local noise in the input image.

3. The method of claim 1, wherein the local noise variation values correspond to values of standard deviation of local noise in the input image.

4. The method of claim 1, wherein the ascertaining comprises determining each of the local noise amplification values based on a respective measure of rate of change in the contrast enhancement as a function of image intensity at a respective local input image intensity value.

5. The method of claim 1, wherein the input image comprises image forming elements, each of the image forming elements has at least one respective value, and the ascertaining comprises determining each of the local noise amplification values based on a respective measure of rate of change of the contrast enhancement as a function of image forming element values at a respective one of the image forming elements.

6. The method of claim 1, wherein the reducing comprises determining a second set of local noise variation values characterizing local noise levels in the contrast enhanced image based on the first set of local noise variation values and the local noise amplification values, and the reducing is based on the second set of local noise variation values.

7. The method of claim 6, wherein the determining comprises for each of the local noise variation values in the second set evaluating a function of a corresponding one of the local noise variation values in the first set and a corresponding one of the local noise amplification values.

8. The method of claim 7, wherein the evaluating comprises for each of the local noise variation values in the second set multiplying the corresponding local noise variation value in the first set and the corresponding local noise amplification value.

9. The method of claim 6, wherein the reducing comprises applying a denoising filter to the contrast enhanced image based on the local noise variation values in the second set.

10. The method of claim 9, wherein the applying comprises selectively applying the denoising filter to local areas of the contrast enhanced image in which respective differences between values of local intensity variation in the local areas and the corresponding local noise variation values in the second set satisfy a maximum threshold predicate, and omitting the application of the denoising filter in local areas of the contrast enhanced image in which respective differences between values of local intensity variation in the local areas and the corresponding local noise variation values in the second set fail to satisfy the maximum threshold predicate.

11. The method of claim 1, further comprising blurring the input image to produce a blurred image, and wherein the performing comprises performing the local contrast enhancement on the blurred image.

12. The method of claim 1, wherein the input image comprises image forming elements, each of the image forming elements has at least one respective value, and the calculating, the performing, the ascertaining, and the reducing are performed for each of the image forming elements of the input image.

13. Apparatus for processing an input image, comprising:

a memory; and

a processing unit coupled to the memory and operable to perform operations comprising

calculating a set of local noise variation values characterizing local noise levels in the input image;

performing local contrast enhancement on image forming element values derived from the input image to produce a contrast enhanced image;

ascertaining values of local noise amplification caused by the local contrast enhancement; and

reducing noise in the contrast enhanced image based on the set of local noise variation values and the local noise amplification values.

14. The apparatus of claim 13, wherein the processing unit is operable to perform operations comprising determining each of the local noise amplification values based on a respective measure of rate of change in the contrast enhancement as a function of image intensity at a respective local input image intensity value.

15. The apparatus of claim 13, wherein the processing unit is operable to perform operations comprising determining a second set of local noise variation values characterizing local noise levels in the contrast enhanced image based on the first set of local noise variation values and the local noise amplification values, and the reducing is based on the second set of local noise variation values.

16. The apparatus of claim 15, wherein in the determining the processing unit is operable to perform operations comprising for each of the local noise variation values in the second set evaluating a function of a corresponding one of the local noise variation values in the first set and a corresponding one of the local noise amplification values.

17. The apparatus of claim 15, wherein the processing unit is operable to perform operations comprising applying a denoising filter to the contrast enhanced image based on the local noise variation values in the second set.

18. The apparatus of claim 17, wherein in the applying the processing unit is operable to perform operations comprising selectively applying the denoising filter to local areas of the contrast enhanced image in which respective differences between values of local intensity variation in the local areas and the corresponding local noise variation values in the second set satisfy a maximum threshold predicate, and omitting the application of the denoising filter in local areas of the contrast enhanced image in which respective differences between values of local intensity variation in the local areas and the corresponding local noise variation values in the second set fail to satisfy the maximum threshold predicate.

19. The apparatus of claim 13, wherein the processing unit is operable to perform operations comprising blurring the input image to produce a blurred image, and in the performing the processing unit is operable to perform operations comprising performing the local contrast enhancement on the blurred image.

20. A non-transitory machine readable medium for arranging graphic objects on a page, the machine readable medium storing machine-readable instructions causing a machine to perform operations comprising:

calculating a set of local noise variation values characterizing local noise levels in the input image;

performing local contrast enhancement on image forming element values derived from the input image to produce a contrast enhanced image;

ascertaining values of local noise amplification caused by the local contrast enhancement; and

reducing noise in the contrast enhanced image based on the set of local noise variation values and the local noise amplification values.