IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND IMAGE PROCESSING PROGRAM

• US 20160117807A1
• Filed: 10/28/2015
• Published: 04/28/2016
• Est. Priority Date: 10/28/2014
• Status: Active Grant

• Current Assignee: Samsung Display Company Limited (Samsung Electronics Co. Ltd.)
• Original Assignee: Samsung Display Company Limited (Samsung Electronics Co. Ltd.)
• Inventors: MATSUMOTO, KAZUHIRO
• Granted Patent: US 9,665,929 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 2207/20208 High dynamic range [HDR] image processing
• 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

An image processing device includes a generation unit and a smoothing processor. The generation unit calculates an edge strength on the basis of a stairs tone strength indicative of differences between a first parameter and a second parameter. The first parameter is a first gradient based on pixel values of a pixel of interest in an input image and adjacent pixels included in a first range with respect to the pixel of interest. The second parameter is one of a variance based on pixel values of the pixel of interest and the adjacent pixels, and a second gradient based on the pixel value of the pixel of interest and adjacent pixels included in a second range. The smoothing processor performs a nonlinear smoothing operation on the pixel value of the pixel of interest and the adjacent pixels included in the first range on the basis of the calculated edge strength.

Claims

1. An image processing device comprising:

a generation unit that calculates an edge strength based on a stairs tone strength indicative of differences between a first parameter and a second parameter, wherein the first parameter is a first gradient based on pixel values of a pixel of interest in an input image and of adjacent pixels included in a first range with respect to the pixel of interest, the second parameter is one of a variance based on pixel values of the pixel of interest and of the adjacent pixels and a second gradient based on pixel values of adjacent pixels included in a second range wider than the first range; and

a smoothing processor that performs a nonlinear smoothing operation on the pixel values of the pixel of interest and the adjacent pixels included in the first range, based on the calculated edge strength.

2. The image processing device of claim 1, wherein the generation unit calculates the variance as the second parameter, and calculates the stairs tone strength K_S based on

3. The image processing device of claim 1, wherein the generation unit calculates a standard deviation as the second parameter based on the variance, and calculates the stairs tone strength K_S based on

4. The image processing device of claim 1, wherein the generation unit calculates the variance as the second parameter, and calculates the stairs tone strength K_S based on K_S=k_σ

·

σ

²−

k_∇

·

!²,

wherein ∇

denotes the first parameter, σ

² denotes the second parameter, k_∇

denotes a compensation coefficient for the first parameter, and k_σ

denotes a compensation coefficient for the variance.

5. The image processing device of claim 2, wherein the generation unit calculates the variance based on the pixel of interest and adjacent pixels included in a range wider than the first range.

6. The image processing device of claim 1, wherein the generation unit calculates the first gradient by applying a band limiting filter having a first pass frequency range, and calculates the second gradient by applying a band limiting filter having a second pass frequency range that is wider at a lower band side than the first pass frequency range.

7. The image processing device of claim 1, wherein the generation unit calculates the second gradient as the second parameter, and calculates the stairs tone strength K_S based on

8. The image processing device of claim 1, wherein the generation unit calculates the second gradient as the second parameter, and calculates the stairs tone strength K_S based on K_S=|∇

₂|−

|∇

₁|,

wherein ∇

1 denotes the first parameter, and ∇

2 denotes the second parameter.

9. The image processing device of claim 1, wherein the generation unit calculates the edge strength by multiplying an absolute value of the first parameter by the stairs tone strength.

10. The image processing device of claim 1, wherein the generation unit calculates the edge strength by multiplying a square of the first parameter by the stairs tone strength.

11. The image processing device of claim 1, wherein the smoothing processor determines an ε

value based on the edge strength, and performs the nonlinear smoothing operation based on an adaptive ε

filter according to the ε

value.

12. The image processing device of claim 11, wherein the smoothing processor determines the ε

value based on a result of comparison between the edge strength and a predetermined threshold value.

13. The image processing device of claim 12, wherein the smoothing processor determines the ε

value to be a minimum value if the edge strength exceeds the threshold value, and determines the ε

value to be a maximum value if the edge strength is less than or equal to the threshold value.

14. An image processing method comprising:

calculating an edge strength based on a stairs tone strength indicative of differences between a first parameter and a second parameter, the first parameter being a first gradient based on pixel values of a pixel of interest in an input image and of adjacent pixels included in a first range with respect to the pixel of interest, the second parameter being one of a variance based on pixel values of the pixel of interest and the adjacent pixels and a second gradient based on pixel values of adjacent pixels included in a second range wider than the first range; and

performing a nonlinear smoothing operation on the pixel values of the pixel of interest and the adjacent pixels included in the first range, based on the calculated edge strength.

15. The method of claim 14, wherein the edge strength is calculated from the first parameter and the stairs tone strength, and the method further comprises

determining an value based on the edge strength, wherein the nonlinear smoothing operation is based on an adaptive ε

filter based on the ε

value.

16. The method of claim 14, wherein the variance is the second parameter, and the stairs tone strength is calculated based on one of

17. The method claim 14, wherein the second gradient is the second parameter, and the stairs tone strength is calculated based on one of

18. A non-transitory recording medium for storing a program for instructing a computer to perform a method for processing an image, the method comprising:

calculating an edge strength based on a stairs tone strength indicative of differences between a first parameter and a second parameter, the first parameter being a first gradient based on pixel values of a pixel of interest in an input image and of adjacent pixels included in a first range with respect to the pixel of interest, the second parameter being one of a variance based on pixel values of the pixel of interest and of the adjacent pixels and a second gradient based on pixel values of adjacent pixels included in a second range wider than the first range; and

performing a nonlinear smoothing operation on the pixel values of the pixel of interest and of the adjacent pixels included in the first range, based on the calculated edge strength.

19. The recording medium of claim 18, wherein the edge strength is calculated from the first parameter and the stairs tone strength, and the method further comprises

determining an ε

value based on the edge strength, wherein the nonlinear smoothing operation is based on an adaptive ε

filter based on the ε

value.

20. The recording medium of claim 18, wherein the variance is the second parameter, and the stairs tone strength is calculated based on one of