Image processing apparatus, image processing method, and computer program

• US 8,750,638 B2
• Filed: 07/22/2010
• Issued: 06/10/2014
• Est. Priority Date: 08/07/2009
• Status: Expired due to Fees

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: Hara, Takayuki
• Primary Examiner(s): Kassa, Yosef
• Application Number:
  US13/386,932
• Publication Number:
  US 20120121203A1
• Time in Patent Office: 1,419 Days
• Field of Search: 382/255, 382/260, 382/274, 382/275, 382/286, 358/3.26, 358/3.27, 358/463
• US Class Current: 382/255
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20028 Bilateral filtering
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: H04N 1/58 Edge or detail enhancement; Noise or error suppression, e.g. colour misregistration correction H04N1/62 takes precedence

Abstract

An image processing apparatus includes a distance calculating unit that calculates a distance between first and second distributions, wherein the first distribution is a distribution of pixel values of pixels surrounding a focus pixel included in an image, the second distribution is a distribution of pixel values of pixels surrounding a pixel included in a window that includes the focus pixel at the center, and the second distribution is obtained for each pixel included in the window; a coefficient calculating unit that calculates, using the distance between the first and second distributions, a filter coefficient for each pixel included in the window; and a filter processing unit that calculates a correction value for the focus pixel by performing a filter process on each pixel included in the window using the filter coefficient.

Claims

1. An image processing apparatus comprising:

a distance calculating unit that calculates a distance between first and second distributions, wherein the first distribution is a distribution of pixel values of pixels surrounding a focus pixel included in an image, the second distribution is a distribution of pixel values of pixels surrounding a pixel included in a window that includes the focus pixel at the center, and the second distribution is obtained for each pixel included in the window;

a coefficient calculating unit that calculates, using the distance between the first and second distributions, a filter coefficient for each pixel included in the window; and

a filter processing unit that calculates a correction value for the focus pixel by performing a filter process on each pixel included in the window using the filter coefficient.

2. The image processing apparatus according to claim 1, wherein the distance between the first and second distributions is a distance between an average of the pixel values of the pixels surrounding the focus pixel and an average of the pixel values of the pixels surrounding the pixel included in the window.

3. The image processing apparatus according to claim 2, wherein, when the image is a color image, the averages of the pixel values that are used to calculate the distance between the first and second distributions are a three-dimensional vector, wherein components of the three-dimensional vector are averages of three color components contained in the pixel values.

4. The image processing apparatus according to claim 1, wherein the distance between the first and second distributions is calculated by dividing a distance between the average of the pixel values of the pixels surrounding the focus pixel and the average of the pixel values of the pixels surrounding the pixel included in the window by the sum of the variance of the pixel values of the pixels surrounding the focus pixel and the variance of the pixel values of pixels surrounding the pixel included in the window.

5. The image processing apparatus according to claim 1, wherein the distance between the first and second distributions is Kullback-Leibler information between the distribution of the pixel values of the pixels surrounding the focus pixel and the distribution of the pixel values of the pixels surrounding the pixel included in the window.

6. The image processing apparatus according to claim 1, wherein the distance between the first and second distributions is either (i) a distance between pixels that have pixel values closest to each other, where one pixel is selected from the pixels surrounding the focus pixel and the other pixel is selected from the pixels surrounding the pixel included in the window;

or (ii) a lowest sum of distances, which is obtained by associating pixel values of the pixels surrounding the focus pixel with pixel values of the pixels surrounding the pixel included in the window in a one-on-one manner so that a sum of the distances between each pixel in each pair of pixels has the lowest value.

7. The image processing apparatus according to claim 1, wherein, when the distance between the first and second distributions is more than or equal to a predetermined value, the coefficient calculating unit sets the filter coefficient to 0.

8. The image processing apparatus according to claim 1, wherein the coefficient calculating unit uses a value of a dependent variable as a weighting for calculating the filter coefficient, wherein the dependent variable is a variable of a monotonically decreasing function that contains the distance between the first and second distributions as an independent variable.

9. An image processing method comprising:

calculating a distance between first and second distributions, wherein the first distribution is a distribution of pixel values of pixels surrounding a focus pixel included in an image, the second distribution is a distribution of pixel values of pixels surrounding a pixel included in a window that includes the focus pixel at the center, and the second distribution is obtained for each pixel included in the window;

calculating, using the distance between the first and second distributions, a filter coefficient for each pixel included in the window; and

calculating a correction value for the focus pixel by performing a filter process on each pixel included in the window using the filter coefficient.

10. A non-transitory, computer readable medium encoding a computer program that cause a computer to execute an image processing method, the method comprising:

calculating a distance between first and second distributions, wherein the first distribution is a distribution of pixel values of pixels surrounding a focus pixel included in an image, the second distribution is a distribution of pixel values of pixels surrounding a pixel included in a window that includes the focus pixel at the center, and the second distribution is obtained for each pixel included in the window;

calculating, using the distance between the first and second distributions, a filter coefficient for each pixel included in the window; and

calculating a correction value for the focus pixel by performing a filter process on each pixel included in the window using the filter coefficient.