Method and apparatus for recovering image

Patent metadata

Field	Value
Publication number	US-10560673-B2
Application number	US-201816161229-A
Country	US
Kind code	B2
Filing date	Oct 16, 2018
Priority date	Jun 20, 2018
Publication date	Feb 11, 2020
Grant date	Feb 11, 2020

Abstract

An image recovery apparatus and method are provided, where the image recovery apparatus calculates a reference color image based on a linear correlation between a color channel image and a luminance channel image, and corrects the luminance channel image based on residual information that indicates a difference between the reference color image and the color channel image, to recover an image.

First claim

What is claimed is:

1. An image recovery method comprising:
acquiring an input image;
determining an image patch comprising pixels corresponding to a partial area of the input image as a color channel image;
generating a reference color image from a luminance channel image based on linear correlation information indicating a linear correlation between the color channel image and the luminance channel image;
calculating residual information based on the reference color image and the color channel image; and
correcting the luminance channel image based on the residual information.

2. The image recovery method of claim 1 , wherein the determining of the color channel image further comprises selecting a blue channel image as the color channel image from the input image.

3. The image recovery method of claim 1 , further comprising:
compressing pixel values of the luminance channel image to a maximum value of a display.

4. An image recovery method comprising:
generating a reference color Image from a luminance channel image based on linear correlation information indicating a linear correlation between a color channel image and the luminance channel image;
calculating residual information based on the reference color image and the color channel image; and
correcting the luminance channel image based on the residual information,
wherein the generating of the reference color image comprises
generating a target color image by filtering the color channel image based on a difference in position between pixels in the luminance channel image and a difference between pixel values of the pixels in the luminance channel image.

5. The image recovery method of claim 4 , wherein the generating of the reference color image comprises:
determining a first linear correlation coefficient based on a covariance between the color channel image and the luminance channel image and a variance of the luminance channel image;
determining a second linear correlation coefficient based on the first linear correlation coefficient, an expected value of the color channel image, and an expected value of the luminance channel image; and
generating the reference color image by applying the first linear correlation coefficient and the second linear correlation coefficient to the luminance channel image.

6. The image recovery method of claim 1 , further comprising:
calculating illuminance information based on a low frequency component of the luminance channel image.

7. The image recovery method of claim 6 , wherein the correcting of the luminance channel image comprises correcting the luminance channel image based on the calculated illuminance information and the residual information.

8. The image recovery method of claim 6 , wherein the calculating of the illuminance information comprises:
generating a target luminance image by filtering the luminance channel image based on a difference in position between pixels in the luminance channel image and a difference between pixel values of the pixels in the luminance channel image; and
determining the target luminance image as the illuminance information.

9. An image recovery method comprising:
generating a reference color image from a luminance channel image based on linear correlation information indicating a linear correlation between the color channel image and the luminance channel image;
calculating residual information based on removing an intensity value of a pixel in the reference color image that corresponds to a pixel in the color channel image from an intensity value of the pixel in the color channel image; and
correcting the luminance channel image based on the residual information.

10. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the image recovery method of claim 1 .

11. An image recovery apparatus comprising:
an image acquirer configured to acquire a color channel image; and
a processor configured to
generate a reference color image from a luminance channel image based on linear correlation information indicating a linear correlation between a color channel image and the luminance channel image,
calculate illuminance information based on a low frequency component of the luminance channel image,
calculate residual information based on the reference color image and the color channel image, and
correct the luminance channel image based on the calculated illuminance information and the residual information.

12. The image recovery apparatus of claim 11 , wherein
the image acquirer is further configured to acquire an input image, and
the processor is further configured to determine the color channel image from the input image.

13. The image recovery apparatus of claim 12 , wherein the processor is further configured to determine an image patch comprising pixels corresponding to a partial area of the input image as the color channel image.

14. The image recovery apparatus of claim 12 , wherein the processor is further configured to select a blue channel image as the color channel image from the input image.

15. The image recovery apparatus of claim 11 , wherein the processor is further configured to compress pixel values of the luminance channel image to a maximum value of a display.

16. The image recovery apparatus of claim 11 , wherein the processor is further configured to generate a target color image by filtering the color channel image based on a difference in position between pixels in the luminance channel image and a difference between pixel values of the pixels in the luminance channel image.

17. The image recovery apparatus of claim 16 , wherein the processor is further configured to:
determine a first linear correlation coefficient based on a covariance between the color channel image and the luminance channel image and a variance of the luminance channel image;
determine a second linear correlation coefficient based on the first linear correlation coefficient, an expected value of the color channel image and an expected value of the luminance channel image; and
generate the reference color image by applying the first linear correlation coefficient and the second linear correlation coefficient to the luminance channel image.

Assignees

• Samsung Electronics Co Ltd
• Univ Yonsei Iacf

Inventors

• Ryu Inhwan
• Kang Moon Gi
• Lee Keechang
• Jee Seung Hoon
• Kim Min Sub

Classifications

• H04N9/78

  for separating the brightness signal or the chrominance signal from the colour television signal, e.g. using comb filter · CPC title

• H04N9/646 Primary

  for image enhancement, e.g. vertical detail restoration, cross-colour elimination, contour correction, chrominance trapping filters · CPC title

• H04N23/86

  for controlling the colour saturation of colour signals, e.g. automatic chroma control circuits · CPC title

• G06T3/4015

  Image demosaicing, e.g. colour filter arrays [CFA] or Bayer patterns · CPC title

• G06T3/4007

  based on interpolation, e.g. bilinear interpolation (image demosaicing G06T3/4015; edge-driven or edge-based scaling G06T3/403) · CPC title