Image processing apparatus and method for increasing sharpness of images

Patent metadata

Field	Value
Publication number	US-9619864-B2
Application number	US-201514840032-A
Country	US
Kind code	B2
Filing date	Aug 30, 2015
Priority date	Mar 25, 2015
Publication date	Apr 11, 2017
Grant date	Apr 11, 2017

Abstract

An image processing apparatus includes a high-frequency component translating unit, a high-frequency component extracting unit, a detail-gain generating unit and an image output unit. The high-frequency component translating unit extracts and translates first high-frequency components of an input image to generate a first image. The high-frequency component extracting unit extracts second high-frequency components to generate a second image. The detail-gain generating unit stores a conversion table and generates detail gains respectively associated with input pixels in the input image according to pixel values of the input pixels and the conversion table. The image output unit calculates a weighted superposition of the first image and the second image and generates a high frequency component of an output image according to the weighted superposition and the detail gains.

First claim

What is claimed is:

1. An image processing apparatus, comprising:
at least one processor connected to a memory, wherein the at least one processor is configured to:
extract and translate a plurality of first high-frequency components of an input image to generate a first image;
extract a plurality of second high-frequency components of the input image to generate a second image;
generate a plurality of detail gains respectively associated with a plurality of input pixels in the input image according to a conversion table and pixel values of the input pixels, wherein the detail gain of a first input pixel of the input pixels is corresponding to a pixel value sum of pixel values of the first input pixel and a plurality of neighboring pixels separate from the first input pixel by at least one pixel; and
calculate a weighted superposition of the first image and the second image and to generate a high-frequency component of an output image according to the detail gains and the weighted superposition.

2. The image processing apparatus of claim 1 , wherein the at least one processor is configured to blend the high-frequency component of the output image into the input image to generate the output image.

3. The image processing apparatus of claim 1 , wherein the at least one processor is configured to extract the plurality of neighboring pixels separate from the first input pixel by at least one pixel and to search the conversion table with a pixel value sum of the first input pixel and the neighboring pixels to generate the detail gain associated with the first input pixel, wherein the first input pixel is one of the input pixels.

4. The image processing apparatus of claim 3 , wherein the at least one processor is configured to extract pixels separate from the first input pixel by one pixel from up, down, left, and right directions as the neighboring pixels.

5. The image processing apparatus of claim 1 , wherein the conversion table corresponds to a periodic continuous function.

6. The image processing apparatus of claim 1 , wherein a value range of the detail gains is from 0 to N.

7. The image processing apparatus of claim 1 , wherein the at least one processor is configured to filter the input image to generate a plurality of high-frequency translation images with at least two overshoot high-pass filters, and the at least one processor is configured to generate the first image by selecting pixels with highest absolute pixel values from the high-frequency translation images, and the high-frequency translation images comprise the first high-frequency components, respectively.

8. The image processing apparatus of claim 7 , wherein the overshoot high-pass filters are associated with a plurality of masks respectively, and coefficient values at the center and periphery of the masks are positive while the rest of the coefficient values of the masks are negative.

9. The image processing apparatus of claim 1 , wherein the at least one processor is configured to filter the input image to generate a plurality of high-frequency images with at least two high-pass filters and is configured to generate the second image by selecting pixels with highest absolute pixel values from the high-frequency images, and the high-frequency images comprise the second high-frequency components, respectively.

10. An image processing method, comprising:
extracting and translating a plurality of first high-frequency components from an input image to generate a first image;
extracting a plurality of second high-frequency components from the input image to generate a second image;
generating a plurality of detail gains respectively associated with a plurality of input pixels in the input image according to pixel values of the input pixels and a conversion table, wherein the detail gain of a first input pixel of the input pixels is corresponding to a pixel value sum of pixel values of the first input pixel and a plurality of neighboring pixels separate from the first input pixel by at least one pixel; and
calculating a weighted superposition of the first image and the second image and generating a high-frequency component of an output image according to the detail gains and the weighted superposition.

11. The image processing method of claim 10 , further comprising:
blending the high-frequency component of the output image into the input image to generate the output image.

12. The image processing method of claim 10 , wherein the operation of generating the detail gains comprises:
extracting the plurality of neighboring pixels; and
searching the conversion table with the pixel value sum to generate the detail gain associated with the first input pixel, wherein the first input pixel is one of the input pixels.

13. The image processing method of claim 12 , wherein the operation of generating the detail gains comprises:
extracting pixels separate from the first input pixel by one pixel from up, down, left, and right directions as the neighboring pixels.

14. The image processing method of claim 10 , wherein the conversion table corresponds to a periodic continuous function.

15. The image processing method of claim 10 , wherein a value range of the detail gains is from 0 to N.

16. The image processing method of claim 10 , wherein the operation of generating the first image comprises:
filtering the input image to generate a plurality of high-frequency translation images with at least two overshoot high-pass filters, and generating the first image by selecting pixels with highest absolute pixel values from the high-frequency translation images, wherein the high-frequency translation images comprise the first high-frequency components, respectively.

17. The image processing method of claim 16 , wherein the overshoot high-pass filters are associated with a plurality of masks respectively, and coefficient values at the center and periphery of the masks are positive while the rest of the coefficient values of the masks are negative.

18. The image processing method of claim 10 , wherein the operation of generating the second image comprises:
filtering the input image to generate a plurality of high-frequency images with at least two high-pass filters, and generating the second image by selecting pixels with highest absolute pixel values from the high-frequency images, wherein the high-frequency images comprise the second high-frequency components, respectively.

19. An image processing method, comprising:
extracting and translating a plurality of first high-frequency components from an input image to generate a first image;
extracting a plurality of second high-frequency components from the input image to generate a second image;
generating a plurality of detail gains respectively associated with a plurality of input pixels in the input image according to pixel values of the input pixels and a conversion table; and
calculating a weighted superposition of the first image and the second image and generating a high-frequency component of an output image according to the detail gains and the weighted superposition;
wherein the operation of generating the first image comprises:
filtering the input image to generate a plurality of high-frequency translation images with at least two overshoot high-pass filters, and generating the first image by selecting pixels with highest absolute pixel values from the high-frequency translation images, wherein the high-frequency translation images comprise the first high-frequency components, respectively.

20. The image

Assignees

• Realtek Semiconductor Corp

Inventors

• Chen Cheng-Yueh
• Hsieh Chun-Hsing
• Yang Jun-Yu
• Sun Zhi-Gang

Classifications

• G06T5/003

  Physics · mapped topic

• G06T3/4061 Primary

  by injecting details from different spectral ranges · CPC title

• G06T5/73

  Deblurring; Sharpening · CPC title