Image processing apparatus and method

Patent metadata

Field	Value
Publication number	US-9519952-B2
Application number	US-201614993067-A
Country	US
Kind code	B2
Filing date	Jan 11, 2016
Priority date	Mar 12, 2015
Publication date	Dec 13, 2016
Grant date	Dec 13, 2016

Abstract

An image processing apparatus includes a gradient calculation unit, a direction determining unit, a directional interpolation unit, and a blender unit. The gradient calculation unit processes an input image to generate gradient magnitudes and gradient angles associated with input pixels of the input image. The direction determining unit generates interpolation angles and directional confidence values according to the gradient magnitudes and gradient angles. The directional interpolation unit performs directional interpolation on the input image according to the interpolation angles, so as to generate a first image with an image resolution different from the input image. The blender receives the first image and a second image generated from interpolating the input image, and blends the first image and the second image according to the directional confidence values to generate an output image.

First claim

What is claimed is:

1. An image processing apparatus, comprising:
a gradient calculation unit configured to perform an operation on an input image to generate gradient magnitudes and gradient angles respectively associated with a plurality of input pixels of the input image;
a direction determining unit configured to generate a plurality of interpolation angles and a plurality of directional confidence values according to the gradient magnitudes and the gradient angles associated with the input pixels;
a directional interpolation unit configured to perform directional interpolation on the input image according to the interpolation angles, so as to generate a first image with an image resolution different from that of the input image; and
a blender unit configured to receive the first image and a second image generated from interpolating the input image, and configured to blend the first image and the second image into an output image according to weights of the first image and the second image associated with the directional confidence values, wherein the resolution of the second image is the same as that of the first image.

2. The image processing apparatus of claim 1 , wherein the input pixels comprise a plurality of first input pixels in a first range, and the directional interpolation unit is configured to compare the gradient magnitudes and the gradient angles of the first input pixels and other input pixels adjacent to the first input pixels to generate a first interpolation angle of the interpolation angles and one of the directional confidence values.

3. The image processing apparatus of claim 2 , wherein the input pixels comprise a plurality of second input pixels in a second range, and the second range comprises the first range, and the directional interpolation unit is configured to compare the first interpolation angle and the gradient angles of the second input pixels to update the first interpolation angle.

4. The image processing apparatus of claim 1 , wherein the directional interpolation unit is configured to associate a first interpolation angle of the interpolation angles with a first angular interval of a plurality of angular intervals and select the input pixels within a sampling window as a plurality of directional interpolation input pixels, and to selectively perform low-pass filtering on the directional interpolation input pixels according to the first angular interval, and then to generate a pixel value of a first pixel according to pixel values of the processed directional interpolation input pixels, wherein the first pixel is associated with the first interpolation angle and is one of the pixels in the first image.

5. The image processing apparatus of claim 4 , wherein the directional interpolation unit is configured to select one of a plurality of angular sampling windows as the sampling window according to the first angular interval, wherein the angular sampling windows comprise rectangular coordinate sampling windows, vertical diagonal coordinate sampling windows and horizontal diagonal coordinate sampling windows.

6. The image processing apparatus of claim 4 , wherein the directional interpolation unit is configured to associate the first interpolation angle with a second angular interval, and to generate another pixel value of the first pixel according to the second angular interval, and to blend the pixel value and the another pixel value of the first pixel to update the pixel value of the first pixel.

7. An image processing method, comprising:
performing an operation on an input image to generate gradient magnitudes and gradient angles associated with a plurality of input pixels of the input image;
generating a plurality of interpolation angles and a plurality of directional confidence values according to the gradient magnitudes and the gradient angles associated with the input pixels;
performing directional interpolation on the input image according to the interpolation angles, so as to generate a first image with an image resolution different from that of the input image; and
performing interpolation on the input image to generate a second image, and blending the first image and the second image into an output image according to weights of the first image and the second image associated with the directional confidence values, wherein the resolution of the second image is the same as that of the first image.

8. The image processing method of claim 7 , wherein the step of generating the interpolation angles and the directional confidence values comprises:
comparing the gradient magnitudes and the gradient angles of a plurality of first input pixels and other input pixels adjacent to the first input pixels to generate a first interpolation angle of the interpolation angles and one of the directional confidence values, wherein the first input pixels comprise the input pixels located in the first range.

9. An image processing apparatus, comprising:
a gradient calculation unit configured to perform an operation on an input image to generate gradient magnitudes and gradient angles associated with a plurality of input pixels of the input image;
a direction determining unit configured to generate a plurality of interpolation angles according to the gradient magnitudes and the gradient angles associated with the input pixels; and
a directional interpolation unit configured to associate a first interpolation angle of the interpolation angles with a first angular interval of a plurality of angular intervals and select the input pixels within a sampling window as a plurality of directional interpolation input pixels, and to selectively perform low-pass filtering on the directional interpolation input pixels according to the first angular interval, and then to generate a pixel value of a first pixel according to pixel values of the processed directional interpolation input pixels, wherein the first pixel is associated with the first interpolation angle and is one of the pixels in the first image.

10. The image processing apparatus of claim 9 , wherein the directional interpolation unit is configured to select one of a plurality of angular sampling windows as the sampling window according to the first angular interval, wherein the angular sampling windows comprise rectangular coordinate sampling windows, vertical diagonal coordinate sampling windows and horizontal diagonal coordinate sampling windows.

11. The image processing apparatus of claim 9 , wherein the directional interpolation unit is configured to associate the first interpolation angle with a second angular interval, and to generate another pixel value of the first pixel according to the second angular interval, and to blend the pixel value and the another pixel value of the first pixel, so as to update the pixel value of the first pixel.

Assignees

• Realtek Semiconductor Corp

Inventors

• Yang Jun-Yu
• Hsieh Chun-Hsing
• LIN Shang-Yen

Classifications

• G06T5/20

  using local operators · CPC title

• G06T3/4007 Primary

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

• G06T11/60

  Creating or editing images; Combining images with text · CPC title

• G06T2207/20192

  Edge enhancement; Edge preservation · CPC title

• G06T5/73

  Deblurring; Sharpening · CPC title