Guided image filtering for image content

Patent metadata

Field	Value
Publication number	US-9478017-B2
Application number	US-201414328920-A
Country	US
Kind code	B2
Filing date	Jul 11, 2014
Priority date	Jul 12, 2013
Publication date	Oct 25, 2016
Grant date	Oct 25, 2016

Abstract

Apparatus 10 for processing image data comprises an input 11 arranged to receive image data representing an image. The image data comprises a luminance component 13 at a first spatial resolution and at least one chrominance component 14 at a second, lower, spatial resolution. An up-sampling unit 18 up-samples the spatial resolution of the at least one chrominance component 14 to form an up-sampled chrominance component. A guided image filter (GIF) 20 filters the up-sampled chrominance component. The guided image filter comprises a radius parameter which determines an extent of the filter around a pixel. A control unit 15 for the guided image filter 20 determines a value of the radius parameter of the guided image filter. The value of the radius parameter is variable for different pixels of the image, or for different regions of pixels of the image. The control unit 15 can analyze the image data to determine the value of the radius parameter.

First claim

The invention claimed is:

1. A method for performing colour space conversion on an image to up-sample the image from a lower quality colour space to a higher quality colour space, the method comprising:
receiving image data representing the image, the image data comprising a luminance component at a first spatial resolution and at least one chrominance component at a second, lower, spatial resolution;
up-sampling the spatial resolution of the at least one chrominance component to form an up-sampled chrominance component having a higher resolution than the at least one chrominance component;
determining a value of a radius parameter for a guided image filter for filtering the image, wherein:
the value of the radius parameter is variable for different pixels of the image or for different regions of pixels of the image, and
the radius parameter determines an extent of the effect of the guided image filter around each pixel; and

filtering the up-sampled chrominance component using the guided image filter to provide a resulting up-sampled image for display.

2. A method according to claim 1 wherein the guided image filter uses the luminance component as a guide for the filtering of the up-sampled chrominance component.

3. A method according to claim 2 further comprising analysing at least one component of the image data to determine the value of the radius parameter for each pixel of the image, or for each region of pixels of the image.

4. A method according to claim 3 wherein the analysing comprises analysing the luminance component of the image data.

5. A method according to claim 4 wherein the analysing comprises analysing a window of the image data and determining a value of the radius parameter of the guided image filtering process based on the image data for the window of pixels.

6. A method according to claim 3 wherein the analysing comprises analysing a window of the image data and determining a value of the radius parameter of the guided image filtering process based on the image data for the window of pixels.

7. A method according to claim 1 further comprising analysing at least one component of the image data to determine the value of the radius parameter for each pixel of the image, or for each region of pixels of the image.

8. A method according to claim 7 wherein the analysing comprises analysing the luminance component of the image data.

9. A method according to claim 8 wherein the analysing comprises analysing a window of the image data and determining a value of the radius parameter of the guided image filtering process based on the image data for the window of pixels.

10. A method according to claim 7 wherein the analysing comprises analysing a window of the image data and determining a value of the radius parameter of the guided image filtering process based on the image data for the window of pixels.

11. A method according to claim 10 wherein the analysing comprises forming a histogram of values for the window of pixels of the image data.

12. A method according to claim 11 wherein the value of the radius is inversely proportional to a number of brightness/chrominance levels present in the image data for the window of pixels.

13. A method according to claim 10 wherein the window is a block of pixels.

14. A method according to claim 10 wherein a size of the window of pixels is variable.

15. A method according to claim 10 wherein the analysing is performed iteratively for multiple pixels of the image data, wherein each iteration of the analysis centres the window on a different pixel of the image data.

16. A method according to claim 1 further comprising receiving data from an encoder which was used to encode the image data, and determining the value of the radius parameter for each pixel of the image, or for each region of pixels of the image, based on the received data.

17. A method according to claim 1 further comprising determining a value of a further parameter for the guided image filter for filtering the image, wherein the value of the further parameter is variable for different pixels of the image, or for different regions of pixels of the image.

18. A method according to claim 1 which is performed as part of an image encoder, wherein the step of determining a value of a radius parameter for a guided image filter is used to determine data for sending to an image decoder.

19. A computer program product comprising a non-transitory machine-readable medium carrying instructions which, when executed by a processor, cause the processor to perform the method of claim 1 .

20. Apparatus for performing colour space conversion on an image to up-sample the image from a lower quality colour space to a higher quality colour space, the apparatus comprising:
an input arranged to receive image data representing the image, the image data comprising a luminance component at a first spatial resolution and at least one chrominance component at a second, lower, spatial resolution;
an up-sampling unit arranged to up-sample the spatial resolution of the at least one chrominance component to form an up-sampled chrominance component having a higher resolution than the at least one chrominance component;
a guided image filter arranged to filter the up-sampled chrominance component and provide a resulting up-sampled image for display, wherein the guided image filter comprises a radius parameter which determines an extent of the effect of the filter around a pixel; and
a control unit for the guided image filter arranged to determine a value of the radius parameter of the guided image filter, wherein the value of the radius parameter is variable for different pixels of the image or for different regions of pixels of the image.

21. Apparatus according to claim 20 further comprising an analysing unit arranged to analyse at least one component of the image data to determine the value of the radius parameter for each pixel of the image, or for each region of pixels of the image.

Assignees

• Barco Nv

Inventors

• Vermeir Thijs Koen

Classifications

• G06T7/90

  Determination of colour characteristics · 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

• G06T5/20 Primary

  using local operators · CPC title

• H04N19/186

  the unit being a colour or a chrominance component · CPC title

• H04N7/0102

  involving the resampling of the incoming video signal · CPC title