Granted Patent B2

US 12,333,673 · App. 17/773,411 · Granted Jun 17, 2025

Inventors: Valentin Dessy (Mont-Saint-Guibert, BE); Gael Rouvroy (Mont-Saint-Guibert, BE); Pascal Pellegrin (Mont-Saint-Guibert, BE)

Assignee: INTOPIX SA

G06T3/4015H04N9/64

Abstract

Claims (50)

1. An image processor configured to process an image comprising pixels of three different colors (R, G, B) arranged in a Bayer pattern so as to obtain a de-correlated image composed of components of three different types (Y, Cr, Cb) arranged in a pattern corresponding to the Bayer pattern,

whereby, according to the Bayer pattern, the image is a rectangular array of blocks of 2 by 2 pixels that comprise two diagonally disposed pixels of a first color (G), a pixel of a second color (R), and a pixel of a third color (B); and

whereby, according to the pattern corresponding to the Bayer pattern, the de-correlated image is a rectangular array of blocks of 2 by 2 components that comprise two diagonally disposed components of a first type (Y), a component of a second type (Cr), and a component of a third type (Cb),

wherein the processor is configured to:

provide a component of the first type (Y) in the de-correlated image as a substitute for a pixel of the first color (G) in the image, whereby the component of the first type (Y) is a weighted combination of a cluster of pixels in the image that includes the pixel of the first color (G) and neighboring pixels, wherein neighboring pixels of the second and third color (R, B) have an overall positive weighting factor corresponding to an overall addition of neighboring pixels of the second and third color (R, B) to the pixel of the first color (G),

provide a component of the second type (Cr) in the de-correlated image as a substitute for a pixel of the second color (R) in the image, whereby the component of the second type (Cr) is a weighted combination of a cluster of pixels in the image that includes the pixel of the second color (R) and neighboring pixels, wherein neighboring pixels of the first color (G) have an overall negative weighting factor corresponding to an overall subtraction of neighboring pixels of the first color (G) from the pixel of the second color (R), and

provide a component of the third type (Cb) in the de-correlated image as a substitute for a pixel of the third color (B) in the image, whereby the component of the third type (Cb) is a weighted combination of a cluster of pixels in the image that includes the pixel of the third color (B) and neighboring pixels, wherein neighboring pixels of the first color (G) have an overall negative weighting factor corresponding to an overall subtraction of neighboring pixels of the first color (G) from the pixel of the third color (B),

wherein the weighted combination that forms the component of the first type (Y) comprises:

nearest neighboring pixels of the first color (G) that are in an x-configuration with respect to the pixel of the first color that is substituted, and

further neighboring pixels of the first color that are in an +-configuration with respect to the pixel of the first color that is substituted.

2. An image processor according to claim 1 , wherein the image processor is configured to:

first provide respective components of the second type (Cr) as substitutes of respective pixels of the second color (R) and to provide respective components of the third type (Cb) as substitutes of respective pixels of the third color (B), and, subsequently, to

provide respective components of the first type (Y) as substitutes of respective pixels of the first color (G), whereby a component of the first type (Y) is provided on the basis of a weighed combination of, on the one hand, the pixel of the first color (G) for which the component of the first type is a substitute and, on the other hand, components of the second type (Cr) and components of the third type (Cb) that have been provided as substitutes for pixels of the second color and for pixels of the third color, respectively, that form a cluster with the pixel of the first color.

3. An image processor according to claim 1 , wherein the nearest neighboring pixels of the first color (G) that are in an x-configuration have a heavier weighting factor in the weighted combination than the further neighboring pixels of the first color that are in an +-configuration.

4. An image processor according to claim 3 , wherein, in the weighted combination that forms the component of the first type (Y):

the pixel of the first color (G) that is substituted has a weighting factor of +⅞;

the nearest neighboring pixels of the first color that are in an x-configuration have a weighting factor of − 1/16;

the further neighboring pixels of the first color that are in an +-configuration have a weighting factor of − 1/32; and

the neighboring pixels of the second color (R) and the neighboring pixels of the third color (B) have a weighting factor of +⅛.

5. An image processor according to claim 1 , wherein, in the weighted combination that forms the component of the second type (Cr):

the pixel of the second color (R) that is substituted has a weighting factor of +1; and

the neighboring pixels of the first color (G) are four immediately neighboring pixels of the first color that have a weighing factor of −¼.

6. An image processor according to claim 1 , wherein, in the weighted combination that forms the component of the third type (Cb):

the pixel of the third color (B) that is substituted has a weighting factor of +1; and

the neighboring pixels of the first color (G) are four immediately neighboring pixels of the first color that have a weighing factor of −¼.

7. An image processor according to claim 1 , wherein the image processor is configured to apply a wavelet-based filtering to components of the first type (Y) in the de-correlated image so as to obtain high frequency components of the first type (Y H ) and low frequency components of the first type (Y L ), the wavelet-based filtering involving a decomposition of the components of the first type (Y) into a first set of arrays of components of the first type and a second set of arrays of components of the first type, the arrays of components of the first type that belong to the first set being interlaced with the arrays of components of the first type that belong to the second set, whereby the high frequency components of the first type (Y H ) are provided for the components of the first type that belong to the first set of arrays and whereby the low frequency components of the first type (Y L ) are provided for the components of the first type that belong to the second set of arrays.

8. An image processor according to claim 7 , wherein a high frequency component of the first type (Y H ) is a weighted combination between a component of the first type belonging to the first set of arrays and neighboring components of the first type belonging to the second set of arrays, and wherein a low frequency component of the first type (Y L ) is a weighed sum of a component of the first type belonging to the second set of arrays and neighboring high-frequency components.

9. An image processor according to claim 8 , wherein:

in the weighted combination that forms the high frequency component of the first type (Y H ), the component of the first type belonging to the first set of arrays has a weighting factor of +1 and the neighboring components of the first type belonging to the second set of arrays have a weighting factor of −¼; and

in the weighed sum that forms the low frequency component of the first type (Y L ), the component of the first type belonging to the second set of arrays has a weighting factor of +1 and neighboring high-frequency components have a weighting factor of +⅛.

10. An image processor according to claim 7 , wherein the processor is configured to apply a two-dimensional wavelet-based filtering to the low frequency components of the first type (Y L ), to the components of the second type (Cr) in the decorrelated image, and to the components of the third type (Cb) in the decorrelated image.

11. An encoding system comprising an image processor according to claim 1 and an encoder configured to encode at least one of the following: the de-correlated image and a wavelet-based filtered version of the de-correlated image.

12. An encoding system according to claim 11 , wherein the encoder is configured to encode:

the high frequency components of the first type (Y H );

sub bands produced obtained by applying the two-dimensional wavelet-based filtering to the low frequency components of the first type (Y L );

sub bands produced obtained by applying the two-dimensional wavelet-based filtering to, the components of the second type (Cr) in the decorrelated image; and

sub bands produced obtained by applying the two-dimensional wavelet-based filtering to the components of the third type (Cb) in the decorrelated image,

wherein the processor is configured to apply a two-dimensional wavelet-based filtering to the low frequency components of the first type (Y L ), to the components of the second type (Cr) in the decorrelated image, and to the components of the third type (Cb) in the decorrelated image.

13. An inversely-operating image processor configured to process the de-correlated image produced by the image processor according to claim 1 , the inversely-operating image processor being configured to reconstitute the image comprising the pixels of three different colors (R, G, B) that are arranged in the Bayer pattern.

14. A method of processing an image comprising pixels of three different colors (R, G, B) arranged in a Bayer pattern so as to obtain a de-correlated image composed of components of three different types (Y, Cr, Cb) arranged in a pattern corresponding to the Bayer pattern,

whereby, according to the Bayer pattern, the image is a rectangular array of blocks of 2 by 2 pixels that comprise two diagonally disposed pixels of a first color (G), a pixel of a second color (R), and a pixel of a third color (B); and

whereby, according to the pattern corresponding to the Bayer pattern, the de-correlated image is a rectangular array of blocks of 2 by 2 components that comprise two diagonally disposed components of a first type (Y), a component of a second type (Cr), and a component of a third type (Cb),

wherein the method comprises:

providing a component of the first type (Y) in the de-correlated image as a substitute for a pixel of the first color (G) in the image, whereby the component of the first type (Y) is a weighted combination of a cluster of pixels in the image that includes the pixel of the first color (G) and neighboring pixels, wherein neighboring pixels of the second and third color (R, B) have an overall positive weighting factor corresponding to an overall addition of neighboring pixels of the second and third color (R, B) to the pixel of the first color (G),

providing a component of the second type (Cr) in the de-correlated image as a substitute for a pixel of the second color (R) in the image, whereby the component of the second type (Cr) is a weighted combination of a cluster of pixels in the image that includes the pixel of the second color (R) and neighboring pixels, wherein neighboring pixels of the first color (G) have an overall negative weighting factor corresponding to an overall subtraction of neighboring pixels of the first color (G) from the pixel of the second color (R), and

providing a component of the third type (Cb) in the de-correlated image as a substitute for a pixel of the third color (B) in the image, whereby the component of the third type (Cb) is a weighted combination of a cluster of pixels in the image that includes the pixel of the third color (B) and neighboring pixels, wherein neighboring pixels of the first color (G) have an overall negative weighting factor corresponding to an overall subtraction of neighboring pixels of the first color (G) from the pixel of the third color (B),

wherein the weighted combination that forms the component of the first type (Y) comprises:

nearest neighboring pixels of the first color (G) that are in an x-configuration with respect to the pixel of the first color that is substituted, and

further neighboring pixels of the first color that are in an +-configuration with respect to the pixel of the first color that is substituted.

15. A non-transitory machine readable storage medium on which a computer program is stored, the computer program comprising a set of instructions that enables an image processor to carry out the method according to claim 14 .

Assignments (1)

Priority Claims (2)

Continuity (1)

References Cited (38)