Scalable storage-loss optimized framework based method and system for color gamut mapping

Patent metadata

Field	Value
Publication number	US-9910633-B2
Application number	US-201615152239-A
Country	US
Kind code	B2
Filing date	May 11, 2016
Priority date	May 11, 2016
Publication date	Mar 6, 2018
Grant date	Mar 6, 2018

Abstract

A method for a scalable storage-loss optimized (SSLO) system for color gamut mapping is provided. The SSLO system method comprises generating a pre-computed color gamut mapping between an input color gamut and an output color gamut, receiving an input signal including a plurality of input pixels from the input signal domain, wherein the input pixels correspond to a set of first color values represented in a first color space; converting the set of the first color values to be a set of second color values represented in a second color space; generating a set of third color values through a mapping modeling and a color reproduction; converting the set of the third color values to be a set of fourth color values represented in the first color space; and outputting an output signal including a plurality of output pixels corresponding to the fourth color values to the output signal domain.

First claim

What is claimed is:

1. A method for a scalable storage-loss optimized (SSLO) system for color gamut mapping, comprising:
generating a pre-computed color gamut mapping between an input color gamut and an output color gamut, wherein an input device in an input signal domain has the input color gamut, and an output device in an output signal domain has the output color gamut;
receiving an input signal including a plurality of input pixels from the input device in the input signal domain, wherein the input pixels correspond to a set of first color values represented in a first color space;
converting the set of the first color values to be a set of second color values represented in a second color space;
based on the pre-computed color gamut mapping and the set of second color values, generating a set of third color values represented in the second color space through a mapping modeling and a color reproduction, wherein the mapping modeling comprising:
based on the pre-computed color gamut mapping, generating at least one trained model through learning a plurality of models to group colors into clusters;
performing a hierarchical clustering algorithm to model the pre-computed color gamut mapping; and
saving a plurality of color points for each cluster;

converting the set of the third color values to be a set of fourth color values represented in the first color space; and
outputting an output signal including a plurality of output pixels corresponding to the fourth color values to the output device in the output signal domain.

2. The SSLO system method according to claim 1 , wherein generating a pre-computed color gamut mapping between an input color gamut and an output color gamut further includes:
receiving at least one input signal domain reference signal from the input device in the input signal domain and having a set of first reference color values represented in the first color space;
converting the first reference set of color values represented in the first color space to be a set of second reference color values represented in the second color space;
receiving at least one output signal domain reference signal from the output device in the output signal domain and having a set of third reference color values represented in the first color space;
converting the set of third reference color values represented in the first color space to be a set of fourth reference color values represented in the second color space; and
based on the second reference color values and the fourth reference color values, generating the pre-computed color gamut mapping between the input device in the input signal domain and the output device in the output signal domain.

3. The SSLO system method according to claim 2 , wherein:
the pre-computed color gamut mapping is generated based on a color gamut mapping function G(i)=Ω s (Ψ(i))=i′,
wherein i denotes the input pixels corresponding to the first reference color values, i′ denotes the output pixels corresponding to the third reference color values, ψ denotes a color space converting function, and Ω denotes a color gamut mapping function in the second color space.

4. The SSLO system method according to claim 1 , wherein:
three or four color points are saved for each cluster.

5. The SSLO system method according to claim 1 , wherein performing a hierarchical clustering algorithm to model the pre-computed color gamut mapping further including:
calculating a distance between two pixels i and j in the second color space, where i and j is a positive integer respectively; and
based on a linkage criterion, determining a merging or splitting of the clusters.

6. The SSLO system method according to claim 1 , wherein the color reproduction further includes:
based on the trained model, predicting the cluster each input pixel belongs to; and
calculating the third color values through an interpolation based on the color points saved for each cluster.

7. The SSLO system method according to claim 6 , wherein calculating the third color values through an interpolation based on the color points saved for each cluster further including:
performing a linear interpolation to calculate the third color values.

8. The SSLO system method according to claim 7 , wherein:
based on the second color value c i predicted to be belong to a cluster A, the third color value c′ i is determined based on c′ i =w×[c′ 1 , c′ 2 , . . . , c′ n ]′,
wherein w is a weight vector and

w
⁢

⁢

1.
/

dist
A

⁡

(

c
i

)

sum
⁡

(

1.
/

dist
A

⁡

(

c
i

)

)

,

dist A (c i ) is a first distance vector between each color point represented by the first color values in the cluster A and a color point represented by the third color value c′ i , and dist A (c i )=[d(c i , c 1 ), d(c i , c 2 ), . . . , d(c i , c n )],
c n denotes the first reference color value included in the cluster A; and
c n ′ denotes the corresponding fourth reference color value included in the cluster A.

9. A non-transitory computer-readable medium having computer program for, when being executed by a processor, performing a scalable storage-loss optimized (SSLO) system method, the method comprising:
generating a pre-computed color gamut mapping between an input color gamut and an output color gamut, wherein an input device in an input signal domain has the input color gamut, and an output device in an output signal domain has the output color gamut;
receiving an input signal including a plurality of input pixels from the input device in the input signal domain, wherein the input pixels correspond to a set of first color values represented in a first color space;
converting the first of the first color values to be a set of second color values represented in a second color space;
based on the pre-computed color gamut mapping and the set of second color values, generating a set of third color values represented in the second color space through a mapping modeling and a color reproduction, wherein the mapping modeling comprising:
based on the pre-computed color gamut mapping, generating at least one trained model through learning a plurality of models to group colors into clusters;
performing a hierarchical clustering algorithm to model the pre-computed color gamut mapping; and
saving a plurality of color poin

Assignees

• Tcl Res America Inc

Inventors

• Yuan Jianbo
• Wang Haohong
• Ren Xiaobo

Classifications

• H04N5/225

  Electricity · mapped topic

• G06F3/147 Primary

  using display panels · CPC title

• G09G5/06 Primary

  using colour palettes, e.g. look-up tables · CPC title

• H04N9/64 Primary

  Circuits for processing colour signals (H04N9/77 takes precedence; camera processing pipelines for processing colour signals H04N23/84) · CPC title

• H04N1/6058

  Reduction of colour to a range of reproducible colours, e.g. to ink- reproducible colour gamut · CPC title