Method for conversion of a colour space using separate chromatic and luminance look-up tables

• US 20080074717A1
• Filed: 09/17/2007
• Published: 03/27/2008
• Est. Priority Date: 09/21/2006
• Status: Active Grant

• Current Assignee: Thomson Licensing (Vantiva SA)
• Original Assignee: Thomson Licensing (Vantiva SA)
• Inventors: Stauder, Jurgen, Thollot, Julien, Blonde, Laurent
• Granted Patent: US 7,847,989 B2
• Time in Patent Office: Days
• US Class Current: 358/520
• CPC Class Codes: G06T 11/001 Texturing; Colouring; Generation of texture or colour inpainting G06T5/77
• CPC Class Codes: H04N 1/6019 using look-up tables H04N1/6025 takes precedence

Abstract

According to the method,

• starting from the chromatic components x_S, y_S of the colour vector to be converted, the components of an intermediate colour vector R_INT-CR, G_INT-CR, Σ_INT-CR are-interpolated, whose chromatic components, expressed in the source colour space are: x_INT-CR=x_S, y_INT-CR=y_S;
• starting from the luminance component S_S of the colour vector to be converted and from the luminance component Σ_INT-CR of the intermediate colour vector, the luminance component Σ_D of the converted colour vector is interpolated;
• the chromatic components R_D, G_D of the converted colour vector are calculated from its luminance component Σ_D and from the components of the intermediate colour vector R_INT-CR, G_INT-CR, Σ_INT-CR. The conversion methods of the prior art are substantially improved.

Claims

1. Method for converting the n components (x_S, y_S, S_S) of a colour vector expressed in an n-dimensional source colour space into m components (R_D, G_D, Σ

_D) of this same vector expressed in an m-dimensional destination colour space, the said n components comprising a source luminance component (S_S) and (n-1) source chromatic components (x_S, y_S), the said m components also comprising a destination luminance component (Σ

_D) and (m-1) destination chromatic components (R_D, G_D), the said method comprising;

a first operation in which, starting only from the said chromatic components (x_S, y_S) of the said colour vector to be converted expressed in the source colour space, the components of an intermediate colour vector in the destination colour space (R_INT-CR, G_INT-CR, Σ

_INT-CR) are interpolated,

a second operation in which, starting only with the luminance component (S_S) of the said colour vector to be converted expressed in the source colour space and from the luminance component (Σ

_INT-CR) of the intermediate colour vector expressed in the destination colour space, the luminance component (Σ

_D) of this colour vector to be converted is interpolated in the destination colour space, and

a third operation in which the chromatic components R_D, G_D) of the said colour vector expressed in the destination colour space are calculated starting from its luminance component (Σ

_D) and from the components of the said intermediate colour vector (R_INT-CR, G_INT-CR, Σ

_INT-CR) expressed in the destination colour space,

wherein, for the interpolation operation of the second operation, a normalized luminance look-up table is used comprising a plurality of pairs (S_S-N, Σ

_D-N) of values comprising a normalized luminance component (S_S-N) in the source colour space and a normalized luminance component (Σ

_D-N) in the destination colour space.

2. Method according to claim 1, wherein, in order to go from a luminance component in the source space (S_S) or in the destination space (Σ

_D) of the colour vector to be converted to a normalized luminance component in the source space (S_S-N) or in the destination space (Σ

_D-N) of the same colour vector, the following equations are used;

3. Method according to claim 1, wherein, in the source colour space, the chromatic components (X_INT-CR, Y_INT-CR) of the said intermediate vector are equal to the chromatic components (x_S, y_S) of the said vector to be converted.

4. Method according to claim 1, wherein, for the interpolation operation of the first operation, a chromatic look-up table is used that comprises a plurality of multiplets x_S-Ci, y_S-Ci, S_S-Ci, R_D-Ci, G_D-Ci, Σ

_D-Ci) with (n+m) values comprising the n components (x_S-Ci, y_S-Ci, S_S-Ci) that define a conversion colour vector in the source colour space and the m components (R_D-Ci, G_D-Ci, Σ

_D-Ci) that define this same conversion colour vector in the destination colour space.

5. Method according to claim 1, where the source space is a calorimetric space independent of any device, and where the destination space is a space dependent on a device.

6. Method according to claim 1, where the destination space is a colorimetric space independent of any device, and where the source space is a space dependent on a device.