EP3671706A1 - Method for adjusting the colorimetry parameters of a display device - Google Patents

Abstract

The technical field of the invention is that of methods for adjusting the photometric and colorimetric parameters of a display device composed of bright colored dots. The method according to the invention makes it possible to calibrate the display device according to a reference display device. The method consists, starting from a limited number of colorimetric measurement points, in establishing the corrections to be made to the control values of each colored point so that the light emission of said point corresponds to that of the reference display device. This application takes into account the measurement points closest to the colored point. The tolerance on the corrections is sufficiently low so that the colorimetric differences between the real light emission of the display device and the reference emission are not perceptible by a human observer.

Description

The general technical field of the invention is that of display devices for applications requiring very high colorimetric and photometric performances. The preferred field of the invention is that of aeronautical dashboards.

In the aeronautical field, a dashboard has several display screens. The colorimetry and luminance of the images displayed on these screens are subject to very specific requirements so that the crew members always have the same image quality. In particular, the continuity of the gradations between the different colors must be perfectly ensured. Furthermore, when a display screen is replaced, it is essential that this replacement be done while maintaining the same color and photometric requirements.

These colorimetric performances go beyond the standards applied by the suppliers of display screens, especially when changing the model of display screen or suppliers. A correction is therefore necessary to maintain the photometric and colorimetric performances.

A colored pixel of a screen conventionally comprises a triplet of colored sub pixels, a first red sub-pixel, a second green sub-pixel and a third blue sub-pixel. Each sub-pixel can take 255 values. Therefore, potentially, each pixel can have 255 ³ levels of color or luminance, or just over 16 million levels. It is of course impossible to make an individualized correction for each level which would require a colossal amount of preliminary measurements and very large capacities for storing the necessary corrections on each on-board display device.

Furthermore, it is impossible to establish a mathematical model which, from a limited number of correction measurements would make it possible to calculate all the correction measurements applied to each sub-pixel. Indeed, the sensitivity of the human eye to variations in color or luminance varies very strongly in function of this color or level and does not obey simple rules that are easy to implement, once a few reference points are known.

We know that a color can be described in a two-dimensional space called CIE 1976, this space being taken from CIE space 1931. In this first space, each color is defined by a triplet (X, Y, Z) or a triplet (X Y Z). In the CIE 1976 area, each color is defined by two coordinates (u ', v') defined as follows: $$\{\begin{array}{c}\mathit{uʹ}=\frac{4X}{X+15Y+3Z}=\frac{4x}{-2x+12y+3}\\ \mathit{vʹ}=\frac{9Y}{X+15Y+3Z}=\frac{9y}{-2x+12y+3}\end{array}$$

entre deux teintes voisines vaut : $${\mathrm{\Delta }\mathit{E}}_{\mathit{U}\mathit{V}}^{*}=\sqrt{{\left(\mathrm{\Delta }\mathit{L}*\right)}^2+{\left(\mathrm{\Delta }\mathit{u}*\right)}^2+{\left(\mathrm{\Delta }\mathit{v}*\right)}^2}$$

avec $$\{\begin{array}{c}L*=\{\begin{array}{c}116\sqrt[2]{Y/Y_w-16}\mathit{si\; L}*\ge 8\\ {\left(\frac{29}{3}\right)}^3\left(Y/Y_w\right)\mathit{si\; L}*<8\end{array}\\ \begin{array}{cc}u*& =13L*\left(\mathit{uʹ}-u_w^{ʹ}\right)\\ v*& =13L*\left(\mathit{vʹ}-v_w^{ʹ}\right)\end{array}\end{array}$$

This space is represented on the figure 1 . The line SL curved in bold lines represents the spectrum locus and the straight line LP which closes this curve represents the line of purples. The spectrum locus varies from approximately 420 nm to 680 nm. In this space, a difference in color ${\mathrm{\Delta }\mathit{E}}_{\mathit{UV}}^{*}$

between two neighboring shades is worth: $${\mathrm{\Delta }\mathit{E}}_{\mathit{U}\mathit{V}}^{*}=\sqrt{{\left(\mathrm{\Delta }\mathit{L}*\right)}^2+{\left(\mathrm{\Delta }\mathit{u}*\right)}^2+{\left(\mathrm{\Delta }\mathit{v}*\right)}^2}$$

with $$\{\begin{array}{c}L*=\{\begin{array}{c}116\sqrt[2]{Y/Y_w-16}\mathit{if\; L}*\ge 8\\ {\left(\frac{29}{3}\right)}^3\left(Y/Y_w\right)\mathit{if\; L}*<8\end{array}\\ \begin{array}{cc}u*& =13L*\left(\mathit{uʹ}-u_w^{ʹ}\right)\\ v*& =13L*\left(\mathit{vʹ}-v_w^{ʹ}\right)\end{array}\end{array}$$

The triplet (u ' _w , v' _w , Y _w ) corresponds to the color coordinates of white.

It has been shown that around each color, there is an area called MacAdam ellipse or EMA inside which the differences between colors are not visible to the human eye. As an example, four MacAdam ellipses are represented on the figure 1 . These ellipses are enlarged 10 times in this figure.

The method according to the invention is based on this property. The color correction does not have to be perfect. It is enough that this correction is sufficiently precise to enter the criteria of MacAdam.

• Etape 1 : Mesure, pour le dispositif de visualisation de référence, pour une seconde pluralité de triplets de commande de calibration, des triplets de valeurs colorimétriques d'émission dits triplet de référence, la seconde pluralité étant de trois ordres de grandeur inférieure à la première pluralité ;
• Etape 2 : Mesure, pour le dispositif de visualisation, pour la même seconde pluralité de triplets de commande de calibration, des triplets de valeurs colorimétriques d'émission dits triplets de calibration ;
• Etape 3 : Pour chaque triplet de calibration, calcul des corrections à appliquer au triplet de commande de calibration correspondant de façon que le triplet de valeurs colorimétriques obtenu ait les mêmes valeurs que le triplet d'émission de référence correspondant ;
• Etape 4 : A tout triplet de commande du dispositif de visualisation dit triplet émis, détermination d'un sous-ensemble de triplets de calibration les plus proches dudit triplet émis dans le premier espace colorimétrique ;
• Etape 5 : calcul des corrections à appliquer au triplet émis en fonction des corrections appliquées au sous-ensemble des triplets de calibration déterminé à l'étape 4.

More specifically, the method according to the invention is a method for adjusting the photometric and colorimetric parameters of a display device composed of bright colored dots, each point consisting of a triplet of colored pixels, each point being able to take a first plurality of triplets of control values, said triplets being referenced in a first color space, each triplet of control values corresponds to a triplet of emission color values, said triplets of color values being referenced in a second color space, said device display being calibrated with respect to a reference display device, characterized in that the method comprises the following steps:

• Step 1: Measuring, for the reference display device, for a second plurality of calibration control triplets, triplets of emission colorimetric values called reference triplets, the second plurality being three orders of magnitude smaller than the first plurality;
• Step 2: Measurement, for the display device, for the same second plurality of calibration control triplets, triplets of emission colorimetric values called calibration triplets;
• Step 3: For each calibration triplet, calculation of the corrections to be applied to the corresponding calibration control triplet so that the triplet of colorimetric values obtained has the same values as the corresponding reference emission triplet;
• Step 4: At any control triplet of the display device said triplet emitted, determination of a subset of calibration triplets closest to said triplet emitted in the first color space;
• Step 5: calculation of the corrections to be applied to the triplet sent as a function of the corrections applied to the subset of the calibration triplets determined in step 4.

Advantageously, the second plurality of calibration control triplets comprises a group of control triplets corresponding to specific colors, said group being taken into account, in step 5, in the calculation of the corrections applied to the control triplets of the measurement device. visualization.

Advantageously, the corrections are inversely proportional to the distances separating the triplet emitted from the specific colors, the distances being measured in the first color space.

• Etape 3.1 : Détermination d'une application qui associe à chaque triplet de commande de calibration le triplet d'émission correspondant ;
• Etape 3.2 : Détermination d'un groupement de triplets de commande de calibration les plus proches dudit triplet de commande de calibration dans le premier espace colorimétrique ;
• Etape 3.3 : Détermination, pour chaque triplet de commande de calibration, de la dérivée de l'application en fonction du triplet de calibration et du triplet de référence correspondant ;
• Etape 3.4 : Détermination de l'inverse de la dérivée de l'application, pour le triplet de commande de calibration considéré, en fonction de la différence des valeurs entre les triplets de calibration et les triplets de référence correspondant aux triplets de commande de calibration dudit groupement, l'inverse de cette dérivée étant représentative de la correction à apporter au triplet de commande de calibration.

Advantageously, step 3 comprises the following sub-steps:

• Step 3.1: Determination of an application which associates with each triplet of calibration command the corresponding transmission triplet;
• Step 3.2: Determination of a grouping of calibration control triplets closest to said calibration control triplet in the first color space;
• Step 3.3: Determination, for each calibration control triplet, of the derivative of the application as a function of the calibration triplet and of the corresponding reference triplet;
• Step 3.4: Determination of the inverse of the derivative of the application, for the calibration control triplet considered, as a function of the difference in values between the calibration triplets and the reference triplets corresponding to the calibration control triples of said grouping, the inverse of this derivative being representative of the correction to be made to the calibration control triplet.

Advantageously, the first color space is the space (R, G, B) and that the second color space is the CIE 1976 space.

Advantageously, the first plurality is equal to 255 ³ and that the second plurality is between 10,000 and 15,000.

• [Fig.1] représente l'espace colorimétrique CIE 1976 ;
• [Fig.2] représente l'espace de mesure colorimétrique tridimensionnel (r, g, b) utilisé dans les étapes 1 et 2 du procédé selon l'invention ;
• [Fig.3] représente les corrections à apporter à un triplé de calibration telles que définies dans l'étape 3 du procédé selon l'invention ;
• [Fig.4] représente les corrections à apporter à un triplé quelconque telles que définies dans les étapes 4 et 5 du procédé selon l'invention ;

The accompanying drawings illustrate the invention:

• [ Fig. 1 ] represents the CIE 1976 color space;
• [ Fig. 2 ] represents the three-dimensional colorimetric measurement space (r, g, b) used in steps 1 and 2 of the method according to the invention;
• [ Fig. 3 ] represents the corrections to be made to a calibration triplet as defined in step 3 of the method according to the invention;
• [ Fig. 4 ] represents the corrections to be made to any triplet as defined in steps 4 and 5 of the method according to the invention;

The method for adjusting the photometric and colorimetric parameters applies to a display device composed of bright colored dots. Each point consists of a triplet of colored pixels. Each point can take a number N of triplets of command values. Generally, each pixel being able to take 255 different values, this number N is equal to 255 ³ , that is to say a little more than 16.5 million values. The triplets are referenced in a first color space. Generally, this space is space (R, G, B). Each triplet of control values corresponds to a triplet of emission color values, said triplets of color values being referenced in a second color space. This space is the color space (u ', v', L) as defined above.

The method requires a reference display device, a measurement and calibration bench comprising colorimetric and photometric measurement means, means for memorizing the collected data and means for processing said data in order to extract the calibration corrections to be applied to it. some specific color points. The display device must also include means for storing the calibration corrections and calculation means for calculating the correction to be applied to any color triplet received as a function of the calibration corrections. These different technical means are known to those skilled in the art and do not pose any particular production problem.

The method according to the invention comprises several steps detailed below. The first three steps are implemented on the optical bench. They consist in determining the corrections to be made to a certain number of colorimetric triplets for calibration.

The following steps are implemented in the display device itself when it is operating. They consist in applying, in real time, from the previous correction data, the corrections to be applied to any color triplet emitted.

The first step consists in measuring, for the reference display device, for a number I of calibration control triplets, triplets of emission colorimetric values called reference triplets, the second plurality being three orders of magnitude less than the first plurality. For example, the number I is equal to 12800. The figure 2 illustrates this step. The triplets of points (r _i , g _i , b _i ) of calibration commands are referenced in the color space (R, G, B). the parameter i varies between 1 and I. these triplets are represented by circles on this figure 2 . The calibration control points are located on a three-dimensional grid, part of which is represented by dotted lines on the figure 2 and are regularly spaced on this grid. Each triplet of calibration control points (r _i , g _i , b _i ) therefore associates a triplet of emission colorimetric values called reference triplet whose coordinates in the color space (u ', v', L) are (u _i *, v _i *, L _i *) _TH .

There are certain colorimetric points (r _j , g _j , b _j ) called critical colors whose associated emission must correspond to a perfectly defined colorimetry. In the aeronautical field, these critical colors can correspond, for example, to alarms. By their nature, these critical points do not necessarily correspond to the measurement points of the three-dimensional grid. One of these points (r _j , g _j , b _j ) is represented by a double concentric circle on the figure 2 .

The second step consists in measuring, for the display device, for the same second plurality of calibration control triplets (r _i , g _i , b _i ), triplets of emission colorimetric values known as calibration triplets which are noted (u _i *, v _i *, L _i *) in the color space (u ', v', L).

In a third step, the corrections to be applied to each calibration control triplet (r _i , g _i , b _i ) are calculated so that the triplet of colorimetric values obtained has the same values as the corresponding reference emission triplet . This third step 3 comprises the following intermediate steps:

qui associe à chaque triplet de commande de calibration le triplet d'émission correspondant ;In a first step of this third step, we determine an application Ψ such that: $\mathrm{\Psi }\left(r_i,g_i,b_i\right)=\left(u_i^{*},v_i^{*},L_i^{*}\right)$

which associates with each calibration control triplet the corresponding transmission triplet;

et chaque triplé de référence correspondant est noté ${\left(u_j^{*},v_j^{*},L_j^{*}\right)}_{\mathit{TH}}\mathrm{.}$

In a second step of this third step, we determine, for each triplet ( r _i , g _i , b _i ), i belonging to I, a triplet subfamily close to the triplet ( r _i , g _i , b _i ) . As an example, this subfamily corresponds to the vertices of the cube including the triplet ( r _i , g _i , b _i ). We note this family J _i . The triplets belonging to this family are therefore {( r _j , g _j , b _j )} _{j ∈ J i} . Each corresponding emission triple is noted $\left(u_j^{*},v_j^{*},L_j^{*}\right)$

and each corresponding reference treble is noted ${\left(u_j^{*},v_j^{*},L_j^{*}\right)}_{\mathit{TH}}\mathrm{.}$

On en déduire la matrice dérivée ${\mathrm{\Psi }}_{\left(r,g,b\right)}^{ʹ}\left(r_i,g_i,b_i\right)$

de l'application Ψ(r_i ,g_i ,b_i ) au point (r_i ,g_i ,b_i ). Cette dérivée est égale à la matrice M qui minimise la forme suivante : $${\sum _{j\in J_i}\Vert M\left({\mathit{dr}}_j,{\mathit{dg}}_j,{\mathit{db}}_j\right)-\left(\left(u_j^{*},v_j^{*},L_j^{*}\right)-{\left(u_j^{*},v_j^{*},L_j^{*}\right)}_{\mathit{TH}}\right)\Vert }^2$$

In a third step of this third step, we deduce, for each j ∈ J _i , the deviation vector ( dr _j , dg _j , db _j ). This vector is equal to: $$\left({\mathit{dr}}_j,{\mathit{dg}}_j,{\mathit{db}}_j\right)=\left(r_i,g_i,b_i\right)-\left(r_j,g_j,b_j\right)\mathrm{.}$$

We deduce the derived matrix ${\mathrm{\Psi }}_{\left(r,g,b\right)}^{ʹ}\left(r_i,g_i,b_i\right)$

from the map Ψ ( r _i , g _i , b _i ) to point ( r _i , g _i , b _i ). This derivative is equal to the matrix M which minimizes the following form: $${\sum _{j\in J_i}\Vert M\left({\mathit{dr}}_j,{\mathit{dg}}_j,{\mathit{db}}_j\right)-\left(\left(u_j^{*},v_j^{*},L_j^{*}\right)-{\left(u_j^{*},v_j^{*},L_j^{*}\right)}_{\mathit{TH}}\right)\Vert }^2$$

pour le triplet de commande de calibration considéré (r_i , g_i ,b_i ), en fonction de la différence des valeurs entre les triplets de calibration et les triplets de référence correspondant aux triplets de commande de calibration de la, l'inverse de sous-famille, cette dérivée étant représentative de la correction (dr_i , db_i , dg_i ) apporter au triplet de commande de calibration. On a la relation : $$\left({\mathit{dr}}_i,{\mathit{dg}}_i,{\mathit{db}}_i\right)={\mathrm{\Psi }}_{\left(r,g,b\right)}^{ʹ-1}\left(\left(u_i^{*},v_i^{*},L_i^{*}\right)-{\left(u_i^{*},v_i^{*},L_i^{*}\right)}_{\mathit{TH}}\right)$$

In a fourth step of this third step, the inverse of the derivative of the application noted is determined. ${\mathrm{\Psi }}_{\left(r,g,b\right)}^{ʹ-1}$

for the calibration control triplet considered ( r _i , g _i , b _i ), as a function of the difference in values between the calibration triplets and the reference triplets corresponding to the calibration control triplets of the, the inverse of sub-family, this derivative being representative of the correction ( dr _i , db _i , dg _i ) to bring to the triplet of calibration control. We have the relation: $$\left({\mathit{dr}}_i,{\mathit{dg}}_i,{\mathit{db}}_i\right)={\mathrm{\Psi }}_{\left(r,g,b\right)}^{ʹ-1}\left(\left(u_i^{*},v_i^{*},L_i^{*}\right)-{\left(u_i^{*},v_i^{*},L_i^{*}\right)}_{\mathit{TH}}\right)$$

This last stage of the third stage is illustrated on the figure 3 which represents a calibration triplet ( r _i , g _i , b _i ). with the associated corrections ( dr _i , dg _i , db _i ), the corrected triplet being noted ( r _i , g _i , b _i ) _cor .

The second and third steps are also carried out for critical colors.

These corrections being known for all the triplets of calibration control, it is then necessary to determine the corrections to be made to any triple command (r, g, b) applied to the display device. This is the object of the fourth and fifth stages of the process. These steps are implemented in real time in the display device.

The fourth step of the method consists, for any control triplet (r, g, b) of the display device said triplet emitted, to determine a subset of calibration triplets closest to said triplet emitted in the first color space. As an example, this subset corresponds to the vertices of the cube including the triplet (r, g, b). We denote this subset J _k . The triplets belonging to this family are therefore {( r _k , g _k , b _k )} _{k ∈ J k} . When the command triplet is close to one or more critical colors, these are also determined. This step is illustrated on the figure 4 which represents a triplet of points (r, g, b) with all the triplets of points to be taken into account for the establishment of the corrections. In this example, eight calibration points numbered from T1 to T8 are taken into account as well as two critical colors C1 and C2.

• des corrections (dr1, dg1, db1) appliquées au sous-ensemble des triplets de calibration déterminé à l'étape précédente et,
• lorsqu'elles sont présentes, des corrections (dr2, dg2, db2) issues des couleurs critiques proches.

The fifth step consists in calculating the corrections ( dr , dg , db ) to apply to the triplet emitted. These corrections are based on:

• corrections ( dr 1, dg 1, db 1) applied to the subset of the calibration triplets determined in the previous step and,
• when present, corrections ( dr 2, dg 2 , db 2) from near critical colors.

qui sert de base au calcul de ces corrections est obtenue par interpolation linéaire des valeurs des matrices ${\mathrm{\Psi }}_{\left(r_k,g_k,b_k\right)}^{ʹ},$

interpolation étendue aux valeurs (r_k , g_k , b_k ) proches du triplet (r, g, b).Regarding the corrections ( dr 1, dg 1 , db 1) from the calibration triplets, the derived matrix ${\mathrm{\Psi }}_{\left(r,g,b\right)}^{ʹ}\left(r,g,b\right)$

which is used as a basis for the calculation of these corrections is obtained by linear interpolation of the values of the matrices ${\mathrm{\Psi }}_{\left(r_k,g_k,b_k\right)}^{ʹ},$

extended interpolation to values ( r _k , g _k , b _k ) close to the triplet (r, g, b).

The corrective value (dr2, dg 2, db 2) due to the critical colors is calculated by weighted average of the sum of the influences of the critical colors sufficiently close, the weighting being inversely proportional to the distance to each of these critical colors.

The final function ( dr , dg , db ) is calculated by weighted average over the corrections ( dr 1, dg 1, db 1) and ( dr 2, dg 2, db 2) knowing that the weighting remains proportional to the inverse of the distance to the nearest critical color. Thus, if the triplet (r, g, b) corresponds exactly to a critical color, the final corrective value is exactly that of the critical color in question.

Claims (6)

Method for adjusting the photometric and colorimetric parameters of a display device composed of bright colored dots, each point consisting of a triplet of colored pixels (r, g, b), each point being able to take a first plurality of triplets of control values, said triplets being referenced in a first color space (R, G, B), to each triplet of control values there corresponds a triplet of emission color values (u ', v', L), said triplets of color values being referenced in a second color space, said display device being calibrated with respect to a reference display device, characterized in that the method comprises the following steps: - Step 1: Measurement, for the reference display device, for a second plurality of calibration control triplets (r _i , g _i , b _i ), triplets of emission colorimetric values called reference triplet (u _i *, v _i *, L _i *) _TH , the second plurality being three orders of magnitude less than the first plurality, between 10,000 and 15,000, so that the corrections applied to any control triplet of the display device called triplet emitted (r, g, b) are sufficient to enter MacAdam's criteria for tolerable chromatic variation; - Step 2: Measurement, for the display device, for the same second plurality of calibration control triplets, triplets of emission colorimetric values called calibration triplets (u _i *, v _i *, L _i *); - Step 3: For each calibration triplet, calculation of the corrections (dr _i , dg _i , db _i ), to be applied to the corresponding calibration control triplet so that the triplet of colorimetric values obtained has the same values as the triplet d 'corresponding reference program; - Step 4: At any control triplet of the display device said triplet emitted (r, g, b), determination of a subset of calibration triplets closest to said triplet emitted in the first color space; - Step 5: calculation of the corrections (dr, dg, db) to be applied to the triplet transmitted as a function of the corrections applied to the subset of the calibration triplets determined in step 4. Method for adjusting the photometric and colorimetric parameters of a display device according to claim 1, characterized in that the second plurality of calibration control triplets comprises a group of control triplets corresponding to specific colors, said group being taken into account, in step 5, in the calculation of the corrections applied to the control triplets of the device of visualization. Method for adjusting the photometric and colorimetric parameters of a display device according to claim 2, characterized in that the corrections are inversely proportional to the distances separating the emitted triplet from the specific colors, the distances being measured in the first color space. Method for adjusting the photometric and colorimetric parameters of a display device according to one of the preceding claims, characterized in that step 3 comprises the following substeps: - Step 3.1: Determination of an application (Ψ) which associates with each triplet of calibration command the corresponding emission triplet; - Step 3.2: Determination of a grouping of calibration control triplets closest to said calibration control triplet in the first color space; - Step 3.3: Determination, for each triplet of calibration control, of the derivative (Ψ ') of the application as a function of the calibration triplet and of the corresponding reference triplet; - Step 3.4: Determination of the inverse (Ψ ' ^-1 ) of the derivative of the application, for the calibration control triplet considered, as a function of the difference in values between the calibration triplets and the corresponding reference triplets to the calibration control triplets of said grouping, the inverse of this derivative being representative of the correction to be made to the calibration control triplet. Method for adjusting the photometric and colorimetric parameters of a display device according to one of the preceding claims, characterized in that the first color space is space (R, G, B) and that the second color space is l 1976 CIE space. Method for adjusting the photometric and colorimetric parameters of a display device according to one of the preceding claims, characterized in that the first plurality is equal to 255 ³ . 1

Images