WO2007071604A1

WO2007071604A1 - Image display method for the determination or the adjustment of the electro-optical response of a screen or projector

Info

Publication number: WO2007071604A1
Application number: PCT/EP2006/069668
Authority: WO
Inventors: Laurent Blonde; Luis Montalvo; Jonathan Kervec
Original assignee: Thomson Licensing
Priority date: 2005-12-22
Filing date: 2006-12-13
Publication date: 2007-06-28

Abstract

The invention is an image display method intended for determining or adjusting the electro-optical response of a screen. The invention exploits the fact that, if the screen displays two sets of visual information within a short interval of time, only the composition of these two sets of visual information is perceived by the human eye. For example, if the sets of visual information are two different colours and if the interval of time is less than 50 ms (higher frequency than the fusion frequency of the colours), only the average colour will be perceived by the eye. According to the invention, a colour is displayed in the standard manner within a first region of the images and the same colour is displayed by fusion of colours within another region of the images. These images are displayed with a frequency greater than the fusion frequency of the colours. If the eye detects no difference in colour between these regions, the electro-optical response of the screen is that used for determining the colours to be fused.

Description

IMAGE DISPLAY METHOD FOR THE DETERMINATION OR THE ADJ USTMENT OF THE ELECTRO-OPTICAL RESPONSE OF A SCREEN

OR PROJECTOR

Field of invention

The present invention is in the field of digital video screens or projectors manufactured both for professional applications such as digital cinema and for consumer applications. The invention more particularly relates to a method for determining and/or adjusting the electro-optical optical response of a screen or projector.

Background

The electro-optical optical response of a screen or a projector is the transformation between the electronic input signal supplying it and the visual information resulting from it for the whole range of possible variations of the input signal.

Today, the adjustment of the electro-optical optical response of a consumer television is carried out at the factory via the definition of a few predefined profiles, for example 'Standard', 'Sport' or 'Film' profiles, and/or by the user via the definition of more personalized profiles. These profiles are used to adjust the contrast, the luminance and the colour of the television screen. For professional monitors, these adjustments, then denoted as 'calibrations', are made with respect to electronic references or by physical measurements.

Until now, by allowing the user the possibility of adjusting his screen himself, the television industry has not been too concerned with the correct reproduction of the colours. This does not present a particular problem for the majority of programmes. However, the cinema industry, in particular the directors of photography, attaches a high importance to the faithful reproduction of their work. This is not really possible, or at least not guaranteed, with the projectors or screens currently available on the market. Furthermore, certain processes such as the process preventing the copying of a film by using a camcorder, also called anti-camcorder process, described in the French patent application 2 859 857 use the fusion of colours to tattoo the images to be displayed. This can only work if the screen or the projector is correctly adjusted or, at the very least, is adjusted as is pre-supposed in the anti-camcorder process.

Summary of the invention The aim of the invention is to provide a simple image display method for determining or adjusting the electro-optical optical response of a screen or projector. The invention exploits the fact that, if the screen displays two sets of visual information within a short interval of time, only the composition of these two sets of visual information is perceived by the human eye. For example, if the sets of visual information are two different colours and if the interval of time is less than 50 ms (frequency higher than the fusion frequency of the colours), only the average colour will be perceived by the eye. Two sets of visual information having different luminances may also be used. The two sets of visual information perceived by the human eye depend on the electro-optical optical response of the screen. If this is incorrectly adjusted, the resultant of the two sets of visual information perceived by the human eye will not correspond to the desired value.

The invention relates to an image display method intended for the determination or the adjustment of the electro-optical optical response of a screen or projector, which screen or projector is capable of transforming an electrical input signal representative of an image or sequence of images into visual information. This method comprises the following steps: a) define first and second input signal values corresponding to first and second sets of visual information that are symmetrical with respect to a reference set of visual information based on a reference electro-optical optical response associating a set of visual information with each input signal value, b) define first and second images comprising first image regions of the same shape and same spatial position and second image regions of the same shape and same spatial position, the visual information to be displayed within the first regions of the two images being equal to the reference visual information and the visual information to be displayed within the second regions of the first and second images being respectively said first and second sets of visual information that are symmetrical with respect to said reference visual information, c) display said two images with a frequency that is higher than or equal to a threshold frequency.

When the aim is to determine the electro-optical optical response of the screen or projector, the method comprises an additional step for comparing the visual response of the first and second image regions in such a manner that, if the visual responses of the first and second regions after integration over time of the two images are identical, the reference electro-optical optical response is the electro-optical response of the screen or projector and, if not, all the steps are repeated while modifying the reference electro-optical response.

When the aim is to adjust the electro-optical response of the screen or projector, the method also comprises an additional step for comparing the visual response of the first and second image regions in such a manner that, if the visual responses of the first and second regions are identical, the reference electro-optical response is the electro-optical response of the screen or projector and, if not, all the steps are repeated, modifying the electro-optical response of the screen until the visual responses of said first and second regions are identical. According to a preferred embodiment, the colour components of the first and second sets of visual information are symmetrical with respect to the colour component of the reference set of visual information and the threshold frequency is the fusion frequency of the colours. In this embodiment, the electro-optical response of the screen is the reference electro-optical response for which the colours of the first regions and second regions are identical.

Preferably, the luminance of the first set of visual information is equal to the luminance of the second set of visual information and to that of the reference set of visual information in order to avoid a flickering effect.

Advantageously, in the images, the first and second regions are adjacent and the first region surrounds the second region or vice versa.

According to one particular embodiment for determining the electro-optical response of the screen, several pairs of complementary sets of visual information are defined according to step a) based on several reference electro-optical responses. The pairs are then displayed within separate second regions of the first and second images according to step b). The electro-optical response of the screen is then the reference electro-optical response used for defining the pair of symmetrical sets of visual information displayed within the second regions having a visual response after integration over time of the two images identical to that of the first regions.

The invention will be better understood upon reading the description that follows, presented by way of non-limiting example and with reference to the appended drawings, amongst which:

- Figure 1 shows an example of image pair (I1J2) implementing the invention,

- Figure 2 illustrates the use of several pairs of images for determining the electro-optical response of a screen, - Figure 3 illustrates the use of a single pair of images for determining the electro-optical response of the screen,

- Figure 4 is a flow chart indicating the steps of the method of the invention when it is intended for determining the electro-optical response of a screen or projector, and

- Figure 5 is a flow chart indicating the steps of the method of the invention when it is intended for adjusting the electro-optical response of a screen or projector.

Description of embodiments

The invention will more particularly be described in regard to the fusion of colours and the visual information that is modulated for implementing the invention is the colour.

As in the anti-camcorder process described in the French patent application 2 859 857, the method of the invention exploits the fact that the fusion frequency of the colours is around 20 Hz. This means that, if two colours are displayed within an interval of time less than 50 ms, only the composition of these two colours will be perceived by the eye, and will be perceived without flicker.

According to the invention, at least two images are to be displayed on the screen each comprising at least two types of regions, hereinafter called region of type 1 et region of type 2. Each of these images comprises at the same locations regions of the same type and of the same shape.

Within the regions of type 1 , a given colour is displayed in a standard manner, namely where the same digital excitation triplet (Ro, Go, Bo) is applied to the input of the screen for the two images, which triplet generates at the output of the tube the visual response triplet (R'_O,GO,B'_O). The element R or R' of the triplets relates to the red component of the video signal, the element G or G' relates to the green component and the element B or B' relates to the blue component. In place of the triplet (R'o,GO,B'o), a triplet of coordinates (Xo, Yo, Zo), in the colorimetric space XYZ defined by the ILC (International Lighting Commission), may also be used.

Within the regions of type 2, the same colour is displayed as in the anti- camcorder process, namely where a first digital excitation triplet (R₁₁Gi₁B₁) generating a visual response triplet (R'I .G'I .B'I) is displayed on a first image I₁ and a second digital excitation triplet (R2,G2,B2) generating a visual response triplet (R'2,G'2,B'₂) is displayed on a second image I₂. The triplets (R₁₁G₁₁B₁) and (R₂, G₂₁B₂) are chosen such that the average of the visual responses (R'I .G'I .B'I) and (R'_2lG'_2lB'₂) corresponds, by fusion of the colours, to the visual response (RO,GO,BO) of the regions of type 1 when the reference electro-optical response used to calculate the triplets (R₁₁G₁₁B₁) and (R₂₁G₂₁B₂) is that of the screen. (R₁₁G₁₁B₁) and (R₂₁G₂₁B₂) are chosen such that (R¹! + R'₂)/2 « R'_Ol (C₁ + G'₂)/2 « C₀ and (B¹! + B'₂)/2 « B'_o. The triplets (R'l.G'i.B'i) and (R'_2lG'_2lB'₂) represent sets of visual information that are symmetrical with respect to the visual information (RO₁GO₁BO). More generally, the triplet (RO₁GO₁BO) is the weighted average of the triplets (RN₁C₁₁BN) and (R'_2lG'_2lB'₂).

If the method of the invention is used to determine the electro-optical response of the screen, the method then consists in varying the reference electro-optical response used to calculate the triplets (R₁₁G₁₁B₁) and (R_2lG_2lB₂) until the eye detects no difference in colour between the regions of type 1 and those of type 2.

If the method of the invention is used to adjust the electro-optical response of the screen to a desired value, the triplets (R₁₁G₁₁B₁) and (R_2lG_2lB₂) are calculated based on the desired electro-optical response and the electro- optical response of the screen is varied until the eye detects no difference in colour between the regions of type 1 and those of type 2. In order to make the electro-optical response of the screen vary, the means controlling the luminance, the contrast and the colour of the screen or a colour management device placed upstream of the screen are for example acted on.

This process allows a particular point of the electro-optical response of the screen to be adjusted and/or determined, which may be sufficient if the law characterizing this response is well known. Otherwise, several triplets may be used, successively or on the same pair of images, for sampling the electro- optical response at several points and therefore refining the adjustment and/or the determination of the latter.

In the following part of the description, the method of the invention is more particularly described for the adjustment or the determination of the electro- optical response of a CRT (cathode-ray tube) screen. The electro-optical response of a CRT screen may be defined as a function in which the output signal S is modelled as a function of the input signal E by the equation

S = S_max where γ (gamma) is the transformation parameter which

depends on the adjustment of the tube.

The input signal is for example a digital excitation triplet (R, G, B) supplying the CRT. The corresponding output signal is a triplet (R', G', B') such that:

R K¹ = G ^J¹ = B o' - m \ > )

Within the regions of type 1 of the image, a colour determined by its input triplet (R_o,Go,B_o) is displayed and is transformed by the screen into an output triplet (R'_O,GO,B'_O), each element of the triplet obeying equation (1 ).

Within the regions of type 2, the same colour is displayed by means of two complementary colours characterized by the input triplets (Ri,Gi,Bi) and (R₂,G₂,B₂) at the input of the tube and the triplets (R'i,G'i,B'i) and (R'₂,G'₂,B'₂) at the output of the tube such that: = ^R 0 = G ₀ - B ₀ (2)

Each element of the triplets (R'i,G'i,B'i) and (R'2,G'₂,B'2) also obeys equation (1 )-

Advantageously, the three colours corresponding to the triplets (R'o,GO,B'o), (R'I ,G'I ,B'I) and (R'2,G'2,B'2) have the same luminance value. Thus, if the triplets (R'i,G'i,B'i) are transformed into triplets (Yi₁Uj₁Vi) where Yi is a luminance value and Ui and Vi are chrominance values, the aforementioned condition becomes: Y₀ = Y₁ = Y₂ (3)

This condition allows flickering when the colours are fused to be avoided. If each element of the triplet (R'i,G'i,B'i) has 256 levels, this transformation can be carried out by the following set of equations:

Examples of image sequences comprising such regions are presented in the following section with reference to Figures 1 to 3.

With reference to Figure 1 , a sequence of images displaying the message "γ≠2.2" on a grey background is displayed on the screen. The part of the message "γ=2.2" is displayed in black and the oblique bar of the sign ≠ is displayed with the same value of grey as the image background. The grey background is a region of type 1 and the oblique bar of the sign ≠ is a region of type 2 of the image. The colour assigned to the grey background corresponds to the triplet (R'_O,GO,B'_O) and the region where this colour is displayed is the region Zi in the image h (left-hand side of Figure 1 ) and the region Z₂ in the image I₂ (right-hand side of Figure 1). The regions of type 2 (the oblique bar) in the images \-_\ and I₂ are the regions Z₁' and Z₂'. The oblique bar of the sign ≠ is for example displayed in orange in the region Z1¹ of the first image \-_\ and in blue in the region Z₂' of the following image I₂, the resultant colour obtained by fusion of the colours and perceived by the eye being the grey of the region of type 1. The triplet (Ri,Gi,Bi) of the orange colour and the triplet (R₂,G₂,B2) of the blue colour are therefore defined such that, if γ is for example equal to 2.2, their fusion gives the grey of the region of type 1 of the image. Thus, when this sequence of images is displayed, the oblique bar will visually disappear if the value γ characterizing the electro- optical response of the screen is equal to 2.2 (reference value used to determine the triplets (Ri,Gi,Bi) and (R₂, G₂₁B₂)). If it is different, the oblique bar will be visible. Then, "γ≠2.2" can be read on the screen. The further the value γ will be from the value 2.2, the more visible will be the oblique bar.

If the level 92 is taken, for example, as grey level of the region of type 1 , the maximum grey level being fixed at 255, then Ro'=Go'=Bo'=92, which corresponds to Yo=92 and Uo=V₀=I 28. In order to create the complementary colours orange and blue, the component U may for example be decreased and the component V increased in order to obtain the colour orange and, symmetrically, the component U increased and the component V decreased for the colour blue. The colour orange is for example defined by Yi=92,

Ui=100 and \Λ=156 and the complementary blue by Y₂=92, U₂=156 and

V₂=I OO.

In order to calculate the triplets (R₁₁Gi₁Bi) and (R2,G₂,B₂) to be applied at the input of the tube in order to obtain the two colours defined by the triplets (Yi,Ui,Vi) and (Y₂,U₂,V₂), equation (5) that follows must first of all be applied so as to firstly obtain (R'i,G'i,B'i) and (R'₂,G'₂,B'₂), then equation (1 ) used in order to obtain (Ri,Gi,Bi) and (R2,G₂,B₂). Equation (5) is the reciprocal function of equation (1 ):

In order to calculate (R_o,Go,Bo), equation (1 ) is inverted. Thus, taking γ = 2.2, the following is obtained:

R₁ = 188, G₁ = 152, B₁ = 115 for the colour orange (region of type 2),

R₂ - 126, G₂ - 168, B₂ - 194 for the colour blue (region of type 2) and

R₀ - 160, G₀ - 160, B₀ - 160 for grey (region of type 1 ).

Thus, in order to determine the electro-optical response of the screen, a first pair of images (I₁J₂) can be displayed using a first reference value of gamma in order to determine the triplets (Ri,Gi,Bi) and (R2,G2,B₂) to be used in the regions of type 2, then a second pair of images (I₃J₄) using a second reference value of gamma, then a third pair of images (I₅J₆) using a third reference value of gamma, and so on. This particular embodiment is illustrated in Figure 2. A value of γ equal to 2.1 is used to generate the images (I₁J₂), a value of γ equal to 2.2 to generate the images (I₃J₄) and a value of γ equal to 2.3 to generate the images (I₅J₆)- The value of γ of the screen corresponds to that of the pair of images for which the eye detects no difference between the regions of type 1 and the regions of type 2, i.e. when the eye does not detect the oblique bar.

The following table groups examples of values of triplets (R₀, Go₁B₀), (R₁₁Gi₁B₁) and (R₂₁G₂₁B₂) to be used in order to implement the method of the invention.

According to another embodiment illustrated in Figure 3, a region of type 1 (Z₁ in the image I₁ and Z₂ in the image I₂), and a plurality of regions of type 2 (Zi',Z3',Z₅' in the image I₁ and Z₂ ¹^₁Z₆' in the image I₂), based on several values of gamma, are displayed on a pair of images (I₁J₂). In the example in figure 3, the regions Z₁ and Z₂ represent the grey background of the image.

The regions Z₁' and Z₂' represent the oblique bar of a message "γ≠2.1 " displayed in the top part of the images I₁ and I₂. The colours (R'i,G'i,B'i) and (R'2,G'₂,B'2) displayed in the regions Zi' and Z₂' are calculated based on a value of γ equal to 2.1. The regions Z3' and Z₄' represent the oblique bar of a message "γ≠2.2" displayed in the middle part of the images I₁ and I₂. The colours (R'₃,G'3,B'3) and (R'₄,G'₄,B'4) displayed in the regions Z₃' and Z₄' are calculated based on a value of γ equal to 2.2. The regions Z5' and Z& represent the oblique bar of a message "γ≠2.3" displayed in the lower part of the images I₁ and I₂. The colours (R'₅,G'₅,B'5) and (R'₆,G'6,B'6) displayed in the regions Z₅' and Z₆' are calculated based on a value of γ equal to 2.3. The value of γ of the screen corresponds to that for which the eye does not detect the oblique bar.

Figures 4 and 5 respectively indicate the steps for the method of the invention in the case where it is intended to determine the electro-optical response of a screen or projector and in the case where it is intended to adjust the electro-optical response of a screen or projector to a reference value called reference electro-optical response.

With reference to figure 4, in a first step 10 triplets (R₀,G₀,B₀), (R₁₁Gi₁Bi) and (R₂₁G₂₁B₂) are defined such that (R'i + R'₂)/2 « R'_Ol (G'i + G'₂)/2 « GO and (B'i + B'₂)/2 « BO based on a reference electro-optical response. The following step, 20, consists in defining images I₁ and I₂ comprising regions of type 1 and type 2, for example images comprising regions Z₁, Z₁', Z₂ and Z₂' as previously illustrated in Figures 1 to 3, for displaying these triplets. These images are subsequently displayed at a step 30 with a frequency higher than or equal to a threshold frequency, for example the fusion frequency of the colours if the colour in the regions of type 2 of the images is modulated. At step 40 that follows, the human eye performs an integration over time. If the visual response detected by the human eye in the regions of type 1 (Zi₁Z₂) is equal to that detected in the regions of type 2 (Zi',Z₂'), the reference electro- optical response used to define the triplets (Ri,Gi,Bi) and (R2,G₂,B2) is the electro-optical response of the screen or projector. If not, steps 10 to 40 are repeated (step 50) while modifying the reference electro-optical response.

Figure 5 is identical to Figure 4 as regards the steps 10 to 40. The step 50 is replaced by the step 50'. During this step, if the visual response detected by the human eye in the regions of type 1 (Zi₁Z₂) is not equal to that detected in the regions of type 2 (Zi',Z₂'), steps 10 to 40 are repeated while modifying the electro-optical response of the screen or projector.

It goes without saying that the invention is not limited to the embodiments defined hereinabove. In particular, those skilled in the art will be able to implement the method of the invention using other sequences of images, in particular other messages than the message "γ≠2.1" or "γ≠2.2" or "γ≠2.3", or other regions of type 1 and 2. They will also be able to modulate the luminance in the regions of type 2 instead of modulating the colour. In this case, the display frequency will need to be higher than or equal to 60 Hz, in the case of an interlaced display, or 75 Hz in progressive mode.

Furthermore, the invention is not limited to the adjustment or the determination of the electro-optical response of CRT screens, but to any type of projector (rear or front projection) or screen.

Claims

1. Image display method intended for the determination or the adjustment of the electro-optical response of a screen or projector, which screen or projector is capable of transforming an electrical input signal ((R₁G₁B)) representative of an image or sequence of images into visual information ((R',G',B')), characterized in that it comprises the following steps: a) define first and second input signal values ((Ri,Gi,Bi) and (R2,G2,B2)) corresponding to first and second sets of visual information ((R'i,G'i,B'i) and

(R'2,G'2,B'2)) that are symmetrical with respect to a reference set of visual information ((RO, GO₁BO)) based on a reference electro-optical response associating a set of visual information with each input signal value, b) define first and second images (\-_\, I₂) comprising first image regions (Zi,Zi') of the same shape and same spatial position and second image regions (Z2, Z2') of the same shape and same spatial position, the visual information to be displayed within the first regions (Zi,Zi') of the two images being equal to the reference visual information (RO, GO, BO) and the visual information to be displayed within the second regions (Z₂) of the first and second images being respectively said first and second sets ((R'i,G'i,B'i), (R'2,G'2,B'₂)) of visual information that are symmetrical with respect to said reference visual information, c) display said two images (h, I2) with a frequency that is higher than or equal to a threshold frequency.

2. Method according to Claim 1 , characterized in that, in order to adjust the electro-optical response of said screen or projector, it also comprises a step d) for comparing the visual response of the first and second image regions in such a manner that, if the visual responses of the first and second regions are identical, the reference electro-optical response is the electro-optical response of said screen or projector and, if not, the steps a) to d) are repeated, modifying the electro-optical response of the screen until the visual responses of said first and second regions are identical.

3. Method according to Claim 1 , characterized in that, in order to determine the electro-optical response of said screen or projector, it also comprises a step d) for comparing the visual response of the first and second image regions in such a manner that, if the visual responses of the first and second regions after integration over time of the two images are identical, the reference electro-optical response is the electro-optical response of said screen or projector and, if not, the steps a) to d) are repeated, modifying the reference electro-optical response.

4. Method according to one of Claims 1 to 3, characterized in that each set of visual information comprises a colour component and in that the colour components of the first (Ui,Vi) and second (112,V₂) sets of visual information are symmetrical with respect to the colour component (Uo₁Vo) of the reference set of visual information.

5. Method according to Claim 4, characterized in that the threshold frequency is the fusion frequency of the colours.

6. Method according to either of Claims 4 and 5, characterized in that each set of visual information also comprises a luminance component and in that the luminance components of the first (Yi) and second (Y₂) sets of visual information and the reference set of visual information (Yo) are identical.

7. Method according to any one of Claims 1 to 6, characterized in that said first and second regions are adjacent.

8. Method according to one of Claims 1 to 7, characterized in that said first region surrounds said second region or vice versa.

9. Method according to any one of Claims 3 to 8, characterized in that several pairs of symmetricalal sets of visual information (((R'i,G'i,B'i), (R¹2,G¹2,B¹2)), ((R'3,G'₃,B'3), (R'4,G'₄,B'4)), ((R¹S₁G¹S₁B s), (R'_6lG'_6lB'₆))) are defined according to step a) based on several reference electro-optical responses, and in that said pairs are displayed within separate second regions (Z₁', Z₂', Z₃', Z₄', Z₅', Z₆') of said first and second images (I₁J₂) according to step b), the electro-optical response of the screen being the reference electro-optical response used for defining the pair of symmetrical sets of visual information displayed within the second regions having a visual response after integration over time of the first and second images identical to that of the first regions.