EP1224656A1

EP1224656A1 - Video coding method for a plasma display panel

Info

Publication number: EP1224656A1
Application number: EP00962598A
Authority: EP
Inventors: Didier Doyen; Carlos Correa
Original assignee: Deutsche Thomson Brandt GmbH; Thomson Licensing SAS
Current assignee: Deutsche Thomson Brandt GmbH
Priority date: 1999-09-23
Filing date: 2000-09-11
Publication date: 2002-07-24
Also published as: CN1376295A; US6765548B1; KR100711130B1; JP2003510868A; KR20020047178A; WO2001022396A1; AU7426600A; FR2799040A1; FR2799040B1; CN1187727C

Abstract

The invention concerns a novel video coding technique for a plasma display panel, for improving the use of rotating code. The inventive method consists in coding the highest grey tone level (V1) over the whole of the underscans favouring the underscans with the lowest illumination time. The coding of the lowest grey tone level (V2) is carried out by using the common value (COMMAX) derived from coding the value (V1) if the specific value SPEMAX is greater than the difference (D) of the grey tone levels. If SPEMAX is greater than the difference (D), then the coding of the common value COMMIN corresponds to the lowest value (V2).

Description

Video coding method for a plasma display panel

A method of encoding video for a plasma display panel is provided. More particularly, the invention relates to the coding of the gray levels of a panel of the type with separate display and maintenance.

Plasma display panels, hereinafter called PAP, are flat type display screens. There are two main families of PAP, namely PAP whose operation is of the continuous type and those whose operation is of the alternative type. PAPs generally include two insulating slabs (or substrate), each carrying one or more networks of electrodes and delimiting between them a space filled with gas. The slabs are assembled to each other so as to define intersections between the electrodes of said arrays. Each intersection of electrodes defines an elementary cell to which corresponds a gas space partially delimited by barriers and in which an electric discharge occurs when the cell is activated. The electric discharge causes an emission of UV rays in the elementary cell. Luminophores (red, green or blue) deposited on the cell walls transform UV rays into visible light.

For PAPs of the alternative type, there are two types of cell architecture, one is called matrix, the other is called coplanar. Although these structures are different, the functioning of an elementary cell is essentially the same. Each cell can be in the on or off state. Maintaining in one of the states is done by sending a succession of so-called maintenance pulses for the entire duration during which one wishes to maintain this state. The ignition, or addressing, of a cell is done by sending a larger pulse, commonly called addressing pulse. The extinction, or erasure, of a cell is done by cancellation of the charges inside the cell using a damped discharge. To obtain different levels of gray, the phenomenon of integration of the eye is called up by modulating the durations of the on and off states using sub-scans, or sub-frames, during the display time of a picture. In order to be able to carry out the temporal modulation of ignition of each elementary cell, it is mainly used two techniques called addressing modes. A first addressing mode, called addressing during the display (or Addressing While Displaying) consists of addressing each cell line while the other cell lines are held, the addressing being done line by line in an offset manner. A second addressing mode, called Addressing and Display Separation, consists of addressing, maintaining and erasing all the cells of the panel during three distinct periods. For more details on these two addressing modes, a person skilled in the art can for example refer to American patents Nos. 5,420,602 and 5,446,344. FIG. 1 shows the basic time distribution of the addressing mode with separate display for displaying an image. The total display time Ttot of the image is 16.6 or 20 ms depending on the country. During the display time, eight sub-scans SB1 to SB8 are carried out in order to allow 256 gray levels per cell, each sub-scan making it possible to illuminate or not an elementary cell during an illumination time Tec multiple of a value To. Subsequently, we will speak of the lighting weight p, with p which corresponds to an integer such as Tec = p ^* To. The total duration of a subscanning includes an erasing time Tef, a time d addressing Ta, and the lighting time Tec specific to each subscanning. The addressing time Ta is also decomposable into n times an elementary duration Tae which corresponds to the addressing of a line. The sum of the illumination times Tec necessary for a maximum gray level being equal to the maximum duration of illumination Tmax, we have the following relation: Ttot = m ^* (Tef + n ^* Tae) + Tmax, in which m represents the number of subscans. Figure 1 corresponds to a binary decomposition of the lighting time. This binary representation has many drawbacks. A problem of false contouring (or "contouring") has been identified for a long time.

The problem of false contours comes from the proximity of two zones whose gray levels are very close but whose moments of illumination are decorrelated. The worst case corresponds to a transition between levels 127 and 128: Indeed, the gray level 127 corresponds to an illumination during the first seven sub-scans SB1 to SB7 while the level 128 corresponds to the illumination of the eighth sub-scan. SB8 scan. Two areas of the screen placed next to each other, having levels 127 and 128, are never lit at the same time. When the image is static and the viewer's eye does not move on the screen, integration temporal is done relatively well (if we do not take into account a possible flicker effect) and we observe two zones with relatively close gray levels. On the other hand, when the two zones move on the screen (and / or the viewer's eye moves) the integration time window changes screen zone and is moved from one zone to another for a number of cells. The displacement of the time window for integrating the eye from a level 127 area to a level 128 area leads to integration on cells that are off for the duration of a frame, which results in the appearance of 'a dark outline of the area. Conversely, the displacement of the time window for integrating the eye from a level 128 area to a level 127 area results in integration of cells lit for the duration of a frame, which results in the appearance of a clear outline of the area. When we work on pixels made up of three elementary cells (red, green and blue), this phenomenon results in false colored outlines.

The phenomenon explained occurs on all level transitions where the switched lighting weights are totally or almost totally different. Significant switching is more troublesome than switching due to their importance. The resulting effect may be more or less noticeable depending on the switched weights and their places. Thus, the effect of false contours can also occur with fairly distant levels (for example 63-128) but is much less shocking for the eye because it then corresponds to a very visible level (or color) transition. To remedy this problem of false contours, several solutions have been implemented. One solution is to "break" the highweights which involves adding underscans. Only the total display time of the image Ttot = m ^* (Tef + n ^* Tae) + Tmax remains fixed, which results in a decrease in the time Tmax (because Tef and Tae are incompressible durations) and therefore a maximum screen brightness decrease. It is possible to use up to 10 subscans while having the correct brightness. With 10 subscans, the maximum lighting time Tmax is currently 30% of the total time while the erasing and addressing time is around 70%. FIG. 2 represents an example of addressing using 10 sub-scans SB1 to SB10 in which the most significant are broken in half. In order to reduce important transitions and increase the number of sub-scans without reducing the screen brightness, one technique consists in simultaneously scanning two successive lines for certain lighting values. We then have the following relation Ttot = m (Tef + n ^* Tae) + m ₂ ^* (Tef + n / 2 ^* Tae) + Tmax. The erasure time Tef being negligible compared to an ^* Tae, we have the equivalence Ttot ≡ (m, + m ^) * (Tef + n ^* Tae) + Tmax. These simultaneous sub-scans reduce the addressing time by two, and thus make it possible to add additional sub-scans without reducing Tmax. FIG. 3 represents an example of addressing 11 sub-scans S1 to S11. The sub-scans S1 and S2 corresponding to the lowest lighting times are carried out on two lines at the same time in order to obtain an overall addressing time for these two sub-scans which is equal to the addressing time of a single underscan. If we perform sub-scans common to two successive lines for the weights 1, 2, 4, and 8 of illumination, it is possible to obtain 12 sub-scans in order to suppress the transitions of weights 64. The problem of this solution is however the loss of resolution due to the simultaneous scanning of two lines.

On the principle of sub-scans sweeping two lines at the same time, a solution consists in using a coding with rotating code, or with multiple representation. FIG. 4 illustrates a coding with rolling code using twelve sub-scans S1 to S12 with which the following lighting weights are associated: 1, 2, 4, 6, 10, 14, 18, 24, 32, 40, 48 and 56 An effect of the rolling code is to soften the switching of most significant by reducing the number of weights switched during the switching of a most significant. To obtain the twelve subscans, a simultaneous scan of two lines is carried out for the weights 2, 6, 14 and 24. Such a code also allows multiple representation of the numbers: 34 = 32 + 2 = 24 + 10 = 24 + 6 + 4 = 18 + 14 + 2 = ... etc. This multiple representation of the numbers makes it possible to code the gray levels present on the two scanned lines at the same time so that the weights 2, 6, 14 and 24 are identical. Those skilled in the art can refer to European Application No. 0 874 349 for more details on this technique. Although the softening effect of switching a high weight is attenuated by multiple coding, problems with lower resolution and false outline still arise when it is not possible to have adequate coding between two cells addressed simultaneously. In European application EP-A-0 945 846, it is proposed, among other things, to optimize the use of rotating codes in order to be able to considerably increase the number of underscans. The technique used consists in decomposing the gray levels NG1 and NG2, for two cells placed on two adjacent lines in common value VC and in specific value VS1 and VS2. Thus, we have: NG1 = VS1 + VC, NG2 = VS2 + VC, with, when NG1 <NG2: VS1 = α ^* NG1,

VS2 = D + α ^* NG1, VC = ^_1/2 (NG1 + NG2 - VS1 - VS2), and wherein α is a coefficient set depending on the type of code used, D is equal to the difference NG2-NG1 after rounding, for example rounding to 5. In most cases, rounding to 5 of the difference makes it possible to limit the error to plus or minus 1 bit. However, such a system is limited by a deviation from the maximum difference value. It is then necessary to have for example VS1 = 0, VS2 = maximum value, and VC chosen to minimize the error. The probability of the frequency of occurrence of such a case essentially depends on the choice of the number of specific sub-scans for each cell and on the duration of illumination of said sub-scans.

A second limitation of such a coding method comes from the dispersion of the different codings for the same value. In fact, the coding variations no longer depend on each cell but on each pair of cells independently of the neighboring pair. The false contour phenomenon is greatly attenuated within the pair of cells, but the attenuation of the false contour is less with the neighboring pairs. Those skilled in the art will note, on reading EP-A-0 945 846, that in order to minimize this limitation, it is advisable to use the largest possible common part which has the effect of increasing the probability of error due to the limitation resulting from the deviation from the maximum value.

The invention provides a new scanning technique aimed at improving the use of rolling code. The gray level coding method which is the subject of the invention achieves a coding which favors a choice from two possible codes depending on the gray levels associated with each cell. The two codes use equivalent criteria so that the disparity between the two codes is minimized. According to the invention, the highest gray level is coded as a priority over all of the sub-scans. If the coding on all of the sub-scans is not suitable, the lowest gray level is coded on the sub-scans common to the two cells of the pair. In both cases, priority is given to the sub-scans corresponding to the low light weight. Such a method performs different codings for the same value while keeping a close proximity between the different codes.

The subject of the invention is a method of displaying a video image on a plasma display panel for a display period, said panel comprising a plurality of cells arranged in rows and columns, each cell being switched on for a duration between zero and a maximum display time corresponding to the maximum brightness of a cell for a given brightness adjustment, the total lighting time of a cell being divided into several lighting periods corresponding to different sub- scans among which a distinction is made between first sub-scans specific to an addressing of each cell and second sub-scans common to two cells arranged on adjacent lines, such that, for a pair of cells sharing the same second sub-scans, gray levels NG1 and NG2 of said cells are decomposed into common value VC and into specific value VS1 and VS2 using the following relation: NG1 = VC + VS1 and NG2 = VC + VS2, with NG1 greater than or equal to NG2, said method comprising the following steps:

E1: coding of the highest gray level NG1 over all of the sub-scans, favoring the sub-scans with the lowest lighting time; E2: extraction of the specific value VS1 corresponding to the coding of step E1;

E3: coding of the lowest gray level using the common value VC resulting from step E1 if the specific value VS1 extracted in step E2 is greater than the difference NG1 - NG2. In some cases, the following step is carried out:

E4: coding of the common value VC as being equal to the lowest gray level NG2 if the specific value VS1 extracted in step E2 is less than the difference NG1 - NG2, then calculation of a new value VS1 = NG1 - NG2.

Preferably, if the maximum value encodable using the first sub-scans is less than the difference NG1 - NG2, then the common value VC is equal to the lowest gray level NG2, and the specific value VS1 is equal to the maximum value encodable on the first subscans.

To control the error due to the use of common sub-scans, prior to any coding operation, the value 1 is optionally added and / or subtracted from one or both of the gray levels NG1 and NG2 so that the difference NG1 - NG2, a multiple of five.

According to a particular embodiment, the display durations associated with the first sub-scans correspond to the product of an elementary duration respectively by the factors: 5, 10, 20, 30, 40, 45, and in that the durations d display associated with the second subscans correspond to the product of the elementary duration by the factors respectively: 1, 2, 4, 7, 13, 17, 25, 36.

The invention also relates to a plasma display panel comprising a plurality of cells arranged in rows and columns, each cell being lit for a period of between zero and a maximum display time corresponding to the maximum brightness of a cell for a given brightness setting, the total lighting time of a cell being divided into several lighting periods corresponding to different sub-scans, among which there are first sub-scans specific to an addressing of each cell and second sub-scans -scans common to two cells arranged on neighboring lines, such that, for a pair of cells sharing the same second sub-scans, the gray levels NG1 and NG2 of said cells are broken down into common value VC and into specific value VS1 and VS2 using the following relation: NG1 = VC + VS1 and NG2 = VC + VS2, with NG1 greater than or equal to NG2, said failure au comprising a gray level encoding device comprising:

a first coding circuit for coding the highest gray level NG1 over all of the sub-scans, favoring the sub-scans with the lowest lighting time;

a means for extracting a common value VC and a specific value VS1 leaving the first coding circuit; a selection and calculation circuit for coding the lowest gray level using the common value VC leaving the first coding circuit if the specific value VS1 extracted from the first coding circuit is greater than the difference NG1 - NG2 . The other particularities of the process are also transposed on the device.

The invention will be better understood and other particularities and advantages will appear on reading the description which follows, the description referring to the appended drawings in which: FIGS. 1 to 4 represent temporal distributions of sub-scans during the display of an image according to the state of the art, FIG. 5 represents the temporal distribution of sub-scans according to a preferred embodiment, FIG. 6 represents a gray level coding algorithm according to the invention, FIG. 7 represents a processing circuit implementing the coding algorithm according to the invention, FIGS. 8 to 10 represent details of the circuit of FIG. 7, FIG. 11 represents a plasma display screen implementing the invention.

For representation reasons, the temporal distribution of the sub-scans uses significant proportions which do not correspond to an exact linear scale.

FIG. 5 represents a preferred time distribution for which an embodiment will be described. This time distribution includes first PSB sub-scans specific to each line which make it possible to address each cell of the screen individually. In the preferred example, there are six first PSB sub-scans with which the respective illumination weights 5, 10 20 30 40 and 45 are associated. Such a choice makes it possible to have a maximum difference value of 150 over 256 levels of gray. A statistical study on video images makes it possible to determine that the probability of error due to the maximum value of difference is much less than 5%.

Second DSB subscans simultaneously address two adjacent lines. In the preferred example, there are eight second DSB subscans with which the respective weights 1, 2, 4, 7, 13, 17, 25 and 36 are associated.

The method of coding the gray levels for each pair of cells will now be described using the algorithm of FIG. 6. The algorithm begins with two known gray levels NG1 and NG2 associated with a first and a second cell.

In a first step 101, the absolute value of the difference between NG1 and NG2 is calculated. This difference | NG1 - NG2 | is then rounded to five to minimize the error, the rounded difference being called D below.

In a second step 102, the values V1 and V2 corresponding to the levels NG1 and NG2 are calculated respectively. These values V1 and V2 are determined on the one hand as a function of the rounding performed on the difference | NG1 - NG2 | and on the other hand according to the minimum and maximum values of NG1 and NG2. In the example described, the rounding of the difference and

After the calculation of the values V1 and V2, comes a third step 103 of encoding. The encoding carried out consists in carrying out on the one hand the coding of the value V1 on all of the sub-scans PSB and DSB by favoring the sub-scans corresponding to the low light weights and on the other hand by coding of the value V2 on the second DSB sub-scans by favoring the sub-scans corresponding to the low light weights. After encoding, there is a 6-bit SPEMAX word which corresponds to the first PSB sub-scans used to code the value V1. The word SPEMAX is associated with the value corresponding to the sum of the weights of the first subscans activated in SPEMAX. There is also an 8-bit COMMAX word which corresponds to the second DSB sub-scans used to code the value V1. The word COMMAX is associated with the value corresponding to the sum of the weights of the second subscans activated in COMMAX. Finally, there is an 8-bit COMMIN word which corresponds to the second DSB sub-scans used to code the value V2. The word COMMIN is associated with the value corresponding to the sum of the weights of the second subscans activated in COMMIN. At the end of the third step 103, a first test 104 is carried out.

The first test 104 verifies that the absolute value of the rounded difference D is less than the maximum difference value DMAX, in the present example DMAX = 150. Those skilled in the art will understand that this test can be carried out as soon as the first step 101 has been carried out and it is not necessary to wait for the end of the third step 103 to carry it out. If the first test 104 indicates D> DMAX, then the fourth step 105 is carried out, if not, a second test 106 is carried out.

The second test 106 checks whether the value which corresponds to SPEMAX is less than or not the difference D. If the value of SPEMAX is less than the difference D then a fifth step 107 is carried out, otherwise a sixth step 108 is carried out.

The fourth to sixth steps 105, 107 and 108 are assignment steps which determine three words Si, Sj and COM. The word Si is a six-bit word which corresponds to the coding of the first PSB sub-scans for the cell having the highest gray level. The word Sj is a six-bit word which corresponds to the coding of the first sub-scans PSB for the cell having the lowest gray level. The word COM is an eight-bit word which corresponds to the coding of the second DSB sub-scans which are common to the two cells. The fourth step 105 assigns the word Si so that it corresponds to the realization of all the first sub-scans PSB, the word Sj so that it corresponds to the realization of none of the first sub-scans PSB, and the word COM so that it is identical to the word COMMIN. This fourth step 104 amounts to encoding the values V1 and V2 considering that V1 = V2 + DMAX. We are then in the case of error due to the maximum difference mentioned above. The combination of a maximum difference value (DMAX = 150 in our example) with the coding used creates a error only for the most important value which corresponds to a high luminance and therefore to a lesser perception for the human eye.

The fifth step 107 affects the word Si so that it corresponds to the realization of the first sub-scans PSB whose total illumination weight corresponds to the difference D. The fifth step 107 also affects the word Sj so that it corresponds to none of the first PSB sub-scans, and the word COM so that it is identical to the word COMMIN.

The sixth step 108 assigns the word Si so that it is identical to the word SPEMMAX, and the word COM so that it is identical to the word COMMAX. The word Sj is defined to correspond to an illumination corresponding to the value of the word SPEMAX minus the difference D.

At the end of the fourth to sixth steps 105, 107 and 108, a third test 109 is carried out to determine which gray level NG1 or NG2 is the highest in order to match in the seventh and eighth steps 110 and 111 the words Si and Sj to the words S1 and S2 which correspond to the first sub-scans PSB for the levels NG1 and NG2 respectively.

The algorithm thus described is repeated for each pair of cells for which the second DSB sub-scans are common.

In order to better understand the operation of the process which is the subject of the invention, a few application examples will now be described. For clarity, the different words corresponding to the coding of the sub-scans are represented in the form of a sum of values, each value corresponding to the activation of the sub-scan associated with said value.

First example: NG1 = 130, NG2 = 124

| NG1 - NG2 | = 6 => D = 5 V1 = 130 = 1 + 2 + 5 + 7 + 10 + 13 + 17 + 20 + 25 + 30

COMMAX = 1 + 2 + 7 + 13 + 17 + 25

SPEMMAX = 5 + 10 + 20 + 30

Value (SPEMMAX) = 65

In this first example, the difference D is less than DMAX and D is less than SPEMAX, it is the sixth step 108 which is carried out, there will therefore be:

If = 5 + 10 + 20 + 30 Sj = 10 + 20 + 30 COM = 1 +2 + 7 + 13 + 17 + 25 NG1 being greater than NG2, we obtain: Code (NG1) = 1 +2 + 5 + 7 + 10 + 13 + 17 + 20 + 25 + 30 Code (NG2) = 1 +2 + 7 + 10 + 13 + 17 + 20 + 25 + 30

Second example: NG1 = 62, NG2 = 130

| NG1-NG2 | = 68 => D = 70

V1 = 131 = 4 + 5 + 7 + 10 + 13 + 17 + 20 + 25 + 30 SPEMMAX = 5 + 10 + 20 + 30

Value (SPEMMAX) = 65

V2 = 61

COMMIN = 2 + 4+ 13 + 17 + 25

In this second example, the difference D is less than DMAX and D is greater than SPEMAX, it is the fifth step 107 which is carried out, there will therefore be:

If = 10 + 20 + 40

Sj = 0

COM = 2 + 4 +13 + 17 + 25 NG1 being less than NG2, we obtain:

Code (NG1) = 2 + 4 + 13 + 17 + 25

Code (NG2) = 2 + 4 + 10 + 13 + 17 + 20 + 25 + 40

Third example: NG1 = 53, NG2 = 242 | NG1-NG2 | = 189 => D = 190

V = 242, V2 = 52

COMMIN = 1 +4 + 7 + 13 + 17

In this third example, the difference D is greater than DMAX, it is the fourth step 105 which applies. If = 5 + 10 + 20 + 30 + 40 + 45

Sj = 0

COM = 1 +4 + 7 + 13 + 17

NG1 being lower than NG2, we obtain:

Code (NG1) = 1 +4 + 7 + 13 + 17 Code (NG2) = 1 +4 + 5 + 7 + 10 + 13 + 17 + 20 + 30 + 40 + 45 In these different examples, those skilled in the art can see that the different pairs of gray levels NG1 and NG2 share a large number of common sub-scans, in particular when said levels NG1 and NG2 are very close. To this first effect is added a grouping of coding values of the same gray level. In other words, the generation of false contours due to the use of torque is also reduced.

As an experiment, if we take the pairs comprising on the one hand the gray level 130 and on the other hand one of the gray levels between 0 and 255, the skilled person can perceive that in 75% cases level 130 is coded on the twelve least significant sub-scans. Although level 130 is coded according to sixteen different codes, the distribution of the sub-scans remains grouped and homogeneous, which eliminates any false contour effect. In 22% of possible cases, one of the two most significant sub-scans is used to code the value 130. However, the different codings have homogeneous distributions on the eleven least significant sub-scans which minimize the effect of false outline.

The cases corresponding to the most harmful coding, ie the simultaneous use of the two most significant sub-scans, correspond to 7% of the possible combinations. Another factor decreasing the consideration by the human eye of this coding defect comes from the fact that the couples causing such a defect correspond to a strong transition which attenuates the effect of false contour. A preferred embodiment of the invention will now be described with reference to FIGS. 7 to 10. FIG. 7 shows an encoding device 200, according to the invention, used to code the gray levels in code control for the different PSB and DSB sub-scans. A plasma display panel may include one or more devices of this type depending on the calculation time required and the number of cells present on said panel.

The encoding device 200 has first and second input buses, for example eight bits for receiving the gray levels NG1 and NG2 corresponding to two cells sharing the same second DSB sub-scans. The gray levels NG1 and NG2 can come from either an image memory containing the entire image, or from a decoding device which decodes a video signal and which translates it into gray level for each cell. The encoding device 200 has three output buses which supply the words COM, S1 and S2 which correspond respectively to ignition or non-ignition codes for the second DSB sub-scans, for the first PSB sub-scans. associated with the first gray level NG1 and for the first sub-scans associated with the second gray level NG2.

The encoding device 200 comprises a difference circuit 201 which receives the two gray levels NG1 and NG2 to be encoded and provides on a first output the absolute value of the difference between NG1 and NG2. In addition, on a second output of said difference circuit 201, an information bit SelC indicates what is the gray level NG1 or NG2 which is to be considered as greater than the other.

The difference circuit 201 is for example constituted as shown in FIG. 8. First and second subtraction circuits 301 and 302 receive the gray levels NG1 and NG2 on opposite inputs, so that the first subtraction circuit 301 provides on a result output the difference NG1 - NG2 and that the second subtraction circuit 302 provides the difference NG2 - NG1 on a result output. The second subtraction circuit also has an overflow output (also known as a carry output) which makes it possible to know whether the result of the subtraction is positive or negative and therefore provides the information bit SelC. A multiplexer 303 receives the SelC information bit on a selection input and has first and second inputs connected to the result outputs of the first and second subtraction circuits 301 and 302 respectively. The multiplexer 303 selects the positive result as a function of the information bit SelC so that the output of the multiplexer 303 corresponds to the output of the difference circuit 201.

The encoding device 200 further comprises a first comparison circuit 202 which compares the absolute value of the difference | NG1 - NG2 | with the maximum DMAX difference value which is set by the sub-scans used. The first comparison circuit 202 provides a first selection signal SelA which corresponds to the result of the first test 104. A person skilled in the art can notice that it is not necessary to round to 5 to carry out this comparison because the final result remains equivalent before or after rounding.

A rounding circuit 203 receives the absolute value of the difference | NG1 - NG2 | to round it to 5. A first exit provides the difference rounding D and a second output provides a rounding control bus. The rounding control bus indicates how the values V1 and V2 should be changed. The rounding circuit 203 can be produced using a correspondence table of which a part of the output bits corresponds to the rounded difference D and another part of the output bits corresponds to a command code.

A calculation circuit 204 receives the gray levels NG1 and NG2 and supplies the values V1 and V2 which will be used for coding. The calculation circuit 204 receives for this purpose the information bit SelC to make the highest level NG1 or NG2 correspond to the value V1 and the lowest level NG1 to the value V2. the calculation circuit 204 also receives the control bus coming from the rounding circuit 203 to carry out, if necessary, an addition or a subtraction of a unit on V1 and / or V2.

A first coding circuit 205 receives the value V1 and provides complete coding of this value over all of the sub-scans PSB and DSB. However, a six-bit bus supports the word SPEMAX corresponding to the first PSB subscans and an eight-bit bus supports the word COMMAX corresponding to the second DSB subscans. For the sake of simplicity and speed of calculation, a correspondence table is used which contains optimum coding.

A second coding circuit 206 receives the value V2 and provides coding of this value on the second DSB sub-scans only. The output bus of this second coding circuit 206 supports the word COMMIN. This second coding circuit can also be carried out using a correspondence table. Those skilled in the art will note that only a limited number of different values are actually to be coded and that it is therefore not necessary to use a table having more than seven input bits.

A second comparison circuit 207 receives on the one hand the rounded difference D and on the other hand the word SPEMAX. This second comparison circuit will compare the value associated with the word SPEMAX with the rounded difference D in order to provide on a first output a second selection signal SelB which corresponds to the result of the second test 106. The second comparison circuit also provides on a second output a six-bit word corresponding to the first sub-scans and having an associated illumination which corresponds either to the rounded difference D or to the illumination associated with SPEMAX from which the rounded difference D has been removed. An exemplary embodiment of the second comparison circuit 207 is shown in FIG. 9. The second comparison circuit 207 comprises a decoding circuit 401, a subtraction circuit 402, a multiplexer 403 and a coding circuit 404. The decoding circuit 401 is for example a correspondence table receiving the six bits of the word SPEMAX and providing an eight-bit word corresponding to the value representative of the illumination associated with the word SPEMAX. The subtraction circuit 402 has two inputs receiving respectively the rounded difference D and the word leaving the decoding circuit 401, so that it provides on its output the result of the difference corresponding to the value of SPEMAX minus D. The circuit 402 has an overflow or holdout output that indicates whether the result of the subtraction is positive or negative. Said overflow output thus provides the second selection signal SelB. The multiplexer 403 has two input buses receiving respectively D and the result leaving the subtraction circuit 402. A selection input of the multiplexer 403 receives the second selection signal so that the output bus of the multiplexer 403 provides D if D is greater than the value of SPEMAX or the result leaving the subtraction circuit 402 otherwise. The coding circuit is a correspondence table which has an input connected to the output of the multiplexer 403 to receive either D or SPEMAX - D, in order to provide a six-bit code corresponding to the coding of the input value using the first PSB subscans.

The encoding device 200 comprises a selection circuit 208 which receives on the one hand the different words COMMIN, COMMAX, SPEMAX and the word leaving the second comparison circuit, and on the other hand the first and second selection signals SelA and SelB. Said selection circuit provides, on first and second outputs, first and second words Si and Sj of six bits corresponding to the codings of the first sub-scans for the two gray levels NG1 and NG2, and, on a third output, a third eight-bit COM word corresponding to the common coding of the second sub-scans for the two gray levels NG1 and NG2.

An exemplary embodiment of the selection circuit 208 is shown in FIG. 10. The circuit 208 comprises a decoder 501 and three multiplexers 502 to 503. The decoder circuit 401 receives the first and second selection signals SelA and Sel B and provides the commands necessary for the three multiplexers 502 to 504 to carry out the connections defined in the fourth to sixth steps 105, 107 and 108.

The multiplexer 502 has two inputs which receive the words COMMIN and COMMAX and one output which supplies the word COM. The word COM corresponds either to COMMIN when the first selection signal SelA indicates that D is greater than DMAX or when the second selection signal SelB indicates that the value of SPEMAX is less than D, or to COMMAX when the first and second signals indicate that D is not greater than DMAX and that the value of SPEMAX is not less than D.

Multiplexer 503 has two inputs and one output. One of said inputs receives the six-bit word from the second comparison circuit 207, the other of said inputs receives a six-bit word corresponding to the zero value encoded for the first subscans, and the output provides the word Sj. The word Sj corresponds either to the zero value coded on six bits when the first selection signal SelA indicates that D is greater than DMAX or when the second selection signal SelB indicates that the value of SPEMAX is less than D, or to the word leaving the second comparison circuit 207 COMMAX when the first and second signals SelA and SelB indicate that D is not greater than DMAX and that the value of SPEMAX is not less than D.

Multiplexer 504 has first to third inputs and an output. The first input receives a word corresponding to DMAX, ie corresponding to the maximum possible illumination using the first PSB sub-scans. The second entry receives the word SPEMAX. The third input receives the outgoing word from the second comparison circuit 207. The output provides the word Si which corresponds either to the word corresponding to DMAX when the first signal SelA indicates that D is greater than DMAX, or to SPEMAX when the first and second signals SelA and SelB indicate that D is not greater than DMAX and that SPEMAX is not less than D, that is to say the word leaving the second comparison circuit when the signals SelA and SelB indicate that D is not greater than DMAX and that SPEMAX is less than D.

The encoding device 200 comprises an output circuit 209 which receives the words Si and Sj to make them correspond either to the words S1 and S2 respectively, or to the words S2 and S1 respectively according to the information bit SelC. The encoding device 200 is then incorporated into a display panel 600 to allow the display of image 601, as shown in FIG. 11.

Such an encoding device 200 can be produced according to different variants. By way of example, if a person skilled in the art considers that the computation time is too low, it is for example possible to adopt a structure of the pipeline type. To this end, it is possible, for example, to add storage registers 210 as indicated in FIG. 7 to perform the calculation in two stages, which allows an overall reduction in the calculation time for an image.

Many other alternatives are available to those skilled in the art. In the preferred example, correspondence tables are used to carry out the coding and decoding for reasons of simplicity of implementation and therefore of reliability. It goes without saying that these correspondence tables can be replaced by calculation circuits, in particular if it is chosen to implement such a device using circuits of the microcontroller type.

More generally, a person skilled in the art can also be content to carry out the method of the invention only using programmed circuits essentially comprising a processor and a memory. The device thus produced will have a completely different structure from the device shown.

Also, in the present description of the invention, reference is made to a coding using six first subscans whose respective weights are 5, 10, 20, 30, 40 and 45 and eight second subscans whose respective weights are 1, 2, 4, 7, 13, 17, 25 and 36. This coding has been chosen for the present description because it allows good results to be obtained. For the sake of clarity, no other types of coding have been referred to during the description, but it is obvious that other types of coding may be used with the same method provided that certain numerical values are changed.

By way of additional example, one can for example use a coding for sixteen sub-scans comprising four first sub-scans whose respective weights are 5, 10, 20 and 35 and ten second sub-scans whose respective weights are 1, 2, 4, 6, 9, 12, 15, 19, 23, 27, 31 and 36, taking care to modify the different quantities (DMAX value, number of coding bits) which have been used in the description.

Claims

claims

1. A method of displaying a video image (601) on a plasma display panel (600) for a display period, said panel comprising a plurality of cells arranged in rows and columns, each cell being lit for a duration between zero and a maximum display time corresponding to the maximum brightness of a cell for a given brightness adjustment, the total lighting time of a cell being divided into several lighting periods corresponding to different sub -scans among which there are first sub-scans (PSB) specific to an addressing of each cell and second sub-scans (DSB) common to two cells arranged on adjacent lines, such that, for a pair of cells sharing the same second sub-scans (DSB), the gray levels NG1 and NG2 of said cells are broken down into common value VC and into specific value VS1 and VS2 using the following relation ante: NG1 = VC + VS1 and NG2 = VC + VS2, with NG1 greater than or equal to NG2, characterized in that the following steps are carried out: E1: coding of the highest gray level NG1 over all of the sub -scans (PSB and DSB) by favoring the sub-scans whose lighting time is the lowest;

E2: extraction of the specific value VS1 corresponding to the coding of step E1;

E3: coding of the lowest gray level NG2 using the common value VC resulting from step E1 if the specific value VS1 extracted in step E2 is greater than the difference NG1 - NG2.

2. Method according to claim 1, characterized in that the following step is carried out: E4: coding of the common value VC as being equal to the lowest gray level NG2 if the specific value VS1 extracted in step E2 is less than the difference NG1 - NG2, then calculation of a new value VS1 = NG1 - NG2.

3. Method according to one of claims 1 or 2, characterized in that, if the maximum encodable value (DMAX) using the first sub-scans (PSB) is less than the difference NG1 - NG2, then the value common VC is equal to the lowest gray level NG2, and the specific value VS1 is equal to the maximum encodable value (DMAX) on the first sub-scans (PSB).

4. Method according to one of claims 1 to 3, characterized in that prior to any coding operation, the value 1 is optionally added and / or subtracted from one or both of the gray levels NG1 and NG2 so that the difference NG1 - NG2 is a multiple of five.

5. Method according to one of claims 1 to 4, characterized in that the display durations associated with the first sub-scans (PSB) correspond to the product of an elementary duration by respectively the factors: 5, 10, 20, 30, 40, 45, and in that the display durations associated with the second sub-scans (DSB) correspond to the product of the elementary duration by the factors respectively: 1, 2, 4, 7, 13, 17, 25, 36.

6. Plasma display panel (600) comprising a plurality of cells arranged in rows and columns, each cell being lit for a period of between zero and a maximum display time corresponding to the maximum brightness of a cell for one given brightness setting, the total lighting time of a cell being divided into several lighting periods corresponding to different sub-scans, among which a distinction is made between first sub-scans (PSB) specific to an addressing of each cell and second sub-scans (DSB) common to two cells arranged on neighboring lines, such that, for a pair of cells sharing the same second sub-scans (DSB), the gray levels NG1 and NG2 of said cells are broken down into common value VC and in specific value VS1 and VS2 using the following relation: NG1 = VC + VS1 and NG2 = VC + VS2, with NG1 greater than or equal to NG2, characterized in that q it includes a gray level encoding device (200) comprising:

- a first coding circuit (205) for coding the highest gray level NG1 over all of the sub-scans (PSB and DSB) by favoring the sub-scans with the lowest illumination time; - means (401) for extracting a specific value VS1 leaving the first coding circuit;

- a selection and calculation circuit (207 and 208) for coding the lowest gray level using the common value VC leaving the first coding circuit (205) if the specific value VS1 extracted from the first coding circuit is greater than the difference NG1 - NG2.

7. Panel according to claim 6, characterized in that it comprises a second coding circuit (206) for coding the common value VC as being equal to the lowest gray level NG2, and in that, if the specific value VS1 extracted from the first coding circuit is less than the difference NG1 - NG2, the selection and calculation circuit determines a new value VS1 = NG1 - NG2.

8. Panel according to one of claims 6 or 7, characterized in that it comprises a comparison circuit (202) for comparing the difference NG1 - NG2 with a maximum encodable value (DMAX) on the first sub-scans (PSB ), and in that the selection and calculation circuit determines that VS1 is equal to the maximum encodable value (DMAX) on the first subscans and that the common value is equal to the lowest gray level.

9. Panel according to one of claims 6 to 8, characterized in that it comprises a rounding circuit (203 and 204) placed upstream of the first and second coding circuits, said rounding circuit adding and / or possibly subtracting the value 1 from one or both of the gray levels NG1 and NG2 so that the difference NG1 - NG2 is a multiple of five.