EP1512133A1

EP1512133A1 - Method and device for determining a frequency for sampling an analog signal

Info

Publication number: EP1512133A1
Application number: EP03758016A
Authority: EP
Inventors: Martin Maier
Original assignee: Philips Intellectual Property and Standards GmbH; Koninklijke Philips Electronics NV
Current assignee: Philips Intellectual Property and Standards GmbH; Koninklijke Philips NV
Priority date: 2002-11-21
Filing date: 2003-10-17
Publication date: 2005-03-09
Anticipated expiration: 2023-10-17
Also published as: EP1512133B1; US7257499B2; US20050179571A1; DE10254469A1; DE50304583D1; JP2006506669A; WO2004047063A1; DE10254469B4; TW200419500A; AU2003274038A1; TWI274313B

Abstract

The invention relates to a method and device for determining a frequency for sampling an analog signal, which is provided to a digital display screen (134) for displaying an image on the same. According to the invention, at least two areas (1380 - 1386) that are consecutive in a line direction are firstly established in the image (136) to be displayed. In each of the established areas (1380 - 1386), a sampling phase is determined for which a contrast in the established area (1380 - 1386) is at a maximum or minimum. Afterwards, a local progression of the sampling phase in the line direction is determined on the basis of the determined sampling phases. The sampling frequency is determined on the basis of a fundamental value and on a modification value that is derived from the local progression of the sampling phase.

Description

Method and device for determining a frequency for sampling an analog signal

description

The present invention relates to a method and apparatus for determining a frequency for sampling an analog image, and more particularly to a method for determining a frequency for sampling an analog signal provided to a digital screen to capture an image the digital screen. Furthermore, the present invention relates in particular to a device for generating digital data from analog image data in order to display an image on a digital screen based on the generated image data.

Conventional computers comprise elements, e.g. B. graphics cards to graphics information generated in the computer, e.g. B. pictures, for display on an external device, e.g. B. provide a screen. The graphics cards conventionally used generate, based on the digital signals provided by the computer or its processing unit (CPU), corresponding image data suitable for controlling a screen. In many applications, the display device associated with the computer, the screen, is an analog screen that has a cathode ray tube. In order to provide the necessary data for this application, which only existed until a few years ago, the graphics card comprises a digital / analog converter in order to convert the image data generated by the graphics card into an analog signal, for example an RGB signal, which then activates it of the screen. In addition to the analog image data signals (RGB signals), the horizontal and vertical synchronization signals are also output to the screen, which are necessary for the correct reproduction of the image data on the screen.

Recently, however, so-called digital screens have increasingly been used, for example LCD screens or LCD monitors (LCD = Liquid Crystal Display), which, in contrast to screens with cathode ray tubes, require digital control. In this case, it is possible to digitally process an analog video signal present on an analog video output of a computer / computer in the screen / monitor. This requires digitizing the analog video signal again at a sampling frequency. In order to reconstruct the output data with the most accurate sampling frequency, it is therefore desirable to sample the analog signal with the original frequency and with the correct phase position, i.e. with the frequency and phase position with which the analog video signals at the output of the computer from the digital data in the graphics card were generated. The phase position here denotes the shift of the scanning signal relative to the generated scanning signal, the phase position being generally given in degrees, e.g. B. 0 degrees, which corresponds to no shift, or 180 degrees, which is equivalent to a shift by half a clock period.

In Fig. 1, the course of an analog video signal (see Fig. 1A) at the input of a digital screen is shown schematically. Likewise, an ideal sampling clock for the sampling of this applied analog signal is shown in FIG. 1B. T denotes a period of the sampling clock.

While the generation of images on analog screens using the analog video signals generated by the graphics card is generally problem-free, in particular does not lead to any visible artifacts, the re-scanning of the data originally on a digital len signal-based analog signal is a problem, since it can come to artifacts in the displayed image due to the renewed scanning on the digital screen, which are visible to a viewer. In order to avoid such artifacts, various approaches are known in the prior art, which will be briefly discussed below.

In U.S. For example, patent 6,268,848 describes a method by means of which visible errors in an image which is displayed on a digital monitor can be avoided by using an automatic scanning control system in which one for successive image frames whose image content remains essentially the same Phase of the sampling clock, for re-sampling of the received analog signal, as long as changes are made until a maximum sample value is reached. The phase value reached at the maximum sample value then represents the optimal phase shift of the sampling clock for the sampling of this frame.

The U.S. Patent 6,147,668 describes a digital display unit by means of which display artifacts, which are caused by high-frequency interference in analog display signals due to the aliasing effect, are avoided or minimized. Similar to U.S. -Patent 6,268,848 a modulation is also carried out here in order to apply different phase delays to the scanning clock signal for successive lines or frames, so that due to this modulation the analog display signal for a display on the digital display element is sampled at different sampling points for the same pixel in different frames becomes.

In the approaches described above, as can be seen, only one sampling phase is varied, whereas the sampling frequency remains unchanged. The approaches described in the two U.S. patents above use sampling clocks, which are based on the combined with the analog video provided horizontal and vertical synchronization signals are derived. The synchronization signals represent the reference signal for the digital screen, with which a clock generator is locked in the screen or in the screen controller in order to generate a suitable sampling clock based on the reference signal.

Conventionally, the generation of the reference signal for the clock generator takes place in such a way that, based on the received synchronization signals of the analog signal, a look-up table is accessed from which a suitable / ideal reference value for these synchronization signals is then selected, which is then used by the clock generator as a reference clock or Reference frequency for generating the sampling clock is provided.

The above approaches work well if it is ensured that the synchronization signals or the reference signal which is assigned to the analog signal actually reproduces the frequency of the digital signal on the basis of which the analog signal was generated. In this case, the sampling clock generated by the clock generator in the digital screen or in the control thereof corresponds to this frequency. However, this boundary condition does not apply to all graphics cards and is usually only fulfilled for very sophisticated graphics cards. Other graphics cards, e.g. B. cheaper graphics cards have tolerances that lead to the fact that the frequency used in the graphics card has deviations from the frequency which is signaled to the digital screen as the optimal / ideal sampling frequency. Traditionally, these deviations range from 1% to 5% of the sampling frequency actually signaled to the screen.

In such cases, the approaches described above are for sampling analog signals in digital screens To avoid artifacts or interference in the display of the image, it can only be used to a limited extent, since there is a frequency error in the sampling of the analog signal which requires further correction.

Starting from this prior art, the present invention is therefore based on the object of providing a method and a device which make it possible to generate a sampling frequency for the renewed digitization of an analog signal which is well matched to the frequency of a digital signal which was the basis of the analog signal.

This object is achieved by a method according to claim 1 and by an apparatus according to claim 11.

The present invention provides a method of determining a frequency for sampling an analog signal provided to a digital screen to display an image on the digital screen, comprising the steps of:

(a) specifying at least two areas in succession in the line direction in the image to be displayed;

(b) determining a sampling phase in each of the specified areas for which a contrast in the specified area is maximum or minimum;

(c) determining a local course of the sampling phase in the row direction, based on the sampling phases determined in step (b) in the defined areas _f and

(d) Determining the sampling rate, based on a basic value and a modification value which is derived from the local course of the sampling phase determined in step (c). When determining the sampling phase in accordance with step (b), a sampling phase is determined in each of the defined areas with which the best or worst scanning is achieved, and the contrast in the defined area is thus maximum or minimum

The present invention also provides an apparatus for generating digital data from analog image data to display an image on a digital screen based on the generated image data

an A / D converter that includes a data input for receiving the analog image data, a data output for outputting the digital image data, and a clock input;

a clock generator having a clock output for outputting a clock signal and a control input for receiving a clock frequency control signal;

a phase shifter including a clock input for receiving the clock signal from the clock generator, a clock output for outputting a phase-shifted clock signal to the clock input of the A / D converter, and a control terminal for receiving a control signal defining a phase shift; and

a controller having an input for receiving the digital data from the A / D converter, a first control output for outputting the clock frequency control signal to the clock generator, and a second control input for outputting the signal defining the phase shift to the phase shifter, the control device being operative to based on the digital data provided at the entrance

to define at least two areas in succession in the line direction in the image to be displayed, to determine a sampling phase in each of the areas for which a contrast in the defined area is maximum or minimum,

to determine a local course of the sampling phase in the row direction based on the determined sampling phases,

determine the sampling frequency based on a basic value and a modification value which is derived from the local course of the sampling phase, and

- To generate the clock frequency control signal according to the determined sampling frequency.

According to a preferred embodiment of the present invention, the sampling phase, which has the maximum or minimum contrast in a defined range, is generated by generating a plurality of reference values with different sampling phases and the same sampling frequency, the reference value being the sum of the absolute differences of successive intensity values in the defined range is defined. A maximum or minimum reference value is selected from the reference values generated in this way, a maximum or minimum contrast being defined by the maximum or minimum reference value.

According to another preferred exemplary embodiment of the present invention, the sampling phase, which has the maximum or minimum contrast in a defined area, is generated by carrying out a first measurement in each of the regions under consideration at a defined sampling phase and a defined sampling frequency to get a first reference value for each of the ranges. A second measurement is then performed in each of the areas under consideration to obtain a second reference value for each of the areas. For each of the areas under consideration, a difference between the reference values obtained by the first measurement and the second measurement is generated. This measurement is carried out on a plurality of different sampling phases / phase values in order to obtain a plurality of difference values. Finally, the maximum difference value, which indicates a minimum contrast, or the minimum difference value, which indicates a maximum contrast, is selected from the plurality of obtained difference values for each of the regions under consideration. Alternatively, any number of measurements can be carried out for each of the areas and for each of the sampling phases, on the basis of which multiple difference values for each area are then obtained.

According to a first preferred exemplary embodiment, the determination of the local course and the sampling frequency first comprises the determination of a straight line which runs through the determined best or worst sampling phases. The slope is then determined for this straight line. The modification value is determined based on the slope of the straight line, and the sampling frequency is obtained by adding the basic value and the modification value, a sign of the modification value depending on whether the straight line is rising or falling, that is to say the slope has a positive or negative sign. In an alternative embodiment, straight sections and jumps are determined in the course of the sampling phases and the number of jumps in the course is detected. The modification value then corresponds to the number of hops, and the sampling frequency is again obtained by adding the basic value and the modification value. To determine the sign of the modification value, determine whether the straight sections are rising or falling in the local course.

Preferred developments of the invention are defined in the subclaims. Preferred exemplary embodiments of the present invention are explained in more detail below with reference to the accompanying drawings. Show it:

Fig. 1 shows the course of an analog signal in Fig. 1A on

Input of a digital screen, and an ideal sampling clock for sampling the analog input signal in Fig. 1B;

2 shows a block diagram of an apparatus for generating a sampling frequency according to a preferred exemplary embodiment of the present invention;

3 is a representation of a screen with an active image, in which a plurality of measuring ranges which are used for frequency determination according to the present invention are shown;

FIG. 4 shows an example for the determination of a bad reference value (FIG. 4A) and a good reference value (FIG. 4B), which is used to determine the sampling phases; and

5 shows the local course of the best sampling phases for the plurality of areas in FIG. 3.

A preferred exemplary embodiment of the device according to the invention is explained in more detail below with reference to FIG. 2. A detailed description of the preferred exemplary embodiments of the method according to the invention is then given with reference to the block diagram shown in FIG. 2.

FIG. 2 shows the block diagram of a control device as used, for example, in the input stage of a digital screen, for example an LCD screen. The device comprises an analog / digital converter (ADC) 100, which receives at an input 102 an analog input signal, for example an analog video signal, from a graphics card of a computer or computer. At a clock input 104, the analog / digital converter 100 receives a clock signal, on the basis of which the analog / digital converter carries out a sampling of the analog signals received at the input 102. The generated, digitized signals are provided by the analog / digital converter 100 at its data output 106. The data generated by the analog / digital converter 100 are provided at the output 106 of the same of a data line 108. The clock signal present at the clock input 104 of the analog / digital converter 100 is carried on a clock line 110. The data line 108 and the clock line 110 extend further to the display element of the digital screen in order to provide the same with the data signals and clock signals required for the display.

Furthermore, the arrangement according to FIG. 2 comprises a clock generator 112, which receives a clock frequency control signal at a control input 114. At an output 116 of the clock generator 112, the latter outputs a clock signal generated as a function of the control signal present at the control input 114.

A phase shifter 118 is provided which receives the clock signal generated by the clock generator 112 at an input 120. Furthermore, the phase shifter 118 has a control input 122, at which it receives a control signal which defines a phase shift with which the clock signal received by the clock generator 112 is to be applied. The phase-shifted clock signal is then provided at an output 124 of the phase shifter. The output of the phase shifter 124 is connected to the input of the analog / digital converter 100 via the clock line 110. Furthermore, the device comprises a regulation / control 126, which receives the data signal generated by the analog / digital converter at a first input 128, which is connected to the data line 108. The control operates to provide the clock frequency control signal at a first control output 130. The controller 126 is likewise effective in order to provide the signal for the phase shifter 118 that defines the phase shift at a second control output 132.

The controller 126 operates according to the method according to the invention, the control signals required for the execution of the method according to the invention for the clock generator and the phase shifter being carried out, for example, on the basis of sequence controls / algorithms implemented in the controller 126. The controller 126 further comprises a signal processing unit in order to process and evaluate the data signals received at the input 128.

A preferred exemplary embodiment of the method according to the invention is explained in more detail below with reference to the device shown in FIG. 2.

As described above, the method according to the invention assumes that an ideal sampling frequency signaled to the digital screen for resampling the analog input signal by the analog / digital converter 100 was not the actual frequency of the digital signal on which the analog signal was based. Rather, it can be expected that due to the tolerances of the graphics card used to generate the analog signal, there will be deviations from the ideal frequency in the range of at most 1% to 5%. This deviation makes it necessary to carry out a modification of the ideal sampling frequency in order to carry out a re-sampling / redigitization of the analog input signal in such a way that an image defined by the analog input data is properly, in particular which can be displayed on the digital screen without any visible errors.

In order to determine the frequency required for sampling the input data generated by a specific device (graphics card), areas of the analog signal that are repeated are considered according to the invention. In fact, static frames are used for the method according to the invention, and in this same frame, for example, a single or several screen lines are viewed. For the method according to the invention, therefore, the same image / the same frame is preferably provided for multiple scanning for determining the optimal scanning frequency. Furthermore, the period of the sampling clock provided to the analog-to-digital converter 100 is an integer divisor of the duration of the repetitive region of the analog signal, the horizontal period being a multiple of the pixel period which is achieved by means of a PLL circuit is produced.

The sampling frequency and also the sampling phase can now be determined from the digital video data on the data line 108 by means of the control and measurement loop shown in FIG. 2.

The method according to the invention for determining the sampling frequency is based on a method for determining the best / worst sampling phase, but is independent of how this best / worst sampling phase is actually determined. For example, to determine the best or worst sampling phase, US patents 6,268,848 and 6,147,668 mentioned in the introduction to the description can be used, which disclose two approaches for determining the best / worst sampling phase. For the frequency determination, either a method which determines the worst sampling phase or a Method that determines the best sampling phase can be used.

In the following description of the preferred exemplary embodiments it is assumed that a method is used for the frequency determination which determines the best sampling phase. A method which is based on the determination of the worst sampling phase can be used analogously.

To carry out the method according to the invention, a “measurement” (sampling) of the analog data of the stationary frame present at the input 102 of the analog / digital converter 100 is first carried out with a freely selected sampling frequency. An error is then calculated on the basis of the data signals received, which indicates the deviation of the selected sampling frequency from the known, ideal sampling frequency (see above). Regarding the freely selected sampling frequency, it should be noted that this can in principle be chosen arbitrarily. However, in order to obtain a result in a short time than after a short calculation time , the freely selectable sampling frequency is chosen to approximately correspond to the expected deviation. Preferably, the freely selectable sampling frequency is chosen to correspond to an expected frequency. For example, for a graphics card used, deviations from the optimal frequency are in the range from ± 1% to ± 5% expected, the freely selected sampling frequency is preferably selected in this range around the optimal sampling frequency.

After the repeating analog signal range is M sampling clocks wide, the sampling frequency can be specified as M clocks, with M being the number of pixels per horizontal line of the digital screen in the preferred exemplary embodiment. To determine the frequency, that is to say to determine the actual sampling frequency, a plurality of N regions (N> 2) are now selected in the active screen region. 3 shows a screen which shows an active image in which a plurality of measuring ranges are shown.

3 schematically shows the display area 134 of the digital screen which, as described above, is M pixels wide, that is to say M pixels in each horizontal line. An active image 136 shown on the screen 134 is also shown in FIG. 3. A plurality of measuring ranges 138 ₀ to 138 _{6 are} shown in the active image 136. These areas 138o to 1386 are used for frequency determination. The best sampling phase is determined in these areas, as will be described below. In the exemplary embodiment shown in FIG. 3, seven areas 138 ₀ to 138 _{6 are} shown, but the present invention is not restricted to this number. In fact, it is sufficient if at least two areas are selected, but the accuracy increases with the number of selected areas. The ranges 138 ₀ to 138δ are also selected with respect to the position depending on the expected frequency error, namely in such a way that they have a predetermined distance in the row direction depending on the expected frequency error. Two areas arranged consecutively or adjacent in the line direction should have a distance which is less than or equal to the predetermined distance, which is generally defined as a function of the assumed error in the scanning in a corresponding number of pixels.

Furthermore, the areas are preferably selected such that image areas are determined in which the best scanning phase is easiest to determine, which is very easily possible, for example, in areas with high contrast. As can be seen from Fig. 3, it is not mandatory required that all measuring ranges 138 ₀ to 1386 are assigned to the same line of the image. As shown in the specific application, these can actually be arranged in different lines.

In the areas 138o to 138 ₆ determined in FIG. 3, for example, a best sampling phase is first determined according to the invention. The best sampling phase is determined using the method described in more detail below.

A so-called reference value RW is calculated over the defined ranges 138 ₀ to 138 _{6 of} the repetitive range of the digitized input signal. For the same sub-areas - the analog signal repeats itself - the associated reference values are determined with different sampling phases. In this case, the controller 126 (see FIG. 2) operates to keep the frequency control signal constant at the output 130 and to provide different phase shift signals at the output 132 for the different calculation sections. The maximum or largest reference value is then obtained for the best phase setting in a range, whereas the minimum / lowest reference value is set for the worst phase setting.

The reference value is calculated from the sum of the absolute difference between two successive sample values of all sample values located in one of the measuring ranges. The measuring range can be small up to a measurement of two samples or extend over several lines of a frame.

The reference value is calculated according to the following calculation rule:

RW = ∑ | X „- X _{n + 1} | n

RW = reference value, n = number of samples in the area under consideration, x = intensity value of a sampled pixel.

This reference value is therefore a value which increases as the contrast increases. The best sampling phase is the sampling phase in which the contrast assumes a highest / maximum value. The advantage of the method for reference value calculation just described is that no line or image memory is required here in order to recognize whether the contrast becomes better or worse with the changed phase.

To make small differences, e.g. B. to suppress analog noise, it can be provided to add up only those differences that are greater than a predetermined threshold.

4 shows an example for the determination of a good and a bad reference value. FIG. 4A shows a sampling of the analog input signal with a fixed sampling frequency (see period T), in which the sampling phase is selected such that two adjacent digital values of 0.8 and 0.3 result in the sampling, which results in leads to a reference value of 0.5.

Fig. 4B shows the sampling of the same analog signal with the same frequency (see period T), but with a sampling phase, which lead to a digital sample of 1.0 and an adjacent sample of 0.0, so that a large reference value of RW = 1.0 results, which reflects a high contrast between the two sampled points in the analog signal. Thus, FIG. 4A shows a scan with a bad scan phase, and FIG. 4B shows the scan with a good scan phase. Assuming that the reference value achievable in FIG. 4B is the maximum reference value, this is then used as the basis for the further process for the area under consideration. With an execution form of the present invention, in which the worst sampling phase would be used instead of the best sampling phase, instead of using the reference value determined in FIG. 4B, the reference value determined in FIG. 4A would continue to be used as the minimum reference value.

In an alternative exemplary embodiment, different measurements can be carried out for the same sub-areas with the same phase setting in order to obtain a plurality of reference values for each of the areas. The differences between the different reference values are then formed in each area. A maximum difference value indicates the worst phase setting in one area and a minimum difference value indicates the best phase setting in one area. The reason for this is the sampling clock jitter, since the analog signal changes the least in the area of the best sampling, so there is also the smallest difference. More specifically, according to the invention, in this exemplary embodiment, a first measurement is first carried out in each of the areas under consideration at a fixed sampling phase and a fixed frequency. A second measurement is then carried out in each of the areas under consideration. The difference between the measured values obtained by the first and second measurements is then generated. The foregoing steps are repeated at different phase settings to obtain a plurality of difference values from which the maximum difference value indicating a minimum contrast or the minimum difference value indicating a maximum contrast value are selected for each area.

Now that the best sampling phase or worst sampling phase has been generated and determined for each of the regions, the frequency determination is now carried out based on the sampling phases thus detected. For this purpose, the measured values obtained are graphically represented in a coordinate system. For this purpose, as shown in FIG. 5, as an x value (Abscissa) the number of the mean sample value of the measuring range is used, and the specific sampling phase assigned to this area is plotted on the y-axis (ordinate). This results in the best / worst phase values recorded in FIG. 5 for the sampled values under consideration, which were determined in the manner described above.

The points relating to the best sampling phases plotted over the x axis are then connected with a straight line, as shown in FIG. 5, and the gradient S of the straight line is now determined in degrees per sample value using known mathematical methods. For example, the slope is determined according to the following calculation rule:

Δ deg

S = Δx

When calculating S, however, jumps have to be taken into account, in which the sampling phase values jump between a minimum (0 deg) and a maximum (360 deg) value, as is indicated in FIG. 4.

After the slope of the straight line has been determined, the correct sampling frequency can be determined according to the following calculation rule:

With:

M = ideal sample value,

ΔM = modification value,

S = slope, and

Mn = corrected frequency value, Alternatively, the corrected or correct sampling frequency can also be determined by determining the number of jumps in the course of the sampling phases in the M sampling cycles. This value then corresponds to the absolute value of ΔM. The sign is determined by determining whether the straight line is rising or falling.

If the sampling frequency is set correctly, the best sampling phase results for each of the N areas.

Claims

claims

1. A method for determining a frequency for sampling an analog signal, which is provided to a digital screen (134) to display an image (136) on the digital screen (134), comprising the following steps:

(a) specifying at least two areas (138 ₀ to 138 ₆ ) successively in the row direction in the image (136) to be displayed;

(b) determining a sampling phase in each of the specified ranges (138 ₀ to 138g) for which a contrast in the specified range (138 ₀ to 138 ₆ ) is maximum or minimum;

(c) determining a local course of the sampling phase in the row direction, based on the sampling phases determined in step (b) in the defined regions (138 ₀ to 138 ₆ ); and

(d) determining the sampling frequency (Mn) based on a basic value (M) and a modification value (ΔM) which is derived from the local course of the sampling phase determined in step (c).

2. The method of claim 1, wherein step (c) comprises the following steps:

(c.l) determining a straight line on which the sampling phases determined in step (b) lie; and

(c.2) determining the slope (S) of the straight line; and

where step (d) comprises the following steps: (dl) determining the modification value (ΔM) based on the slope (S) of the straight line; and

(d.2) Determine the sampling frequency (Mn) by adding the basic value (M) and modification value (ΔM).

3. The method according to claim 2, wherein the modification value (ΔM) is determined according to the following calculation rule:

With

ΔM modification value,

S slope of the straight line and

M basic value.

The method of claim 1, wherein step (c) comprises the following steps:

(c.l) determining straight sections and jumps in the local course of the sampling phase; and

(c.2) determining the number of jumps in the local course of the sampling phase, the course alternating with a jump between a maximum value and a minimum value of the sampling phase; and

where step (d) comprises the following steps:

(dl) determining the modification value (ΔM) based on the number of jumps; and (d.2) Determining the sampling frequency (Mn) by adding the basic value (M) and the modification value (ΔM), the sign of the modification value (ΔM) being positive or negative, depending on whether the straight sections of the local course the sampling phase is rising or falling.

5. The method according to any one of claims 1 to 4, wherein step (b) for each of the defined areas (138 ₀ to 138 ₆ ) comprises the following steps:

(b.l) determining a plurality of reference values (RW) for different sampling phases at the same sampling frequency, the reference value

(RW) is defined by the absolute difference of at least two successive intensity values (X _n , X _n -i) in the defined ranges (138o to 138 ₆ ); and

(b.2) selecting a maximum reference value or a minimum reference value from the plurality of specific reference values (RW), wherein a maximum reference value defines a contrast with a maximum value, and wherein a minimum reference value defines a contrast with a minimum value.

6. The method according to claim 5, in which the reference value is determined according to the following calculation rule:

with RW = reference value n = number of samples in the considered Area, and X = intensity value of a sampled pixel.

Method according to Claim 6, in which a difference value only contributes to the reference value (RW) if the difference value exceeds a predetermined threshold.

Method according to one of Claims 1 to 4, in which step (b) comprises the following steps:

(bl) performing a first measurement in each of the areas under consideration (138 ₀ to 138 g) at a specified sampling phase and a specified

Sampling frequency to obtain a first reference value;

(b.2) taking a second measurement in each of the regions under consideration (138o to 138 ₆ ) at the specified sampling phase and the specified sampling frequency in order to obtain a second reference value;

(b.3) for each of the regions under consideration, generating a difference between the first and the second reference value;

(b.4) repeating steps (b.l) through (b.3) at different phase settings to obtain a plurality of difference values; and

(b.5) for each of the areas under consideration (138 ₀ to 138 ₆ ), selecting the maximum difference value indicating a minimum contrast or the minimum difference value indicating a maximum contrast from the plurality of difference values obtained.

9. The method according to any one of claims 1 to 8, wherein step (a) comprises defining a plurality of areas (138 ₀ to 138β), the number of areas being determined depending on an accuracy of the resulting sampling frequency (Mn) , and wherein the positions of the areas (138 ₀ to 138 ₆ ) in the row direction have a predetermined distance, which is determined depending on an expected frequency.

10. The method according to any one of claims 1 to 9, wherein the areas defined in step (a) are arranged in the same and / or different lines of the image (136).

11. The method according to any one of claims 1 to 10, wherein the areas (138 ₀ to 138 ₆ ) defined in step (a) are arranged in image areas with high contrast.

12. Device for generating digital data from analog image data in order to display an image (136) based on the analog image data on a digital screen (134)

an A / D converter (100) which includes a data input (102) for receiving the analog image data, a data output (106) for outputting the digital image data and a clock input (104);

a clock generator (112) comprising a clock output (116) for outputting a clock signal and a control input (114) for receiving a clock frequency control signal;

a phase shifter (118) which has a clock input (120) for receiving the clock signal from the clock generator (112), a clock output (124) for outputting comprises a phase shifted clock signal to the clock input (104) of the A / D converter (100) and a control terminal (122) for receiving a control signal defining a phase shift; and

a controller (126) having an input (128) for receiving the digital data from the A / D converter (100), a first control output (130) for outputting the clock frequency control signal to the clock generator (112) and a second control input (132) for outputting the signal determining the phase shift to the phase shifter (118), the control device (126) being operative to calculate based on the digital data provided at the input (128)

to define at least two areas (138 ₀ to 138 ₆ ) successive in the line direction in the image (134) to be displayed,

to determine in each of the areas a sampling phase for which a contrast in the defined area is maximum or minimum,

determine a local course of the sampling phase in the row direction based on the determined sampling phases,

determine the sampling frequency (Mn) based on a basic value (M) and a modification value (ΔM) which is derived from the local course of the sampling phase, and

generate the clock frequency control signal corresponding to the determined sampling frequency (Mn).

13. The apparatus of claim 12, wherein the controller for determining the sampling phase in each of the areas comprises a plurality of samples of each area (138o to 138 ₆ ) in order to obtain a plurality of reference values (RW) for each of the ranges, which is defined by the absolute difference of at least two successive intensity values,

wherein during the plurality of samples the controller changes the signal indicating the phase shift with every sample and keeps the clock frequency control signal constant, and

in which the controller selects a maximum reference value or a minimum reference value from the plurality of reference values (RW) obtained for each area (138 ₀ to 138β).

14. The apparatus of claim 12, wherein the controller is operative to determine the sampling phase

perform a first measurement in each of the considered ranges (138 ₀ to 138 ₆ ) at a defined sampling phase and a defined sampling frequency in order to obtain a first reference value,

a second measurement in each of the examined areas (138 ₀ to 138 ₆ ) at a defined one

Sampling phase and a fixed sampling frequency to obtain a second reference value,

- to repeat the first measurement and the second measurement at different phase settings, and

for each of the areas under consideration (138 ₀ to 138g) select the maximum difference value, which indicates a minimum contrast, or the minimum difference value, which indicates a maximum contrast, from the plurality of obtained difference values.