Method and device for phase correction of a vertically distorted digital image

Abstract

The invention relates to the phase correction of a vertically distorted digital picture, which is divided into two half pictures in accordance with the interlaced scanning method. To take into account a vertical decimation factor (vscale), which changes in the vertical direction of the digital picture, it is proposed that a phase correction (vphscor) be generated for the second half picture as a function of the particular vertical position in the second half picture. This phase correction must be considered in generating the second half picture. It is derived directly from information (vinc) about the change of the vertical decimation factor.

Description

[0001] The present invention relates to a method according to the generic part of claim 1 and to a device according to the generic part of claim 5 for correcting the phase of a vertically distorted digital picture, especially of a digital television picture.

[0002] In presently customary systems of television technology, pictures are displayed by the so-called interlaced scanning method. As is shown in FIG. 4, the interlaced scanning method is characterized in that the picture being displayed (full picture) is decomposed into lines, such that the lines of this full picture are divided into two temporary sequential half pictures. The lines with an originally odd line number are displayed in the first half picture, and the lines with an originally even line number are displayed in the second half picture. The standard for displaying the picture is specified in the guideline ITU-R 601.

[0003] As is clear from FIG. 4, the number of lines of each of the two half pictures is half as great as the number of lines of the full picture. A so-called vertical decimation factor (VDEC) can be defined for the interlaced scanning method. This factor corresponds to the ratio of the number of lines of the full picture to the number of lines of the half pictures. In the example of FIG. 4, the vertical decimation factor thus is 2.

[0004] FIG. 4 furthermore clearly shows that a phase offset PHS exists between the lines of the first half picture and those of the second half picture. This phase offset amounts to one line. In general, the phase offset is calculated as follows: 1$\begin{array}{cc}\mathrm{PHS}=\frac{\mathrm{VDEC}}{2}& \left(1\right)\end{array}$

[0005] This phase offset must be taken into account at the beginning, when generating the second picture.

[0006] The preceding description assumes a vertical decimation factor VDEC which is constant over the entire full picture or the entire half picture. However, for certain applications and for displaying certain effects, it may be desirable to create digital pictures by the interlaced scanning method, with a vertical distortion which changes as a function of vertical position. For example, expansion or compression of a picture or display of a vertical panorama effect requires a change of the vertical decimation factor or expansion factor over the picture. The vertical decimation factor VDEC thus becomes a function of the line of the half picture, i.e. VDEC=VDEC(l), where (l) designates the particular line of the half picture. For displaying other effects, it is also conceivable to make the vertical decimation factor also dependent on other parameters, for example the pixel position, etc. However, to calculate a phase correction factor for the second half picture, in order to take into account the phase offset between the second half picture and the first half picture, only the dependence of the vertical decimation factor VDEC on the line is relevant. The following therefore holds for the phase offset: 2$\begin{array}{cc}\mathrm{PHS}\left(1\right)=\frac{\mathrm{VDEC}\left(1\right)}{2}& \left(2\right)\end{array}$

[0007] It is necessary to perform a phase correction for the second half picture, since otherwise an undesirable picture distortion, which is perceived as interference, occurs in the vertical direction during the course of the picture.

[0008] It is therefore the object of the invention to propose a method and a corresponding device to generate half pictures distorted in the vertical direction with the least possible complication. In particular, when generating digital pictures with the interlaced scanning method, and in the case of a vertical distortion which changes in dependence on the vertical position, the phase correction signal for the second half picture should be generated as easily as possible.

[0009] According to the invention, this object is achieved by a method with the characteristics of claim 1 and by a device with the characteristics of claim 5. The subclaims each define preferred and advantageous embodiments of the present invention.

[0010] According to the invention, the phase correction and the phase correction signal for the second half picture are derived from a signal which contains information about the change of the so-called vertical decimation factor in the vertical direction of the digital picture and of the second digital half picture. If the vertical decimation factor changes as a function of the line, the phase correction and the phase correction signal are correspondingly determined anew for each individual line of the second half picture.

[0011] In particular, the phase correction vphscor can be determined in accordance with the following relation, where vinc(l) designates the vertical increment of the vertical decimation factor, i.e. the change of the vertical decimation factor in the vertical direction: 3$\begin{array}{cc}\mathrm{vphscor}\left(1\right)=\frac{1}{2}·\mathrm{vinc}\left(1\right)& \left(3\right)\end{array}$

[0012] The entire circuit needed for the phase correction can be constructed of only two adders with feedback, one limiter, a multiplexer, and an additional adder.

[0013] The present invention can be used generally in the field of digital picture processing, especially in the field of digital television technology. A unit for picture processing with vertical picture distortion using the interlaced scanning method, based on the present invention, can be situated both before and after an appropriate picture memory, i.e. the principle on which the present invention is based can be applied generally to a vertical distortion both before and after a picture memory.

[0014] The present invention is explained in more detail below, in terms of preferred embodiments, having reference to the attached drawing.

[0015] FIG. 1 shows a simplified block circuit diagram of a unit for vertical picture distortion, constructed in accordance with the present invention,

[0016] FIG. 2 shows a detailed picture of the circuit section which contributes to the generation of the phase correction signal in FIG. 1, in accordance with a preferred embodiment of the present invention,

[0017] FIGS. 3A and 3B show in general the arrangement of the units shown in FIG. 1, before and after a picture memory, and

[0018] FIG. 4 shows a chart for elucidating the decomposition of a full picture into the lines of two half pictures with the so-called interlaced scanning method.

[0019] The unit, shown in FIG. 1, for vertical picture distortion comprises a circuit section 1, below called panorama generator, a circuit section 2, below called phase correction section, and a circuit section 3, below called interpolation phase generator. The circuit sections 1-3, in combination, are used to generate a phase correction signal vphase(l) for the second half picture of a digital full picture. Its input data are conducted to an interpolation stage 4. In dependence on the phase correction signal vphase(l), which can also be designated as the vertical interpolation phase, the interpolation stage 4 calculates the decimated picture lines belonging to the digital picture data of a digital full picture, in accordance with the particular instantaneous value of the so-called vertical decimation factor VDEC(l), and outputs the picture data corresponding to these decimated picture lines of the two half pictures of the digital full picture in the form of output data.

[0020] The vertical interpolation phase vphase(l), created by the interpolation phase generator 3, designates the weighting between two original lines of the digital full picture, and serves as the basis for the interpolation stage 4, to be able to calculate the decimated picture lines of the two digital half pictures and the corresponding digital picture data. The interpolation phase generator 3 calculates the vertical interpolation phase vphase(l) as a function of a vertical scale factor vscale(l), which is generated by the panorama generator 1, and of a vertical phase correction vphscor(l), which is generated by the phase correction circuit 2. The vertical scale factor vscale(l) is a measure of the vertical decimation factor VDEC(l). A vertical increment vinc(l) is conducted as input signal both to the panorama generator 1 and to the phase correction circuit 2. This vertical increment describes the change of the vertical decimation factor, i.e. vinc(l)=0 if ΔVDEC(l)=0. The values and signals vinc(l), vscale(l), vphscor(l), and vphase(l) are each a function of the line 1 of the digital full picture and of the particular second digital half picture under consideration.

[0021] The structure of the individual circuit sections 1-3, shown in FIG. 1, will be explained in more detail below with reference to FIG. 2.

[0022] As can be seen from FIG. 2, the interpolation phase generator 3 in principle consists of a digital oscillator, which is formed by an adder 9, fed back through a register 11. The reference symbols k, m respectively designate the bit width of the data words transmitted within the interpolation phase generator 3. In addition, the interpolation phase generator 3 has an adder 10, situated in its feedback path, by means of which the output value of the adder 9 is added to the vertical phase correction value vphscor(l) of the phase correction circuit 2.

[0023] Without considering this vertical phase correction vphscor(l), the vertical decimation factor VDEC(l) is defined as follows, as a function of the vertical scale factor vscale(l) of the interpolation phase generator 3: 4$\begin{array}{cc}\mathrm{VDEC}\left(1\right)=\frac{2^k+\mathrm{vscale}\left(1\right)}{\mathrm{vscale}\left(1\right)}& \left(4\right)\end{array}$

[0024] Similarly as with the phase correction for the second half picture with a constant vertical decimation factor, so also, with a variable vertical decimation factor, the phase must be corrected for each new value of the vertical interpolation phase vphase(l). The phase correction is here calculated as follows: 5$\begin{array}{cc}\mathrm{vphscor}\left(1\right)=\frac{1}{2}·\frac{d}{dl}\mathrm{VDEC}\left(1\right)& \left(5\right)\end{array}$

[0025] One thus obtains: 6$\begin{array}{cc}\mathrm{vphscor}\left(1\right)=\frac{1}{2}·\frac{d}{dl}\mathrm{vscale}\left(1\right)& \left(6\right)\end{array}$

[0026] We here assume the convention VDEC(l)=vscale(l). A mathematical simplification and approximation then yields: 7$\begin{array}{cc}\mathrm{vphscor}\left(1\right)=\frac{1}{2}·\Delta \mathrm{vscale}\left(1\right)& \left(7\right)\end{array}$

[0027] If the vertical increment, as already described, is now described by the change of the vertical scale factor and of the vertical decimation factor, that is if:

(8) vinc(l)=Δvscale(l)

[0028] the phase correction can be derived directly from the vertical increment, as follows: 8$\begin{array}{cc}\mathrm{vphscor}\left(1\right)=\frac{1}{2}·\mathrm{vinc}\left(1\right)& \left(9\right)\end{array}$

[0029] The phase correction signal vphscor(l) for the second half picture can thus be derived directly from the increment vinc(l), which serves as the basis for the vertical scale factor vscale(l). The signal of the vertical scale factor vscale(l) on its part serves as the control signal for generating the phase information vphase(l) for the interpolation by the interpolation stage 4.

[0030] Due to the above relation (8), the following relation holds:

(10) vscale(l)=vscale(l−1)+vinc(l)

[0031] or

(11) vinc(l)=vscale(l)−vscale(l−1)

[0032] The structure of an accumulator, shown in FIG. 2, thus results for the panorama generator 1, which generates the vertical scale factor vscale(l). In principle, this accumulator is formed by an adder 4, fed back through a register 6.

[0033] In addition, a limiter is inserted into the feedback path, to prevent overflow of the register 6. The reference symbols n, u and v again designate the word width of the respectively transmitted data words.

[0034] As shown in FIG. 2, the vertical phase correction vphscor(l) can be realized by the combination of a multiplexer 7 with an evaluator 8 and by adding the adder 10, already explained above, to the phase generator 3. Information about the specific instantaneous half picture position, i.e. about the specific and continuously instantaneously processed half picture, is continuously conducted to the multiplexer 7. If, at this instant, the first half picture is involved, the adder 10, which has been added to the interpolation phase generator 3, is ineffective, since the multiplexer 7 outputs the value “0” On the other hand, for the second half picture, the value vinc(l)/2 is conducted to the adder 10 through the multiplexer 7 and the evaluator 8. One half of the vertical increment vinc(l) is thus added to each phase value calculated by the adder 9 of the interpolation phase generator 3, so as to obtain the final phase vphase(l). The evaluator 8 can be realized simply by a bit shift operation, i.e. by shifting the bits of the data word vinc(l) by one place to the right or toward a less significant bit.

[0035] In this way, the phase of the second half picture can be corrected by a circuit with an extremely simple structure.

[0036] As can be seen from the diagrams of FIGS. 3A and 3B, the inventive unit for vertical picture distortion can be situated both before and after a picture memory. FIG. 3A shows a general diagram of an inventive unit 100 for vertical picture distortion by the interlaced scanning method in the acquisition path of an arrangement for digital video signal processing, in which the inventive unit 100 for vertical picture distortion is situated behind a data acquisition section 101 and before a picture memory 102. FIG. 3 shows a general diagram of the arrangement of an inventive unit 100 for vertical picture distortion by the interlaced scanning method in the display path of an arrangement for digital video signal processing, where the inventive unit 100 for vertical picture distortion is situated after a full picture memory 103 and before a unit 104, which is used to display the picture by the interlaced scanning method.

Claims

1. A method for correcting the phase of a vertically distorted digital picture, by which the lines of the digital picture are assigned to a first half picture and to a second half picture in accordance with an imaging factor (vscale) which changes in the vertical direction of the digital picture, by which the lines of the first half picture and the lines of the second half picture are displayed successively, and by which a phase correction (vphscor) is generated for the second half picture, as a function of the respective vertical position in the second half picture, this phase correction (vpscor) [sic] being taken into account in the generation of the second half picture, characterized in that the phase correction (vphscor) for the second half picture is derived from information (vinc) about the change of the imaging factor (vscale) in the vertical direction of the digital picture.

2. The method of claim 1, characterized in that the imaging factor (vscale) changes as a function of the line in the vertical direction of the digital picture, and that the phase correction (vphscor) is determined anew for each line of the second half picture.

3. The method of claim 1 or 2, characterized in that the phase correction (vphscor) for the half picture is determined as follows from the value (vinc), which describes the vertical change of the imaging factor (vscale) in the vertical direction of the digital picture: vphscor={fraction (1/2 )}·vinc.

4. The method of one of the claims 1-3, characterized in that the method for correcting the phase of a vertically distorted digital video picture, is applied especially to a digital video picture.

5. A device for correcting the phase of a vertically distorted digital picture, with a line determination device (4), to which are conducted the picture data of a digital picture, to assign the lines of the digital picture to a first half picture and to a second half picture in accordance with an imaging factor (vscale), which changes in the vertical direction of the digital picture, such that the lines of the first half picture and the lines of the second half picture are displayed sequentially, and with phase correction means (1-3) to generate a phase correction signal (vphase) for the line determination device (4), such that the phase correction signal (vphscor) depends on the specific vertical position in the second half picture and is taken into account by the line determination device (4) when generating the second half picture, characterized in that the phase correction means (1-3) are designed in such a way that they derive the phase correction signal (vphscor) for the second half picture from an increment signal (vinc), which describes the change of the imaging factor (vscale) in the vertical direction of the digital picture.

6. The device of claim 5, characterized in that the imaging factor (vscale) changes as a function of the line in the vertical direction of the digital picture, and that the phase correction means (1-3) are designed to generate the phase correction signal (vphscor) anew for each line of the second half picture.

7. The device of claim 5 or 6, characterized in that the phase correction means comprise a first circuit section (1) to generate, in dependence on the increment signal (vinc), a control signal (vscale) which determines the imaging factor, and that the phase correction means comprise a second circuit section (2) to generate the phase correction signal (vphscor) in dependence on the increment signal (vinc), and that the phase correction means comprise a third circuit section (3) to generate a phase signal (vphase), which is conducted to the line determination device (4) and to which corresponds to a phase offset between the first half picture and the second half picture, and which is generated in dependence on the control signal (vscale) of the first circuit section (1) and the phase correction signal (vphscor) of the second circuit section (2).

8. The device of claim 7, characterized in that the first circuit section (1) comprises an adder (4), fed back through a register (6), and that the increment signal (vinc) is conducted as an input signal to said adder (4).

9. The device of claim 8, characterized in that a limiter (5) is situated in the feedback path of the fed back adder of the first circuit section (1).

10. The device of one of the claims 7-9, characterized in that the second circuit section comprises a multiplexer device (7) and an evaluation device (8) with the evaluation factor {fraction (1/2 )}, such that the multiplexer device receives as its input signals the increment signal (vinc) and the fixed value “0” and is driven by a control signal which describes the half picture position, and such that the output signal of the multiplexer device (7) is conducted, via the evaluation device (8), to the third circuit section (3).

11. The device of claim 10, characterized in that the third circuit section comprises an adder (9), fed back through a register (11), and that the input signal of this adder (9) is the control signal (vscale) which determines the imaging factor and which is generated by the first circuit section (1), and that another adder (10) is situated in the feedback path of the fed back adder, and that the output signal of the evaluation device (8) of the second circuit section (2) is conducted, as the phase correction signal (vphscor), to said further adder (10).