FR2665603A1 - Method of correcting defects in television signals during their transmission, and implementation device - Google Patents

Abstract

The method of the invention consists in making the transmitter transmit a reference line which is not visible on the screen. On reception, this reference line (7) is sampled, it is stored (8), an error signal (C) is produced on the basis of the difference between this line and its theoretical value, and a correction coefficient is deduced therefrom for each of the samples of the following lines.

Description

METHOD FOR CORRECTING SIGNAL DEFECTS
TELEVISION ON THEIR TRANSMISSION
AND DEVICE FOR IMPLEMENTING
The present invention relates to a method for correcting faults in television signals during their transmission, as well as to a device for implementing this method.

 The defects created during the transmission of television signals, such as the phenomenon of "line-tilt" (inclination of normally horizontal lines) or that of "linear distortion" (distortion affecting linearity) do not generally cause discomfort visible at reception when it is a conventional television system. On the other hand, such faults can become very troublesome in the case of pay-TV systems, therefore with scrambled transmitted signal. In this case, and in particular when using scrambling of the "line cut and rotate" type (line permutation around a cut point determined randomly) or "discrete 2", these defects can greatly affect the quality descrambled images.

 The "line tilt" phenomenon of a video signal is characterized by an inclination of a few% of this signal. For example, a video line with constant luminance, normally represented by a horizontal line, is then a slightly rising or falling line. This phenomenon can have its origin either in the transmission equipment, or in the receiver's demodulator.

 The distortion affecting the linearity, which will hereinafter be called non-linearity distortion, is generally in the form of overshoots extending over the entire length of the line. This phenomenon has the same origins as "line tilt".

 The subject of the present invention is a method for correcting faults caused in particular by the aforementioned phenomena, but also by other phenomena affecting the transmission (parasitic reflections, etc.), a method which is simple to implement, easily adaptable to existing equipment, and which cancels or makes them negligible, in particular in scrambled television systems.

 The present invention also relates to a device for implementing the method of the invention, which is simple, inexpensive, and easy to implant.

 The method according to the invention consists in causing the transmitter to transmit, during at least part of the frames of the video signal, at least one reference line in the part not visible on the screen of the frames, then on reception at sample and store this reference line, produce an error signal from the difference between this stored line and its theoretical value, average the error signal over several frames, deduce a correction coefficient for each of them samples of the line, and to apply this correction coefficient to at least part of the lines of the following frames.

According to an advantageous aspect of the invention, the correction coefficient is weighted as a function of the luminance of the corresponding line. Advantageously, the average luminance level of this line is determined.

 The correction device according to the invention comprises, in each receiver to be corrected, an analog / digital converter followed by a storage device, a logic correction circuit and a digital / analog converter. Advantageously, this device comprises a device for detecting the average level of luminance.

The present invention will be better understood on reading the detailed description of an embodiment taken as a nonlimiting example and illustrated by the appended drawing in which:
FIGS. 1 to 3 are respectively a timing diagram of an unscrambled video signal assigned to "line-tilt", a timing diagram of a scrambled video signal assigned to "line-tilt", and a timing diagram of an affected video signal non-linearity distortion,
FIGS. 4 and 5 are respectively a timing diagram of a reference line transmitted in accordance with the invention, and a timing diagram of this same line as received with distortion,
FIG. 6 is a timing diagram of the error signal stored in accordance with the invention, and relating to the signals of FIGS. 4 and 5,
FIGS. 7 and 8 are respectively a timing diagram of any video line received and affected by distortion, and a timing diagram of this line after correction in accordance with the invention,
FIG. 9 is a simplified diagram of an error signal production circuit according to the invention,
FIGS. 10 to 13 relate to the correction of an echo in accordance with the invention, and are respectively timing diagrams of a video line affected by an echo, of the error signal produced according to the invention, of this weighted error signal, and of the corrected signal according to the invention, and
- Figure 14 is a block diagram of a correction device according to the invention.

 The method of the invention explained below applies as well to the correction of television signal errors of a conventional system as to that of a scrambled television system, and can even be used to correct errors caused by phenomena other than those mentioned above, for example phenomena such as image echoes (often due to parasitic reflections near the receiving antenna).

 There is shown in Figure t a simplified video signal t as transmitted by a television transmitter. The visible part of this signal is a simple horizontal line. The line-tilt phenomenon causes, upon reception, a slight tilting of this line by more or less a few degrees or fractions of a degree and gives the inclined line 2, which in this case is a descending line from left to right.

 If the video signal t is scrambled on transmission, it must, in the absence of any distortion, give a line which is also horizontal. In the presence of the line-tilt phenomenon we obtain the broken line 3 represented in FIG. 2.

 In the case of FIG. 3, it is also assumed that the transmitter sends a video signal with constant luminance which should, in the absence of any distortion, also give the reception a horizontal line 1 ′. In the presence of non-linearity distortion, we obtain, on reception, curve 4 in the form of overshoots ("overshoot").

 According to the invention, the television transmitter transmits a reference line placed in a non-visible area of the image, for example at the start of the frame or in the VBI. In the example chosen and shown in Figure 4, we send on line number 16 a signal with constant brightness, which would correspond to a white line if it were visible on the screen, and whose brightness is equal to 75% of the white level.

 FIG. 5 shows this same reference line as received in a television receiver with a distortion mainly due to a non-linearity distortion.

 Knowing that the reference line is necessarily a straight horizontal line, it is easy to obtain by difference the error signal affecting the visible part of the received line on the screen (between points A and B). The form of this error signal is represented in FIG. 6 (in fact, this signal is digitized, and the digitization being done at high frequency, this signal has been represented in the drawing as if it were analog).

In the receiver, the visible part is sampled (between
A and B) of this line and it is stored in a register whose capacity is at least sufficient to store a complete line. Sampling is controlled at the rate of a clock signal capable of sampling pixels. Advantageously, the frequency of this clock signal is of the order of magnitude of 10 MHz or more, in order to obtain good resolution.

 The error information thus memorized is averaged under the control of a microprocessor for example, and this, for several frames. This operation can be done, for example, when the receiver is powered up or at determined time intervals after tests. We know that the non-linearity distortion depends, to a certain extent, on the luminance level of the image. Experiments have shown that around 60% of the distortion depends on the level of luminance. Consequently, it is very advantageous to apply a correction factor depending on the luminance level. This correction factor K is determined by measuring the average luminance level on a row of rank n and it is applied to the following row n + l , the factor K being between O and l. Detection of the average level of luminance of a video line is done in a manner obvious to those skilled in the art by rectification and integration.

To correct the errors affecting the received signal, the error signal E is entered for N frames (N = a few tens for example) and the average value pixel by pixel is determined. From the following frame (N + 1), the correction thus determined is applied pixel by pixel to each visible line of each frame. The result is then converted to analog form
FIG. 7 shows a video signal line V
o affected by various distortions. As shown diagrammatically in FIG. 9, the corresponding error signal E is multiplied by -K to give the weighted correction signal C = -KE We add C to V0 and we obtain the corrected signal V as shown in FIG. 8.

The method of the invention also makes it possible to correct the defects due to echoes. FIG. 10 shows a simplified example of a video line V ′ (white line, the part of which
o visible on the screen is delimited by the references C and D) assigned an echo whose duration is such that, in the example of the drawing, the beginning of this line (synchronization pulse and level) is echoed at a distance L from its normal location, which creates the parasite P. The method of the invention makes it possible to produce an error signal E '(between C and D) as shown in FIG. 11, and a weighted error signal C' = - K'E '.
After application of this correction, the signal V ′ is obtained as emitted and represented in FIG. 13.

 FIG. 14 shows the block diagram of an exemplary circuit 5 for implementing the method of the invention.

The input terminal 6 of this circuit receives the analog signals V to be corrected. Terminal 5 is connected to a
o analog / digital converter 7, the output of which is connected to a line or frame storage circuit 8, which in the present example is a FIFO memory. The output of memory 8 is connected to a logic circuit 9 executing, on each line, the calculation of C = -KE under the control of a microprocessor 10. Terminal 6 is also connected to a circuit 11 for detecting the medium level of luminance of each line or of each frame, and to a time base circuit 12 supplying line and pixel synchronization signals. The respective outputs of circuits 11 and 12 are connected to circuit 9.

 The output of circuit 9 is connected to a digital / analog converter 13, the output of which is connected to an input of an adder 14, the other input of which is connected to terminal 6.

At the output of the adder 14, the corrected video signals V = V + C are collected. Of course, if the video signals received are
o in digital form, the converter 7 is useless.

Claims (9)

 1. Method for correcting faults in television signals during their transmission, characterized in that the transmitter transmits, during at least part of the frames of the video signal, at least one reference line in the part not visible on the screen of the frames, then on reception, this reference line is sampled, an error signal is produced from the difference between this line and its theoretical value, which is deduced therefrom a correction coefficient for each of the samples of this line, and that this correction coefficient is applied to at least part of the lines of the following frames.  2. Method according to claim 1, characterized in that the error signal is averaged over several frames.  3. Method according to one of the preceding claims, characterized in that the correction coefficient is weighted as a function of the luminance of the corresponding line.  4. Method according to claim 3, characterized in that one determines the average level of luminance of the line.  5. Device for correcting faults in television signals during their transmission, characterized in that it comprises in each receiver applying the correction an analog / digital converter (7) followed by a storage device (8), d '' a logic correction circuit (9, t0) and a digital / analog converter (13).  6. Device according to claim 5, characterized in that the storage circuit is a FIFO.  7. Device according to claim 5 or 6, characterized in that it further comprises a mean luminance level detection circuit (11).  8. Device according to one of claims 5 to 7, characterized in that the logic correction circuit is controlled by a microprocessor (10).  9. Television receiver characterized in that it comprises a correction device according to one of claims 5 to 8.