DE3017929C2 - Signal processing circuit - Google Patents

Description

comb-filtered high-frequency component (which occupies a frequency band above the filter cut-off frequency), from which the color signal components are removed, and a non-comb-filtered (i.e. flat) low-frequency portion in which all luminance signal components are preserved.

It is sometimes desirable to have the vertical detail information to attract or emphasize the playback picture, by adding back to the luminance signal a larger amount of the vertical detail signal than is required to restore the luminance signal in its original form (i.e. with a flat Amplitude curve). The additional vertical detail signal is then used to emphasize vertical detail information in the sense of increasing the resolution for image details. For low level luminance signals however, such emphasis tends to produce objectionable visible effects, if there are smoke disturbances and undesirable together with the vertical detail information of the luminance signal be raised.

In this case, any setting changes that may be present in the video signal are also made for every second Line accentuates this in an undesirable way This is known as AL-SUV (alternate line set-up variation phenomenon Phenomenon is a type of interference in signals with a low level, which is caused by line-by-line changes in the black level expresses itself in the video signal and, for example, through misalignments of the signal processing systems can be caused by the radio transmitter. ALSUV interference is particularly noticeable with video signals low level of about 5% of the maximum expected video signal amplitude and leads to disruptive visible effects on the playback picture, which undesirably enlarge when a vertical detail increase is provided.

A technique to minimize harmful effects of interference and other undesirable components of a video signal uses a method commonly referred to as "signal coring," where small signal amplitudes (including Noise) can be removed, as described, for example, in US Pat. No. 3,715,477.

An advantageous system which effects such elimination of the vertical detail signal in a manner in which the vertical detail information is not impaired (e.g. smeared), in particular with respect to low level detail signal information which is restored in the luminance signal is described in DE-OS 30 17 930. This system advantageously also ensures an increase in the vertical detail information practically without simultaneous Increase in interfering signal components such as noise and the aforementioned ALSUV phenomena.

A system in which vertical detail signals of large amplitude are reduced (amplitude reduction or attenuation) in order to avoid image tube overexposure that would otherwise destroy detailed information or covered, is described in DE-PS 30 17 931.

The invention is based on the object of a signal processing circuit which has a non-linear amplitude transfer function to selectively restore, increase and decrease of video signals of small, medium or large amplitude.

This object is achieved by the characterizing features of claim 1, while further developments and special embodiments of the invention are characterized in the subclaims

The invention has the advantage that the vertical detail signal processing circuit is not affected by changes in the DC voltage level of the luminance component. Because of the manner in which a comb filter derives a comb-filtered luminance signal by using known subtractive signal combination, the comb-filtered color signal has a DC component of zero. This DC voltage component of the comb-filtered color signal therefore does not interfere with the DC bias voltage component which arises at the comb-filtered color output. The processing of the vertical detail signal, which is derived at the comb filter color output, can therefore be centered in a predictable manner around a DC bias component if the comb filter color output is galvanically coupled to the vertical detail signal processing circuit well-defined areas of recovery, elevation, or attenuation;
A circuit according to the invention has an amplifier with an input terminal and an output terminal, between which an amplifier device is arranged. A first feedback branch with a first impedance runs between the input and output of the amplifier device. A second feedback branch with a second impedance and a threshold switch arrangement is also coupled between the output and input of the amplifier device. The switch arrangement has one input connected to the output of the amplifier device and one output connected to the output terminal of the circuit. The switch arrangement in one line state for signal amplitudes with first values within a first range and another state for signal amplitudes with second values that are greater than the first values within a second range.

According to a feature of the circuits according to the invention, a third feedback branch is between The output and input of the amplifier device are provided. The third branch contains a third impedance and an additional threshold switch arrangement with one connected to the output in the second feedback branch provided switch arrangement connected input. The additional switch arrangement takes one line state at signal amplitudes of the first and second values and another Line status for signal amplitudes with third values that are greater than the first and second values, within a third area.

According to a further feature of the circuits according to the invention, the second feedback branch also contains an additional impedance. The switch arrangement the second branch is coupled to the additional impedance to allow power to pass through it To control switching.

According to a further feature of the invention, the circuits according to the invention described here in a color television receiver or similar system for the transmission of vertical picture detail information si-

b5 signal is used with a non-linear transfer function with respect to the aforementioned ranges of the vertical detail signal amplitudes.
In the drawings shows

F i g. 1 is a block diagram of part of a color television receiver with a non-linear signal processing circuit according to the invention,

F i g. 2 shows an embodiment of a non-linear signal processing circuit according to the invention,

3 shows a modified embodiment of the in F i g. 2 circuit shown,

Fig. 4 to 6 amplitude characteristics for understanding the mode of operation of the circuits according to the invention; and

F i g. 7 shows another embodiment of a circuit according to the invention.

According to FIG. 1 provides a source 10 of a composite color video signal, which contains luminance and color components. Video signals to an input of a Kammfiiter i5 well-known training, such as a Charge coupled device (CCD) comb filters as described in US Pat. No. 4,096,516. The luminance and color components are frequency wise within a video signal frequency spectrum nested. The luminance component has a relatively large bandwidth (it ranges from Direct current of frequency 0 to about 4 MHz). The luminance component divides the upper frequency range with the color component, which comprises a color carrier signal of 3.58 MHz, the amplitude with the color information and is phase modulated. The amplitude frequency response of the color filter 15 shows with reference to FIG Luminance comb filtering a peak amplitude transmission at whole multiples of the horizontal line scanning frequency (about 15734 Hz) starting at DC or frequency 0 and an amplitude transfer of 0 for odd multiples of half the line scanning frequency including the color subcarrier frequency of 3.58 MHz. The amplitude frequency response of the comb filter 15 shows with regard to the color signal comb filtering an amplitude peak transmission at odd multiples of half the line frequency including the frequency of 3.58 MHz, and an amplitude transfer of 0 at whole multiples of the line frequency.

A comb-filtered luminance signal (Y) from the luminance output of the comb filter 15 is coupled via a low-pass filter 22 to an input of a signal combination circuit 30. The filter 22 is arranged so that it passes all luminance signals below a cut-off frequency of about 4 MHz, and it is used to remove noise - And clock frequency components of the switching signals that occur during switch operation of the comb filter 15 if it is a CCD comb filter

A comb-filtered color signal (C) from the color output of the comb filter 15 is fed to a color signal processing unit 64 for generating RY, B-Y and G-Y color difference signals and an input of a vertical detail low-pass filter 35. The unit 64 contains a suitable filter which only those signal frequencies from the comb filter 15 passes which occupy the band of the color signal differences. The filter 35 has a cut-off frequency of approximately 1.8 MHz and selectively lets through those signal frequencies contained in the comb-filtered color output signal of the comb filter 15 which are below this cut-off frequency. Signal frequencies in this area represent vertical detail luminance information, which is missing in the comb-filtered luminance signal and must be restored in the luminance signal if one wants to avoid a loss of vertical resolution in the luminance content of the displayed image. Such a vertical detail restoration as well as a vertical detail raising and lowering is effected in the manner described below.

Vertical detail signals from the output of the filter 35 are subjected to a non-fine signal processing scheme! device 50 supplied, which will be described in detail. The amplitude transfer characteristic of the Signal processing circuit 50 is shown in FIG. 5 shown. The following explanations regarding signals positive polarity (+) also apply to signals with negative polarity (-).

The signal processing units within the circuit 50 effect a signal amplitude transmission (amplification) characteristic, for three areas I, ϊ! and ill (see Fig. 5) with respect to the three predetermined ones Ranges of vertical detail signal amplitudes. In a recovery area I for signals low Levels (for example signal amplitudes of about 5% of the maximum expected amplitude) is used for a predetermined gain behavior is provided, so that low level detail signals along with noise and other undesirable components in Area I can be processed without raising. The peak amplitude of vertical detail signals of medium amplitude (e.g. signal amplitudes between 5% and 40% of the maximum expected amplitude) are increased in the increase range II with a gain factor of, for example, approximately 3 is processed, thereby emphasizing the vertical detail information and increases the image resolution for this area. The peak amplitude of vertical detail signals, relative large values (for example between 40% of the maximum expected amplitude and the maximum amplitude) corresponding to images of high contrast, such as writing, is reduced in amplitude, how this shows the amplitude behavior for the area IH in order to avoid excessive contrast and To prevent overexposure to the picture tubes, which would otherwise distort or cover up picture details.

In area 1 (vertical detail recovery), low-level vertical detail information is all in one has been restored to a sufficient extent to allow normal vertical resolution for low levels im Preserve the luminance content of the display image. In this example, and as will be shown, small signal amplitudes processed in region I to a prescribed recovery gain amplified by about 2 The amplification in area I is preferably that amount of signal amplification which is necessary in a given system around small amplitudes of the vertical detail signal in the luminance signal again, so that the luminance signal finally restored one has an essentially flat amplitude curve for detail signals with a small amplitude to point out that the magnitude of the recovery gain depends on various factors including the signal transmission properties between the outputs of the comb filter 15 and a Luminance signal processing circuit 32 coupled circuits which the ultimately restored Process luminance signals, including the relative sizes of those at the outputs of the comb filter 15 appearing signals, for example

The choice of recovery gain, like them

for the amplitude transmission obtaining in area 1 also depends on considerations as to which results are acceptable in a given video signal processing system. For example, if the recovery gain is too great, then small ALSUV signal glitches can also become apparent. If the recovery gain is not sufficient, then considerable comb filtering effects (i.e. signal peaks and 0 values at different frequencies) appear in the vertical detail frequency range below 2 MHz, which lead to the loss of small vertical detail information. Thus, the slope of the amplitude transfer characteristic in area I corresponds to the degree of signal amplification that is necessary to give a desired transfer behavior (i.e. a flat luminance curve) without introducing unacceptable side effects. The signal transmission behavior in area I preferably shows a fixed relationship to the behavior of the signal path which transmits the comb-filtered luminance signals (Y) from the output of the comb filter 15 to the combination circuit 30.

In area II (elevation of vertical details), a suitable degree of vertical detail elevation is provided, by additionally amplifying signals of medium amplitude in a way which is favorable for the vertical resolution of a playback picture is being viewed. Although the peak amplitudes of signals are medium Sizes that are increased, in this example with a larger gain than the recovery gain are amplified, small amplitudes of these medium-sized signals become with the recovery gain processed (i.e. without raising). Also signals of small amplitude, which are not raised are processed with the recovery gain. This way there will be an increase practically no undesirable small signal components including noise and ALSUV interference carried out or reduced to an acceptable minimum, and one avoids smearing of vertical detail information low level.

The processed vertical detail signal from circuit 50 is summed in circuit 30 with the comb-filtered luminance signal (Y) coming from filter 22. The output of the combination circuit 30 corresponds to a restored luminance component of the video signal in which the vertical detail information is restored (area I), increased (area H) and decreased (area 111) The restored luminance component is then supplied to a luminance signal processing circuit 32. An amplified luminance signal Y from the circuit 32 and the color difference signals from the unit 64 are combined in a matrix 68 to form output signals R, B and G representing a color image

F i g. 2 shows a circuit implementation of a non-linear signal processing circuit 50. Output signals from the detail filter 35 are fed as an input signal (Si) via a coupling capacitor 72 and an input resistor 73 to an inverting input of an operational amplifier 75 contained in the circuit 50. A non-inverting input of the amplifier 75 is connected to a Reference potential point (e.g. ground) connected.

Between the output and the inverting input of the amplifier 75 there is a first feedback branch with a feedback resistor 76. A second feedback branch contains the parallel connection of suppression diodes 81 and 82 with a resistor 87, as well as a feedback resistor 78 and a coupling capacitor 77. The diodes 81 and 82 are mutually like that arranged that they conduct a signal which appears at the output of the amplifier 75 at opposite polarities. A third feedback branch contains diodes 83 and 84 connected in parallel, a resistor 79 and the capacitor 77. The diodes 83 and 84 are mutually arranged in such a way that they conduct a supplied signal when the polarities are opposite. An output signal S ₀ from the circuit 50 is fed to the second input of the combination circuit 30 (FIG. 1) via a coupling capacitor 140.

If you neglect the capacitors 90, 91 and 92 for the time being, then the signal processing circuit has 50 shows a non-linear total amplitude transfer function as shown in FIG. 5 shows and effects various Measures of signal amplification for signals with amplitudes in the three areas I, II and Hl according to FIG. 5. Of the The value of the capacitor 77 is selected to be sufficiently large so that the direct voltage across the capacitor 77 is practical equal to the DC voltage level at the output of amplifier 75 after an initial settling time is.

As already mentioned and explained in DE-OS 30 17 930 it is important that the detailed information represented by small amplitude signals be restored is used to prevent the vertical detail content of the luminance signal from being smeared. this function is through the resistor 87 in the circuit of FIG. 2 causes

The diodes 81 and 82 do not conduct with small signal amplitudes. Resistor 87 is a linear element which, in circuit 50, allows small-amplitude detail signals with the prescribed recovery gain (see FIG. 4) in region I to be processed, as is the amplitude characteristic in FIG. 5 shows. The recovery gain A / (approximately 2) obtained by the output signal S ₀ after processing in the region I is given by the following expression, in which 73, / 78 ^and Ra are the corresponding values for the resistors 73, 78 and 87 and Rn corresponds to the parallel connection of the resistor 76 with the series connection of the resistors 78 and 87:

73

R _n + R.

The resistor 87 does not affect the signal amplification for signals of medium and large amplitude in the Areas II and III, since the resistor 87 is short-circuited or bridged when the diodes 81 and 82 conduct when medium or large amplitude signals are supplied.

If the amplitude of the signal appearing at the output of amplifier 75 increases by a moderate amount above the conduction threshold value of diodes 81 and 82 (corresponding to signals that are processed in area II), then diodes 81 and 82 conduct. The signal amplification of circuit 50 im Area II (An) is then determined to be greater than the recovery gain, namely about 3, according to the following expression, in which Rp \ is the value of the par-

9 10

connection of the resistors 76 and 78 corresponds, according to FIG. 5 by the fact that signals of small amplitudes

and R ₇ j corresponds to the value of resistor 73: tude in area I are suppressed because the diodes

β 81 and 82 then do not conduct. This behavior corresponds to

A \\ - - £! ■. a gain of 0 in the range I, whereby signals are smaller

^ 73 5 Amplitude does not appear as output signals S _n because

the diodes Si, 82 and S3, 84 then do not conduct and the

The gain for medium amplitude signals becomes the inverting input (-) of the operational amplifier

represents a virtual mass point through the amplitude transfer function in area 75, so that

II illustrated in Figure 5. In this context sen conditions no signal currents over the resistance noticed that the amplitudes of signals in the middle ίο would be 78, 79 or the capacitors 77, 91, 92 to flow with the restoration circuit output.

Strengthening are transmitted while peak amplitude F i g. 7 shows another embodiment of the signal

which are reinforced in the manner described above. Processing circuit 50 which is similar to that of FIG

The width of the region I for both signal polarities is similar to FIG. 1. Corresponding elements are in the

depends on the resistance ratio of the resistors 76 15 F i g. 2 and 7 denoted by the same reference numerals,

and 73 from. In this arrangement, the input

The amplitude transfer function for detail electrode switched basis of a basic emitter circuit

signals of large amplitude corresponding to the amplifier transistor 75 operated in the area of an inversely

III in Fig.5 associated signals is determined animal input, the emitter of the transistor is applied through the feedback branch with the resistor 79 20 ground and its collector is used as the output electrode and the suppression diodes 83 and 84, which switches at and is across a collector load impedance 112 Peak amplitudes of the signals with large amplitude values at an operating voltage source (+16 V). With open te become leading. The diodes 81 and 82 then conduct the loop, the transistor 75 has a gain that even. This circuit is used to lower the peak - mainly through the value of the load impedance 12 bezenamplitudes of detail signals of large amplitudes - 25 is correct and is high enough to compensate for the gain values in the range IH in that these peak amplitudes are almost amplified with the open loop of an operational amplifier (lower than the recovery gain for example of amplifier 75 in FIG. 2). The signal amplification in the area III men.

(Λ111) is in this example by the following equation vertical detail signals (Si) from the filter 35 become the

given, in which Rn is supplied to the value of the parallel circuit 30 base of the transistor 75 via a circuit,

from the resistors 76,78 and 79 and R ₇₃ the value which an input resistor 73, a capacitor

of resistor 73 corresponds to: gate 105 and a resistor 106 in the illustrated

Arrangement contains the last two elements

yi _m - "** ^2nd te are used for frequency compensation of the frequency

Ä73 35 output of transistor 75 when the loop is open. To be there-

output signal (Si) does not need to be coupled to AC voltage

In this context it should be noted that small pelts should be predictable if there is a stable one Has amplitudes of the signals of large amplitude values with DC voltage level, as in the case of this

the recovery gain, assuming the example is where the input signal galva-

medium amplitudes with a larger than the Viennese 40 is coupled to the base of the transistor 75. The-

the production gain and peak amplitudes with this DC voltage level is used together with the Wi-

less than the restoration amplifiers 73, 76 and! 14 the determination of a good

kung. desired operating point for transistor 75.

Thus, the overall amplitude overtop As in the circuit according to FIG. 2 determines the rolling characteristics of the circuit 50 according to FIG. 4 three 45 status 76 in F i g. 7 shows the width of range I (i.e. the gain ranges for three predetermined signal am restoration gain ranges for both amplitude levels, both for positive (+) and signal polarities) in connection with resistor 73 for negative (-) signal polarities. The frequency response of and the conduction threshold of diodes 81 and 82 are processed in any of these three gain ranges 90, 91 and 92 are properly biased, so that in parallel with the appropriate feedback resistor results in a symmetrical transfer function.The desired width can be measured as shown.Output signal (S ₀ ) is via a capacitor 140. In the present example, the frequency bandwidth is AC-coupled to prevent a direct current flow of the processed signals of small amplitudes (area 55 which could disturb the desired symmetry of the Ge-I) proportional to the reciprocal of the capacitor 90. Total transfer function (FIG. 5). The frequency bandwidth of the processed signals mean frequency response for signals which are processed in any of the ler amplitudes (range H) is proportional to the recirculation of the three gain ranges can be prokvalue of the sum of the values of the condensates through the use of capacitive feedback in ren 90 and 91. The frequency bandwidth of the desired way to be processed can be selected, as it is in connection with signals of large amplitude (area III) is proportional to F i g. 2 had been explained. tional to the reciprocal of the sum of the values of the For both circuits according to the F i g. 2 or 7, capacitors 90, 91 and 92. the entire in FIG. 5 transfer function shown

The in F i g. 3 circuit shown agrees with the one modified to meet the requirements of a

conditions according to FIG. 2, with the exception that the Wi- 65 given system must be satisfied. For example, the

The status 87 in FIG. 3 is not present. The amplitude transmission in area I can be suppressed if

the transfer function of the circuit according to FIG. 3 the resistor 78 is removed and the diodes 81 and

is in Fig. 6 It differs from the one- 82 as well as short-circuiting the resistor 87 (by adding

Π

e.g. switches a wire across resistor 87). Signal production at signals low Levels in range 1 can be suppressed by removing resistor 78, as shown in FIGS. 3 and 6 is shown. Signals of large amplitude in the area 111 are not attenuated if the diodes 83 and 84 and the resistor 79 can be removed.

It should also be noted that the described Vertical detail signal processing circuitry is not affected by changes in the DC voltage level of the Luminance component is influenced. Because of the way a comb filter is a comb filtered Derives the luminance signal by applying a known subtractive signal combination, if the comb-filtered color signal has a direct voltage component of zero. This DC voltage component of the comb-filtered color signal is therefore disturbing not the DC bias component that arises at the comb filter color output. The processing the vertical detail signal, which is derived at the comb filter color output, can therefore be more predictable Way to be centered around a DC bias component when the comb filter color output is galvanically coupled to the vertical detail signal processing circuit. Since the reference level to the the signal processing takes place, is firmly predictable, one obtains well-defined restoration, enhancement or attenuation areas.

For this purpose 3 sheets of drawings

35

40

45

50

55

60

Claims (7)

1 2 ρ _. The invention relates to a signal processing switchgear claims: device as described in the preamble of claim j. 1. Signal processing circuit with an offset In particular, it is a more powerful circuit, between the output and input of which is a first feedback branch for increasing or attenuating vertical image detail, which provides a first impedance in a color television receiver, which has a (76) contains, and a second feedback comb filter for separating the luminance and color branch, which contains a second impedance (78), contains components of a color television signal, are switched, characterized in that DE-OS 2264282 discloses a circuit arrangement in the second feedback branch in series with voltage with non-linear transfer characteristic for Fernder second impedance (78) between the output of the 10 sehsignale known which the gamma correction, i.e. amplifier (75) and the output (S ₀ ) of the signal to compensate for the curvature of the characteristic of a remote processing circuit (50) serves in both directional picture tube. For this purpose, there is a non-linear Vergen conductive switch (81, 82), which is used more strongly with two negative feedback branches, that it has a linear impedance for signals of both polarities and the other a non-linear impedance within a first amplitude range Also contains from the book nen line condition and within a second, "semiconductor circuit technology" by Tietze and Schenk, above the first amplitude range Springer-Verlag, Berlin-Heidelberg-New York 1974, takes another line condition Page 262, a precision Full wave rectifier with 2. Signal processing circuit according to claim 1, known earthed output, which is kinked with a characterized in that it operates in parallel to the second characteristic curve and, for this purpose, also uses a counter-impedance (78), the series connection of a third amplifier coupled to it pedanz (79) with an additional, in both lines, in a color television system, such as the switch (83, 84) that is conductive in the fences, the system developed by the United States, the luminous signals of either one or the other polarity are dense. and color components of a color television signal in terms of frequency within the first and second amplitude ranges within the video frequency spectrum ches a line state and nested within a third, with the luminance compot, above the first and second am- nents, with integer multiples of the horizontal time range, taking a different line state and the color components at un. wheel-number multiples of half the line scanning frequency 3. Signal processing circuit according to claim 1, 30 are located. There are different comb filter arrangements characterized in that in parallel with the first to separate the interleaved in terms of frequency Switch (81,82) has a resistor (87). Luminance and color components of the video signal 4. Signal processing circuit according to claim 3, known, for example in US-PS 41 43 397 (inventor characterized in that parallel to the second of D.D. Holmes) and US-PS 40 96 516 (inventor Impedance (78) the series connection of a third impedance 35 D. H. Pritchard) and the literature teledance (79) cited there with an additional one in both rows. lines conductive switch (83, 84) is located at Si A comb-filtered luminance signal, which is on If either one polarity or the other occurs, the luminance output of the comb filter is over within the first and second amplitude ranges, its entire band is subject to the comb filtering effect. ches a conduction state and within a third 40 thrown. The comb filtering over the high frequency th, above the first and second lying on the tape part, which is shared with the color signal components If the amplitude range becomes a different line state, this has the desired effect of removing occupies. Luminance signal components. An extension of the 5. Signal processing circuit according to claim 4, ser comb filtering in the low-frequency band part, characterized in that it is shared for signals within 45 not with the color signal components, half of the first amplitude range a first Si is not necessary for the desired distance gain, which is greater than zero, for voltage of the color signal components and brings only one Signals within the second amplitude range have unnecessary suppression of the luminance signals a second signal amplification which has larger components. Components at the bottom of the no than the first, and for signals within the third 50 split band removed in this way, In the amplitude range, a third signal gain corresponds to a vertical detail luminance information that is smaller than the first signal gain. It is desirable to retain such vertical details. worth seeing to a loss of vertical resolution in the 6. Signal processing circuit according to claim 4, luminance content of the reproduced image to be characterized in that one of the feedback 55 avoid. branches a filter (76, 90; 78, 91 j 79, 92) for an arrangement to preserve the vertical details Determination of its frequency characteristic contains. formation uses a low-pass filter that is applied to the output 7. Signal processing circuit according to claim 4, the comb filter is connected to which the characterized in that the first switch (81, comb filtered color components appear 82) in the second feedback branch at Signa- 60 upper limit frequency of this filter is below the len within the second and third amplitude range occupied by the color signal components to prevent a current flow through the (and be just below for illustration the resistor (87) is conductive, and that the additional selected from 2 MHz). The filter selectively couples signals before switch (83, 84) for signals in the first and two below the color band from the color output of the th amplitude range to prevent a comb filter from being connected to a combination circuit, where the se-current flow through the third impedance (79) signals non-conductive coupled with the comb-filtered tend is. Luminance output signals from the comb filter sum- be mated. The combined signal contains one