JPH0584995B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a color television receiver, particularly a so-called projection type receiver that obtains an image by projecting three primary color image lights onto a large screen from a projection tube. Regarding.

[Conventional technology]

Due to manufacturing considerations, CRTs with large image surfaces are currently limited to about 40 inches. Beyond that, a method using a projection tube is currently practical. In the case of a large, high-quality screen, high quality cannot be obtained simply by increasing the size of the screen, so as the number of scans increases, the requirements for image quality become stricter. In particular, with the projection type, the current density of the projection tube beam is about 5 to 10 times that of the direct view type, so the beam becomes thicker.
In addition, a decrease in resolution due to the influence of the lens and flare caused by the high-brightness projection tube, lens, etc. were causes of deterioration in image quality.

Perhaps because the projection-type large screen image receiver is still in the development stage, the means for the flare correction and contour correction described above have not yet been fully established as fully established technologies. Conventionally,
As a flare correction method proposed for high-definition televisions, "Improvement of Image Quality of Projection Displays for High-Definition Televisions - Removal of Flare Interference by SAW Filters -" Television Society 1982 National Conference SP1
−14, Kanazawa et al.'s video signal is once AM modulated,
There is a method using an analog filter that passes through a SAW filter and demodulates again. This method uses a modulated signal wave with ±
By applying attenuation at 1MHz, low frequency components are attenuated for flare correction. However, this method requires the use of a high frequency modulation carrier frequency of 100 MHz or more, and even if the horizontal flare component of the screen can be removed, if you try to apply it to the vertical component, you will have to use a very accurate one-line delay line. This is difficult to implement as a large number of them are required.

The most common method for contour correction is to extract the contour components of the image and add them to the original signal, but most of the methods only emphasize the horizontal direction of the screen, and the vertical direction is emphasized using some method of line delay. There are few examples because it requires creating a

By the way, in order to prevent resolution degradation, contour correction that emphasizes contours and flare correction that suppresses flare are two methods: the former emphasizes high-frequency components including differentials;
In the latter case, a screen with a lot of flare has a shape in which the MTF (resolution characteristics) is lifted in the low range, so it gives an attenuation characteristic to the low frequency components. Therefore, although contour correction and flare correction can be performed in parallel in terms of frequency, it is difficult to achieve this if digital and analog methods are mixed, or if one method is unified. Therefore, there is no example in which both types of correction processing were performed.

[Problem that the invention seeks to solve]

As mentioned above, large-screen projection receivers have not yet reached the stage of fully adopting the necessary flare correction and contour correction means to obtain high-quality images. Here, "completely" means that contour/flare correction can be performed for both vertical and horizontal components. In the analog system, there are problems such as uniformity of filter characteristics and temperature-related fluctuations in the delay line, and in a large-scale system, it is difficult to adjust the timing of all devices. In principle, a digital system can sufficiently deal with the above problems and is highly flexible in terms of design. However, there are difficulties as the scale increases. The problem lies in how to implement the digital correction device.

In view of the above-mentioned circumstances, an object of the present invention is to provide an image quality improvement device that removes flare components in the horizontal and vertical directions of the screen and at the same time performs corrections that emphasize the outline of the screen, all by digital means and on a small scale. It lies in realizing it in form.

[Means for solving problems]

The image quality improvement device of the present invention is provided for each series of three primary color video signals in a projection display type television receiver, inputs each video signal, performs A/D conversion, performs contour/flare correction, and performs D/D conversion. /
It is converted into A and output.

The contour/flare correction section is configured to generate a contour/flare correction signal, which is provided in parallel with the digital video signal input, and includes a compensation delay circuit and a subsequent gain adjustment circuit that changes the gain depending on the input level of the digital video signal. It consists of a creation circuit and a synthesis circuit that synthesizes the outputs of the compensation delay circuit and the gain adjustment circuit.

The contour/flare correction signal generation circuit performs contour correction and flare correction of the image in parallel. A high-pass FIR filter using a line memory and a high-pass FIR filter using an A/D conversion clock register as one delay in the horizontal direction are used. (b) Flare correction is performed in the vertical and horizontal directions of the image. It has two high-pass IIR filters using line memory as one delay in the vertical direction and an inverter that inverts one field's worth of information alternately in series, and in the horizontal direction. This is implemented by a circuit having two stages of high-pass IIR filters using A/D conversion clock registers as one delay and two stages of inverters that invert one line of information alternately in series.

Here, the filters and the coefficient circuit in the gain adjustment circuit use RAM and write data during the video blanking period, and during the video period, the input signal of the coefficient circuit is the address signal of the RAM,
The data is read out and becomes an output signal.

[Effect]

In the present invention, correction is provided separately for each series of three primary color video signals, so that appropriate correction can be obtained for each signal. Since the correction is all done digitally, it is input to the contour/flare correction section after A/D conversion.

The correction signal generation circuit includes a contour correction signal generation circuit and a flare correction signal generation circuit in parallel.
Contour and flare corrections are performed in parallel. The contour correction and flare correction are performed in parallel in the vertical and horizontal directions. As will be explained in detail in the embodiment, an FIR filter is used as the contour correction filter, and an IIR filter is used as the flare correction filter, giving them the characteristics of a high-pass filter. After correction, a gain adjustment circuit provided after the signal generation circuit generates a correction signal corresponding to the level of the digital video signal.

A synthesis circuit combines the above correction signal with the output that has passed through the compensation delay circuit. Next, by D/A converting the output of this synthesis circuit, a three primary color video signal with improved image quality can be obtained.

Although a coefficient circuit is required for the filter and gain adjustment circuit, it is possible to change the coefficient value by writing data to the RAM using RAM.

〔Example〕

An embodiment of the present invention will be described with reference to the drawings.
The basic configuration of the embodiment is shown in FIG. The video signals of the three primary colors are corrected by the same and independent circuits. In terms of the R signal, a digital video signal is generated by the A/D converter 11a, a correction signal generated by the correction signal generation circuit 13a and the gain adjustment circuit 17a, and a signal obtained by delaying the digital video signal by the compensation delay circuit 12a. and is synthesized by the synthesis circuit 14a. The compensation delay is performed by the correction signal generation circuit 13a.
This should be taken into account the delays caused by such factors. The digital video signal corrected in this way is sent to the D/A converter 15.
A is output as an analog signal. Note that in the following explanation, the explanation will be made without distinguishing between R, G, and B signals, so the suffixes in the symbols will be omitted.

Incidentally, due to the characteristics of the cathode ray tube, the input signal to the projection tube is a nonlinear signal obtained by raising the input signal to the gamma power. When filtering such input signals to create and add correction signals, the linearity of the correction signal itself is lost, and the sensitivity of the correction filter decreases in dark areas of the screen where the signal level is low, making it difficult to improve image quality in dark areas. effectiveness decreases. In order to improve this point, in the configuration shown in FIG. 1, the gain adjustment circuit 17 is made to continue with the correction signal generation circuit 13. The digital video signal is input to the gain adjustment circuit 17 via the delay device 16 as a control signal that compensates for the signal delay of the correction signal generation circuit 13. When the signal level is low, the gain of the gain adjustment circuit 17 is increased to compensate for the decrease in sensitivity of the correction filter in the dark areas. Conversely, the gain is lowered where the signal level is high. This improves the image quality even in dark areas of the screen.

Next, the correction signal generation circuit 13 will be explained. FIG. 2 is a circuit block diagram in which contour correction and flare correction are performed in parallel and independently. Since they can be executed independently of each other, by parallelizing them, the signal delay caused by correction is reduced. Further, for each correction, vertical correction and horizontal correction are performed in parallel.

Before explaining the overall configuration of FIG. 2, each filter will be explained. 21 and 22 are contour correction FIR filters for vertical and horizontal correction, respectively. Since the number of lines or dots involved in contour correction is small, it can be made small-scale by using an FIR filter (transversal filter) that can be converted into a linear phase as a digital filter. For example, as shown in FIG.
01a to 201d, and an adder circuit 202. The adder circuit 202 allows the coefficient circuits 201a to 201a to
Linear phase characteristics are obtained by matching the addition phases of 201d. The delay element 200 is a line memory in the case of vertical correction, and is a line memory in the case of horizontal correction.
This is a register for the D conversion clock.

Next, 23 and 24 are special flare correction filters for vertical and horizontal correction, respectively. Since flare correction involves a large number of lines and dots, an IIR filter (recursive filter) is used to reduce the scale. However, special means are required to obtain a linear phase with an IIR filter. In the present invention, an equivalent linear phase is obtained using a two-stage IIR filter in which the IIR filter output is inverted, the inverted output is further input to an IIR filter with the same characteristics, and the output is inverted. In the following, the composite
It is abbreviated as IIR filter.

FIG. 4 is a block diagram of the composite IIR filter 23 for vertical flare correction. IIR filter 230a
is a delay element 231, coefficient circuits 232a to 232d connected to each tap and input terminal, and an adder circuit 2.
It is of a well-known format synthesized in 33 steps. Here, the delay element 231 is a line memory. The output of the IIR filter 230a is inverted field by field by a field inverter 234a, and further
The signal is input to an IIR filter 230b, and its output is inverted by a field inverter 234b. The phase delay of the IIR filter 230a is reversed and the phase delay of the IIR filter 230a is reversed.
30b, the phase advances, so phase compensation is achieved.

Figure 5 shows composite IIR filter 2 for horizontal flare correction.
4 is a block diagram of FIG. The circuit configuration is the same as that of the composite IIR filter 23 for vertical flare correction. However, the difference is that the delay element 241 is a register for the A/D conversion clock, and is used as line inverters 244a to 244b.

The configuration of the filter has been described above, and the contour correction and flare correction involve emphasizing the high-pass characteristic and lowering the low-pass characteristic as frequency characteristics, and the filter is a high-pass type filter.

Since the filter has been explained above, the correction signal generation circuit 13 shown in FIG. 2 will be explained in general below.
The digital signal input is first input to the compensation delay device 25, and signals given a predetermined amount of delay are sent from the taps to signal lines l ₁ , l ₂ , m ₁ , and m _{2 ,} respectively. Signal line
The inputs from l ₁ and l ₂ are subjected to contour correction by a vertical contour correction FIR filter 21 and a horizontal contour correction FIR filter 22, respectively, and then synthesized by a synthesis circuit 28a. Inputs from m ₁ and m ₂ are subjected to flare correction by a vertical flare correction compound IIR filter 23 and a horizontal flare correction compound IIR filter 24, respectively, and then synthesized by a synthesis circuit 28b. Coring circuit 2
7a and 27b will be described later, but the outputs of the combining circuits 28a and 28b are further combined by a combining circuit 26 and output as a correction signal.

Signal lines l ₁ , l ₂ , output from the compensation delay device 25
m ₁ and m ₂ provide different amounts of delay to the digital signal. Determination of this amount of delay is quite complicated because it is determined so that the phases of the input signals to be combined are all matched in all of the combining circuits 28a, 28b, and 26. In this circuit, contour and flare corrections are performed in parallel in the vertical and horizontal directions. Another method is to perform the vertical and horizontal directions in series,
Compensation delay device becomes simple. However, if high-pass filters are used in series for vertical and horizontal correction, in the case of a quadrilateral wind pattern on the screen, the vertical and horizontal corrections are related, and the polarity is opposite to the one that should be improved. A polarity correction signal appears on the diagonal outer side diagonally to the inner corner of the window. In the present invention, since the vertical and horizontal directions are performed in parallel, the corrections are not related to each other, and good correction can be obtained even at the four corners of the window pattern.

The above correction outputs can be immediately synthesized in the synthesis circuit 26 to obtain a correction signal output, but in the circuit shown in FIG. 2, the signals are synthesized after passing through the coring circuits 27a and 27b. The correction signal emphasizes the high-frequency components of the signal, and this is particularly noticeable in contour correction. At this time, noise in the same high-frequency region is also emphasized, and the S/N tends to deteriorate in small areas of the screen. Therefore, the coring circuits 27a and 27b are circuits that set the correction signal to zero to prevent the noise from being emphasized in areas where the signal level is low and where noise is a problem. In other words, a dead zone is provided near zero in the correction signal, but this is not necessarily necessary when the screen to be projected is small.

The circuit configuration of the present invention has been described above, but the present invention requires a large number of coefficient circuits for various filters, gain adjustment circuits, coring circuits, etc. The coefficient values of each coefficient circuit vary, and often require adjustment and setting for each receiver. Therefore, it is necessary to be able to make adjustments at the final stage of manufacturing the receiver, and furthermore, even when the receiver is in practical use, changes in the installation environment will slightly affect the image quality. It is preferable that the user etc. can freely make adjustments in such cases as well.

In the present invention, a coefficient circuit that meets the above-mentioned demand is created by using an arbitrarily writable RAM.
FIG. 6a schematically shows a coefficient circuit, and FIG. 6b shows a circuit using RAM. In this circuit, data is written from a separate CPU during the video blanking period. In the figure, the mode of the selector 31 is set to the CPU address side, and data is transferred from the CPU to the RAM 32 via the buffer 33 to a predetermined address.
Write in. During the video period, the mode of the selector 31 is set to the coefficient circuit input side, and data multiplied by the coefficient is read out via the buffer 34 and sent out as the coefficient circuit output. Note that writing to RAM may be done in DMA mode. By setting data to be arbitrarily written from the CPU to the RAM 32, the coefficients of the coefficient circuit can be made variable.

〔Effect of the invention〕

As detailed above, the image quality of a large-screen projection television receiver can be significantly improved by performing contour and flare correction in parallel. The effects of the present invention include the following.

(1) Since correction is performed for each of the three primary color video signals, the correction is complete for any image. When a correction signal is extracted only from the G signal as in the conventional example, the correction effect is small for an image with a small G signal component, but the quality is high.

(2) The former is used as a contour/flare correction filter.
By using an FIR filter and a composite IIR filter for the latter, it is possible to create a small-scale circuit configuration with a linear phase and a small number of elements. This is also why it is possible to correct the three primary colors separately as shown in (1).

(3) Although the filter characteristics are high-pass type,
Since vertical correction and horizontal correction are performed in parallel, there is no mutual relationship, and good correction can be obtained even at the four corners of a quadrilateral window pattern. Furthermore, since all the filters are connected in parallel, there is little delay in the correction signal generation circuit.

(4) By providing a gain adjustment circuit and adjusting its gain according to the signal level, and increasing the gain when the signal level is low, it is possible to prevent the sensitivity of the correction filter from decreasing in dark areas of the screen due to the gamma characteristics of the video signal. Compensated.

(5) Coefficients can be set arbitrarily by using RAM as coefficient circuits used in filters, gain adjustment circuits, etc. Not only is it easy to adjust and set during the manufacturing stage, but it can also be made variable even while the receiver is in operation, making it possible to maintain optimal image quality in any environment.

[Brief explanation of drawings]

The drawings show one embodiment of the present invention, and FIG. 1 is a basic configuration block diagram, FIG. 2 is a configuration block diagram of a correction signal generation circuit, FIG. 3 is a configuration diagram of an FIR filter used for contour correction, and FIG. 4 - Fig. 5 is a block diagram of a composite IIR filter used for flare correction, and Fig. 6 is a diagram showing an example of a coefficient circuit. 11a-11c...A/D converter, 12a-1
2c... compensation delay circuit, 13a-13c... correction signal creation circuit, 14a-14c... synthesis circuit,
15a-15c...DA/converter, 16a-16
c...Delay device, 17a-17c...Gain adjustment circuit, 21-22...Contour correction FIR filter, 23
~24...Flare correction composite IIR filter, 25...
Compensation delay device, 26, 28a, 28b... synthesis circuit, 27a-27b... coring circuit, 200
...Delay element, 201a-201d...Coefficient circuit, 202...Addition circuit, 230a-230b...
...(High-pass type) IIR filter, 231, 241
...Delay element, 234a-234b...Field inverter, 240a-240b...(high-pass type) IIR filter, 244a-244b...Line inverter, 30...Coefficient circuit, 31...Selector, 32... ...RAM, 33, 34...batshua.

Claims (1)

[Scope of Claims] 1. In a projection display type television receiver, each of the three primary color video signal series is provided, each video signal is input and A/D converted, and after contour/flare correction, D/A The image quality improvement device converts and outputs the image, and the contour/flare correction section changes the gain depending on the input level of the digital video signal in a compensation delay circuit and a subsequent stage provided in parallel with the digital video signal input. The contour/flare correction signal generation circuit includes a contour/flare correction signal generation circuit equipped with a gain adjustment circuit that performs image contour correction, and a synthesis circuit that synthesizes the outputs of the compensation delay circuit and the gain adjustment circuit. (b) Contour correction is performed using a high-pass FIR filter that uses line memory as a 1-delay in the vertical direction and a high-pass FIR filter that uses line memory in parallel in the vertical and horizontal directions of the image, and in the horizontal direction. (b) Flare correction is performed by a high-pass FIR filter using the register of the A/D conversion clock as a 1-day delay, and (b) Flare correction is performed in parallel in the vertical and horizontal directions of the image, and in the vertical direction as a 1-day line memory. It has two stages of high-pass IIR filters using a high-pass IIR filter and an inverter that inverts one field's worth of information alternately in series. It is made up of a circuit that has two stages each of a pass-through IIR filter and an inverter that inverts one line of information, alternately arranged in series.The filters and the coefficient circuit in the gain adjustment circuit utilize RAM, and the video blanking period 1. A television picture quality improvement device, wherein data is written into the RAM, and during a video period, an input signal of the coefficient circuit is an address signal of the RAM, and the data is read out and becomes an output signal.