JP3410646B2 - Image quality correction circuit - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image quality correction circuit suitable for use in an image reproducing apparatus, and more particularly to a high-frequency component of an input image and a method for controlling the brightness of the input image. The present invention relates to an image quality correction circuit that varies the degree of contour enhancement in response. 2. Description of the Related Art As a conventional technique, an image quality correction circuit used in a DVD (Digital Versatile Disk) reproducing apparatus will be described as an example with reference to FIGS. FIG. 5 shows a DVD using a conventional image quality correction circuit.
FIG. 6 is a schematic configuration diagram showing a reproduction signal processing system of the reproduction apparatus, FIG. 6 is a waveform diagram of a main part in a conventional image quality correction circuit, and FIG. 7 is an explanatory diagram showing limiter characteristics of a limiter circuit in the conventional image quality correction circuit. Referring to FIG. 5, a reproduction signal processing system of a conventional DVD reproduction apparatus includes a disk 11 on which an image and an audio signal are recorded, an RF amplifier 12 for amplifying a reproduction signal reproduced by a pickup, and A digital demodulation / error correction circuit 13 for digitally demodulating the signal amplified by the RF amplifier 12 and for correcting an error of the demodulated data;
MPEG2 for expanding data compressed by the PEG2 method
A decoder 14, a Y / C separation circuit 15 for separating the expanded data into luminance data and color difference data, and a filter circuit 1 for extracting a specific frequency component of the separated luminance data
6 is provided. Further, an upsampler 17 for up-converting (raising) the sampling frequency of the luminance data generated by the filter circuit 16 to an arbitrary multiple, and a limiter circuit 18 for limiting the output data of the upsampler 17 in amplitude.
A down-sampler 19 for returning the sampling frequency of the amplitude-limited data to the sampling frequency before the up-conversion, and an adder 20 for adding the output data of the down-sampler 19 and the output luminance data of the Y / C separation circuit 15
And the output luminance data of the adder 20 and the Y / C separation circuit 15
And a Y / C mixing circuit 21 for mixing the output color difference data. Here, the filter circuit 16, upsampler 17, limiter circuit 18, downsampler 19, and adder 20 constitute an image quality correction circuit. Next, the operation of the reproduction signal processing system of the DVD reproduction apparatus will be briefly described. A reproduction signal reproduced from the disk 11 is amplified by the RF amplifier 12. The amplified signal is output to a digital demodulation / error correction circuit 13 where it is digitally demodulated, the error of the demodulated data is corrected, and the output is output to an MPEG2 decoder 14. It should be noted that this data is data compressed by the MPEG2 method at the time of recording. The output data is expanded by the MPEG2 decoder 14 and decoded into data before compression. The decoded data is supplied to the Y / C separation circuit 1
5 to be separated into luminance data and color difference data. The separated luminance data is output to an image quality correction circuit, which will be described in detail later, becomes data in which the outline is emphasized, and is output to the Y / C mixing circuit 21. Also, the separated color difference data is output to the Y / C mixing circuit 21 and
The C mixing circuit 21 mixes the input luminance data and color difference data, and outputs image data with an enhanced edge to the next stage. Next, the above-described image quality correction circuit will be described. The luminance data shown in FIG. 6A separated by the Y / C separation circuit 15 is supplied to the filter circuit 16, the upsampler 17, the limiter circuit 18, and the downsampler 1 described above.
9, is input to an image quality correction circuit composed of an adder 20, one of which is input to the adder 20, and the other is input to the filter circuit 16. First, the luminance data input to the filter circuit 16 is passed through a filter circuit 16 generally using a BPF (band-pass filter) or LPF (low-pass filter) at a specific frequency shown in FIG. The components are extracted. The extracted data is output to an upsampler 17 composed of a zero interpolation circuit 17a and an LPF 17b, subjected to zero interpolation processing, and up-converted (increased) to an arbitrary multiple of the sampling frequency to reduce aliasing noise. The signal is output to the limiter circuit 18. An output waveform from the upsampler 17 is shown in FIG. This output waveform is shown in FIG. 6 if the LPF 17b is an ideal filter.
As shown in (c), the waveform is the same as the output waveform of the filter circuit 16, but in practice there is no ideal filter, and the waveform is actually output as a slightly dull waveform. The data input to the limiter circuit 18 is limited in amplitude to a predetermined level and output to the downsampler 19. The limiter characteristic of the limiter circuit 18 is shown in FIG.
As shown in (1), it has a non-linear characteristic, and outputs as it is when the input amplitude does not exceed L, outputs L when it exceeds L, and outputs as it is when the input amplitude does not exceed -L. , -L, the output is -L. The output waveform of the limiter circuit 18 is shown in FIG. The data whose amplitude has been limited by the limiter circuit 18 is output to a down sampler 19 comprising an LPF 19a and a thinning circuit 19b, and is returned to the sampling frequency before the up-conversion of the sampling frequency. Is output. In addition, down sampler 19
Is not a rectangular wave as shown in FIG. 6 (e), because the waveform is slightly dull by the LPF 19b. An adder 20 adds the output data of the downsampler 19 and the output luminance data of the Y / C separation circuit 15 and outputs the image data whose outline is enhanced as shown in FIG. I do. As described above, by providing the image quality correction circuit, it is possible to obtain image data with emphasized outlines and to improve the image quality. In this case, the high frequency components of the luminance data are extracted. Then, the amplitude is limited by the limiter circuit 18 and the contour is emphasized by adding the data with the limited amplitude to the input image data. There was a problem that the degree decreased. Also, contour enhancement is performed irrespective of the input luminance data level, and the contour enhancement level is constant whether the input image is a bright image or a dark image. Therefore, when the contour enhancement level is set to a high level, the display screen becomes glaring when the input image is bright, and when the contour enhancement level is set to a low level, the display screen becomes blurred when the input image is dark. Was. Therefore, for example, Japanese Patent Application Laid-Open No. 3-237889.
JP, JP-A-6-141204, JP-A-6-2
Japanese Patent No. 45103 proposes a technique in which the gain of an outline emphasis component is controlled in accordance with the high-frequency component peak level of input image data. An image correction device using this technique is described below.
8 and 9, the same parts as those of the conventional example described with reference to FIGS. 5 to 7 are denoted by the same reference numerals, and description thereof will be omitted. Here, FIG. 8 is a block diagram showing another conventional image quality correction circuit, and FIG. 9 is a waveform diagram for explaining the operation of a peak level detection circuit in another conventional image quality correction circuit. Referring to FIG. 8, a conventional image quality correction circuit comprises:
A gain control amplifier 22 for amplifying the output data of the upsampler 17; a peak level detection circuit 23 for detecting a peak level of the output data of the upsampler 17 and outputting a gain control signal for controlling the gain of the gain control amplifier 22; It has. Next, the operation of the image quality correction circuit will be described. The high frequency component data input to the gain control amplifier 22 is amplified and the gain is controlled based on the gain control signal from the peak level detection circuit 23, and is output to the downsampler 19. On the other hand, the high frequency component data input to the peak level detection circuit 23 is between zero level and zero level (positive levels ab, cd, e) as shown in FIG.
~ F, g ~ h, i ~ j, k ~ 1, negative levels m ~ a, n
To c, f to o, p to g, q to i, and l to r) (positive levels D1, D2, D3 and negative levels shown in FIG. 9 are the same). When the peak level detecting circuit 23 detects a peak level between zero level and zero level of the high frequency component data, it calculates the following coefficients. Here, for example, the output level of the gain control amplifier 22 is D, and the above-mentioned peak levels D1, D
2. In the relationship with the D3 level (the same applies to the negative level), when the relationship of D1 = D D2 = 1 / 2D D3 = 2D holds, the coefficient of the D1 area = 1 The coefficient of the D2 area = 1/2 D3 area Becomes = 2. Accordingly, the peak level detection circuit 23 sets the gain control amplifier 22 such that the D1 region has an amplifier gain of 1 and the D2 region has an amplifier gain of 1/2 = 0.
5, the D3 region outputs a gain control signal for setting the amplifier gain to 2. Thus, the gain of the gain control amplifier 22 is controlled in accordance with the peak level of the high frequency component of the input image data to make the output constant, and is added to the input image data to enhance the contour. It is possible to perform sufficient contour emphasis without requiring a circuit, and to prevent the occurrence of whiteout in the contour emphasis portion, thereby realizing improvement in image quality. Further, JP-A-4-22275, JP-A-5-64038, and JP-A-6-46293 disclose a technique for controlling the gain of an outline emphasis component in accordance with the level of input image data. Proposed. According to this, while preventing the display screen from glaring due to prominent contour emphasis portions when the input image is bright, sufficient contour emphasis is performed even when the input screen is dark, so that It is possible to perform appropriate appropriate contour emphasis. However, in the conventional image quality correction circuit, when the degree of edge enhancement is varied in accordance with the peak level of the high frequency component of the input image data, the entire image is corrected. Since light and dark (low frequency components) are not taken into account, there is a problem that, for example, when the input image is dark, sufficient contour emphasis cannot be obtained, and optimal contour emphasis for the entire screen cannot be obtained. On the other hand, according to the level of the input image data,
In the case of varying the degree of contour emphasis, the peak level of the contour emphasis component (high frequency component) is not taken into consideration, so that an overshoot or undershoot may occur depending on the input image. There was a problem that it did not become edge enhancement. The present invention has been made in view of the above-described points, and varies the degree of contour emphasis in accordance with the high-frequency components and brightness of an input image, thereby achieving overshoot and undershoot. It is an object of the present invention to provide an image quality correction circuit capable of preventing occurrence of shoots and obtaining sufficient contour enhancement even when an input image is dark. An image quality correction circuit according to the present invention comprises a filter circuit for extracting a high frequency component of input image data, and a sampling frequency of the high frequency component extracted by the filter circuit. Up to any multiple
And upsampler, a peak level detecting circuit for detecting a peak level between the zero level from the zero level of the high frequency component output from the up-sampler, based on the detection result of the peak level detection circuit, said upsampler
A constant output gain control amplifier high frequency components are more output gain controlled to be constant level, an input level detecting circuit for detecting the level of the input image data, the detection result of the input level detection circuit based on the gain control amplifier for gain control of the high frequency components with a predetermined level in the constant output gain control amplifier, sampling of the output of the gain control amplifier
To the sampling frequency before up-conversion.
A down-sampler to be returned; and an adder for adding output data of the down-sampler and the input image data. Thus, based on the peak level of the high frequency component detected by the peak level detection circuit, the high frequency component level is controlled to be constant by the constant output gain control amplifier, and then detected by the input level detection circuit. based on the level of the input image data, and controls the high-frequency component level by a gain control amplifier, since added to the input image data, to prevent the occurrence of overshoot or undershoot, and a dark input image At times, it is possible to obtain sufficient contour enhancement. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS. 1 to 4. The same parts as those in the above-mentioned conventional example are denoted by the same reference numerals and the description thereof will be omitted. Here, FIG. 1 is a block diagram showing an image quality correction circuit of the present embodiment, FIGS. 2 and 3 are explanatory diagrams showing gain control control in the image quality correction circuit of the present embodiment, and FIG. 4 is an image quality correction circuit of the present embodiment. FIG. 9 is an explanatory diagram showing another example of the gain control characteristic based on the input level in FIG. Referring to FIG. 1, the image quality correction circuit of the present embodiment is based on a peak level detection circuit 23 for detecting the peak level of the output data of the upsampler 17 and a detection result of the peak level detection circuit 23. A constant output gain control amplifier 31 for amplifying the output data of the upsampler 17 to a constant level, an input level detection circuit 32 for detecting the level of the input image data, and a detection result of the input level detection circuit 32. A gain control amplifier 33 for controlling the gain of the output data of the constant output gain control amplifier 31 and outputting the output data to the downsampler 19; Next, the operation of the image quality correction circuit of this embodiment will be described. Here, the description will be made based on the background that the input image data is the luminance data separated by the Y / C separation circuit, corresponding to the above conventional example. First, the input image data shown in FIG. 2A is input to the filter circuit 16 and the input level detection circuit 32, and is also input to the adder 20. In general, a specific high frequency component is extracted from the image data input to the filter circuit 16 using a BPF (Band Pass Filter) or an HPF (High Pass Filter). The extracted high frequency component data is output to an upsampler 17 comprising a zero interpolation circuit 17a and an LPF 17b, subjected to zero interpolation processing, and up-converted (increased) to an arbitrary multiple of the sampling frequency. After the aliasing noise is reduced, the signal is output to the constant output gain control amplifier 31 and output to the peak level detection circuit 23. The peak level detection circuit 23 detects the maximum value between the zero level and the zero level of the input high frequency component data and calculates a coefficient for keeping the output of the constant output gain control amplifier 31 constant. Then, a gain control signal for controlling the gain of the constant output gain control amplifier 31 is output based on the calculation result. The gain of the high frequency component data input to the constant output gain control amplifier 31 is controlled based on the gain control signal from the peak level detection circuit 23 so that the output level becomes constant. The input level detection circuit 32 detects the level of the input image data and outputs a gain control signal having characteristics as shown in FIG. The high-frequency component data set to a constant level by the constant output gain control amplifier 31 is input to the gain control amplifier 33, and after the gain is controlled based on the gain control signal from the input level detection circuit 32, the LPF
The signal is output to the downsampler 19 composed of a thinning circuit 19a and a thinning circuit 19b. The data input to the downsampler 19 is returned to the sampling frequency before the sampling frequency is up-converted, and output to the adder 20. The adder 20 adds the output data of the downsampler 19 and the input image data, and outputs the image data whose outline is emphasized to the next stage. For example, in the peak level detection circuit 23,
As shown in FIG. 2B, between zero level and zero level (positive levels ab, ef, gh, jk,..., Negative levels bc, dee, hi) , K to l,...), That is, the positive levels D1, D2, and D2 shown in FIG.
D, D3..., The negative levels −D1, −D shown in FIG.
2, -D, -D3 ... are detected. As described above, when the peak level between the zero level and the zero level of the high frequency component data is detected, the peak level detection circuit 23
Then, the following coefficient is calculated. Here, the output level of the constant output gain control amplifier 31 is set to D or -D, and D1 = 1 / 2D in relation to the peak levels D1, D2, D3 and -D1, -D2, -D3. D2 = 2D D3 = 4D -D1 = -1 / 2D -D2 = -2D -D3 = -4D If the following relationship is satisfied, the coefficient of the D1, -D1 area = 2 or 1 The coefficient of the D2, -D2 area = A coefficient of 1/2 D, -D area = 1 A coefficient of D3, -D3 area = 1/4 is calculated. The peak level detection circuit 23 which has calculated such a coefficient provides the constant output gain control amplifier 31 with an amplifier gain of 2 or 1 D2, -D2 for the constant output gain control amplifier 31 based on the calculation result. In the region, the amplifier gain is 1/2 (= 0.5). In the D and -D regions, the amplifier gain is 1 D3 and in the -D3 region, the amplifier gain is 1/4 (= 0.2).
5) Output the gain control signal to be. In the above description, the reason why the amplifier gain in the D1 and -D1 regions is set to 2 or 1 is that the peak level is detected and the amplifier gain is uniformly controlled based on the detection result. This is because there is a method of controlling only the level equal to or higher than the D level which is the output level. Further, the input level detection circuit 32
As shown in (a), when N is the maximum level, the level of the input image data (for example, 0.7N,
0.6N, N, 0.8N, 0, 0.2N), and based on this detection level, a control gain K is calculated from a gain control characteristic as shown in FIG. Then, a gain control signal for controlling the gain of the gain control amplifier 33 is output. Referring to FIG. 3, the amplifier gain control of this embodiment will be described in more detail. FIG. 3A shows an output waveform of the upsampler 17. As described above, the output waveform of the upsampler 17 is controlled by the gain control signal from the peak level detection circuit 23 as follows, and the D1 and -D1 regions have the amplifier gain of 2 or 1 D2. , -D2 region has an amplifier gain of 1/2 (= 0.5) D, -D region has an amplifier gain of 1 D3, -D3 region has an amplifier gain of 1/4 (= 0.2)
5) In the constant output gain control amplifier 31, FIG.
Is corrected so that the peak level is constant. FIG. 3B is a diagram when the amplifier gain in the D1 and -D1 regions is set to 2. Further, the following control is performed according to the characteristic shown in FIG. 2C by the gain control signal from the input level detection circuit 32 (here, K = 1). 0.3 -D11 region has an amplifier gain of 0.4 -D21 region has an amplifier gain of 0.4 D21 region has an amplifier gain of 0 D 'region has an amplifier gain of 0 -D' region has an amplifier gain of 0.2 D31 region has an amplifier gain of 0 .2 -D31 region, amplifier gain 1 The gain control amplifier 33 outputs data as shown in FIG. Here, K = 1 has been described for easy understanding, but it is needless to say that K may be any value. In the above embodiment, the gain control characteristic based on the input level is shown in FIG.
Although the one shown in FIG. 4C was used, the invention is not limited to this.
Curve control as shown in FIG. Further, at the time of special reproduction or the like, as shown in FIG. 4B, it is possible to give a special effect to a display image by performing control to lower the amplifier gain at a higher input level. In the above embodiment, since the peak level of the high frequency component of the input image data is set to a constant level and the degree of edge enhancement is controlled in accordance with the input image data level, the gain control is performed. Amplifier 33
In the above, the calculation error of the gain control can be made as small as possible. That is, in the ordinary digital operation, the lower bits (below the decimal point) are truncated. However, as in this embodiment, the constant output gain control amplifier 3
By setting the peak level of the high frequency component to a constant level at 1 and then inputting it to the gain control amplifier 33, truncation of lower bits can be suppressed, so that contour emphasis control closer to the design can be performed. Is possible. In the above description, when the image quality correction circuit is applied to an image reproducing apparatus, the description has been made of the luminance signal processing system of the reproduction processing system. However, it is needless to say that the image signal correction circuit may be used for the chroma signal processing system. No. As described above, the image quality correction circuit of the present invention uses the constant output gain control amplifier 31 according to the peak level between the zero level and the zero level of the high frequency component of the input image data and the input image data level. Since the gain of the gain control amplifier 33 is controlled, it is possible to prevent overshoot and undershoot from occurring, and to obtain sufficient contour emphasis even when the input image is dark. According to the image quality correction circuit of the present invention, the high frequency component level is kept constant by the constant output gain control amplifier based on the peak level of the high frequency component detected by the peak level detection circuit. after control, based on the level of the input image data detected by the input level detection circuit,
The gain control amplifier controls the high-frequency component level and adds it to the input image data, so that overshoot and undershoot are prevented and sufficient edge enhancement is obtained even when the input image is dark. Is possible. Further, since the high frequency component is controlled to a constant level by the constant output gain control amplifier and then input to the gain control amplifier, the calculation error of the gain control of the contour emphasis component is suppressed, and It is possible to perform contour enhancement control close to the design.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of an image quality correction circuit according to the present invention. FIG. 2 is a waveform chart for explaining gain control control in one embodiment of the image quality correction circuit of the present invention. FIG. 3 is a waveform chart for explaining gain control control in one embodiment of the image quality correction circuit of the present invention. FIG. 4 is an explanatory diagram showing another example of a gain control characteristic based on an input level in an embodiment of the image quality correction circuit of the present invention. FIG. 5 is a schematic configuration diagram showing a reproduction signal processing system of a DVD reproduction device using a conventional image quality correction circuit. FIG. 6 is a main part waveform diagram in a conventional image quality correction circuit. FIG. 7 is an explanatory diagram showing limiter characteristics of a limiter control circuit in a conventional image quality correction circuit. FIG. 8 is a block diagram showing another conventional image quality correction circuit. FIG. 9 is a waveform diagram for explaining an operation of a peak level detection circuit in another conventional image quality correction circuit. [Description of Signs] 16 Filter circuit 17 Up sampler 19 Down sampler 20 Adder 23 Peak level detection circuit 31 Constant output gain control amplifier 32 Input level detection circuit 33 Gain control amplifier

Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N 5/208 H04N 5/781 H04N 9/68 G06T 5/20

Claims (1)

(57) Patent Claims 1. A filter circuit for extracting high frequency components of the input image data, high frequency components of the sump extracted by the filter circuit
Upsun to convert ring frequency to any multiple
A plug, and the peak level detecting circuit for detecting a peak level between the zero level from the zero level of the high frequency component output from the up-sampler, based on the detection result of the peak level detection circuit, the A
A constant output gain control amplifier high frequency components output from Ppusanpura to gain control so that a constant level, an input level detecting circuit for detecting the level of the input image data, the detection result of the input level detection circuit And a gain control amplifier for controlling the gain of a high frequency component that has been set to a constant level by the constant output gain control amplifier, based on the sampling frequency of the output of the gain control amplifier.
Return to the sampling frequency before up-conversion
An image quality correction circuit comprising: a sampler ; and an adder for adding output data of the downsampler and the input image data.