JPH0817480B2 - Video signal processing device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video signal processing device for restoring a high-definition television signal band-compressed by offset subsampling.

[0002]

2. Description of the Related Art The band of high-definition television signals is 20 MH
z or more, and it is necessary to compress the band by some method when transmitting using a satellite or the like. As an effective technique for significantly compressing the band of high-definition television signals, the MU that compresses the signal band by multiple sub-Nyquist sampling
SE (Multiple Sub-Nyquist Sampling Encoding) system has been proposed (Ninomiya et al., "Satellite 1-channel transmission system for high-definition television (MUSE)" Technical Report TEBS 95-2, pages 37-42).

This MUSE system performs offset sub-sampling between fields and frames as shown in FIG. 5, transmits 1/4 sample points of the original signal sample points, and completes 4 fields. A picture is transmitted, and the band of the original signal is compressed to 1/4. On the receiving side, sample points that have not been transmitted (non-sample points) are interpolated from the transmitted sample points to restore the original signal. For example, for a still image area, interpolation is performed using sample points of the current field and all three past fields. On the other hand, for the moving image area, interpolation is performed only at the sample points in the transmitted field. As described above, since the still image portion and the moving image portion have different interpolation methods, the receiving side detects the movement of the image and switches the interpolation method to restore the image.

By the processing on the transmitting side and the receiving side as described above, high-definition television broadcasting can be enjoyed at home. However, the above signal processing circuit (MUSE decoder)
Is very expensive and it will take a considerable amount of time for it to be popularized for home use. Therefore, there is a device that converts a high-definition television signal into an existing standard television signal by using a system conversion circuit in order to display and record / reproduce on a standard television or VTR that is now widely used in homes. Among them, a video signal processing circuit that performs interpolation processing to restore an image with a simple configuration has been proposed. As a conventional video signal processing device, for example, “video signal system conversion device” Japanese Patent Application No.
No. 2-328707.

FIG. 6 is a block diagram of this conventional video signal processing apparatus, in which 101 is an input terminal for inputting a MUSE signal, and 102 is a non-sample from a sampling point in a field for restoring a band-compressed MUSE signal. An interpolation circuit for interpolating points, 103 and 104 for field memories, 105 for an intermediate value selection circuit for extracting an intermediate value signal of signals between three adjacent fields, 106 for an adder for performing averaging processing, 1
Reference numeral 07 is an output terminal for extracting a high-definition television signal from which aliasing distortion has been removed.

In the conventional video signal processing device having the above components, the MUS input to the input terminal 101 is used.
The E signal is interpolated from sample points in the field to non-sample points in order to restore the band-compressed signal in the interpolation circuit 102. The signal interpolated in the field corresponding to the moving image processing of the MUSE decoder by the interpolation circuit 102 has a signal passing characteristic as shown in the frequency characteristic diagram of FIG. The high-definition component of the high-definition television signal due to the offset sub-sampling of the MUSE method as shown in the frequency characteristic diagram of b) is reproduced as aliasing distortion and deteriorates the image quality. Therefore, the signals subjected to the interpolation processing are connected to the field memories 103 and 10 connected in cascade.
4, the intermediate value signal of the signals between the three adjacent fields is extracted by the intermediate value selection circuit 105, and the intermediate value selection circuit 105 outputs the average value signal of the signals having correlation between the fields. Output signal and frame memory 10
The output signal of No. 3 and the output signal of No. 3 are averaged by the adder 106, and a signal from which aliasing distortion caused by inter-field offset subsampling is removed is obtained at the output terminal 107.

As shown in FIG. 8, the intermediate value selection circuit 105 compares the magnitudes of two input signals from each of the three input signals and detects the signal having an intermediate value in terms of amplitude. , The selection means is switched according to the detection result, and the intermediate value signal is output. A logical value table for selecting an intermediate value in the intermediate value selection circuit 105 is shown in (Table 1).

[0008]

[Table 1]

The operation of removing the aliasing distortion will be described below with reference to FIG. FIG. 9 shows an example of a signal waveform for explaining the operation of removing the aliasing distortion.

9A and 9B at a still picture area where there is no movement of the input television signal or a change point of the movement.
Since the intermediate value signal X (indicated by ● in the figure) extracted by the intermediate value selection circuit 105 becomes one of the preceding and succeeding signals, the signal Y of the current field (indicated by ◎ in the figure) and the adder By averaging in 106 and taking the sum between the fields, the aliasing component due to the inter-field offset subsampling is removed. On the other hand, in the moving image area in which the input TV signal is moving, the intermediate value signal X extracted by the intermediate value selecting circuit 105 becomes the signal Y of the current field as shown in FIG. Signal itself is obtained and the image quality is not deteriorated. Therefore, since the averaging of the signals between the fields having strong correlation between the three adjacent fields is adaptively performed, the aliasing distortion due to the inter-field offset sub-sampling of the MUSE method can be removed without affecting the moving image. You can

[0011]

However, in the above-mentioned configuration, the aliasing distortion caused by the inter-field offset sub-sampling is performed by performing optimum inter-field averaging processing between three adjacent fields in the adder 106. The 30 Hz flicker component, which appears to have jagged distortion attached to the vertical lines of the image, disappears, but the band-compressed MUSE signal also has aliasing distortion due to interframe offset subsampling. Included, 30 in the diagonal component of the image
There was a problem that flicker of Hz remained.

The present invention solves the above-mentioned problems, and removes not only the aliasing distortion caused by the inter-field offset subsampling but also the aliasing distortion caused by the inter-frame offset subsampling which is visually conspicuous by aliasing to a low frequency band. It is an object of the present invention to provide a video signal processing device capable of performing the above.

[0013]

In order to achieve the above object, the present invention inputs a high-definition television signal band-compressed by offset sub-sampling, and samples in a field to restore the sub-sampled signal. Interpolation means for interpolating non-sampled points from points, first and second field memories cascaded for storing the interpolated signal, and input / output of two field memories to be adjacent to each other 3 Intermediate value selecting means for extracting an intermediate value signal of signals between fields, a first adder for averaging the output of the first field memory and the output of the intermediate value selecting means, and an input of the first field memory And a second adder for averaging between frames from the output of the second field memory, and 2
The field difference and the frame difference are detected from the input and output of one field memory, and the output of the first adder and the second addition are performed by the output of the correlation determining unit and the correlation determining unit that outputs the switching signal according to the magnitude of the difference. A signal switching device for switching the output of the signal processing device, a motion detecting means for detecting the motion of the image from the input of the first field memory and the output of the second field memory, and the output of the signal switching device according to the detected motion. A mixer is provided for mixing the outputs of the interpolating means.

A high-definition television signal band-compressed by offset subsampling is input, and first and second cascaded field memories for storing the input television signal and two field memories are connected. First, second and third interpolation means for interpolating non-sampled points from sampled points in the field to recover the sub-sampled signals from the input and output and the outputs of the three interpolation means are adjacent An intermediate value selecting means for extracting an intermediate value signal of signals between three fields, a first adder for averaging the output of the second interpolating means and the output of the intermediate value selecting means, and a first
Second adder for taking an average between frames from the output of the interpolation means and the output of the third interpolation means, and the field difference and the frame difference are detected from the outputs of the three interpolation means, and the magnitude of the difference is detected. In accordance with the output of the correlation discriminating means and the output of the correlation discriminating means for switching the output of the first adder and the output of the second adder, and the output of the first interpolating means and the first interpolating means. 3 from the output of the interpolation means, or the first
Motion detection means for detecting the motion of the image from the input of the field memory and the output of the second field memory, and mixing for mixing the output of the signal switcher and the output of the first interpolation means according to the detected motion. It has a configuration with a container.

[0015]

According to the present invention, in the above structure, the television signal interpolated in the field is supplied to the two field memories, the intermediate value signal between the three adjacent fields and the signal of the current field are summed, and the adjacent signal is added. Either before or after the fold-back caused by inter-field offset sub-sampling by taking the average between correlated fields, and the fold-back caused by inter-frame offset sub-sampling by taking the average between frames. A signal from which distortion is removed is obtained, and which distortion is larger is determined based on the magnitude relationship between the inter-field difference and the inter-frame difference, and switching is performed to remove not only the field but also the aliasing distortion caused by the inter-frame offset subsampling. be able to. Furthermore, a signal in which the movement of the image is detected and the averaging processing between the adjacent fields before and after the correlation and the averaging processing between the frames are switched, and the signal processed in the field By performing the mixing process of and, it is possible to prevent the deterioration of the image quality such as the occurrence of a double image even in the moving image.

Further, the capacity of the field memory can be reduced by the structure in which the field memory is placed in advance so as to store the signal before being interpolated.

[0017]

(Embodiment 1) FIG. 1 is a block diagram of a video signal processing apparatus according to a first embodiment of the present invention. In FIG. 1, 11 is an input terminal for inputting a MUSE signal, 12 is an interpolation circuit for interpolating non-sampled points from sampled points in the field to restore a band-compressed signal, 13 and 14 are field memories, 15 Is an intermediate value selection circuit for extracting an intermediate value signal of signals between three adjacent fields, 16 and 17 are adders for performing averaging processing, 18 is a field difference detection circuit for detecting aliasing distortion due to interfield offset subsampling, Reference numeral 19 is a frame difference detection circuit that detects aliasing distortion due to interframe offset subsampling, 20 is a correlation determination circuit that determines which of field and frame aliasing distortion is larger, 21 is a signal switcher, 22 is a motion detection circuit, 23 is a mixer, 24
Is an output terminal for taking out a high-definition television signal from which aliasing distortion is removed.

The operation of the video signal processing device having the above components will be described. MU input to the input terminal 11
The SE signal is interpolated from sample points in the field to non-sample points in order to restore the band-compressed signal in the interpolation circuit 12. Next, the interpolated signal is supplied to two field memories 13 and 14 connected in cascade,
After the intermediate value signal of the signals between the three adjacent fields is extracted in the intermediate value selection circuit 15 from the input / output signals of the two field memories 13 and 14, it is averaged with the output signal of the field memory 13 by the adder 16. A signal from which aliasing distortion caused by inter-field offset subsampling is removed is obtained. The processing up to this point is similar to that of the conventional video signal processing device. In this embodiment, the output signal of the interpolation circuit 12 and the output signal of the field memory 14 are subjected to averaging processing between frames by an adder 17 to remove aliasing distortion caused by inter-frame offset subsampling. Is obtained. Input / output signals of the two field memories 13 and 14 are supplied to a field difference detection circuit 18 and a frame difference detection circuit 19 to detect aliasing distortion due to interfield offset subsampling and aliasing distortion due to interframe offset subsampling. The correlation discriminating circuit 20 discriminates which of the folding distortion between the field and the frame is larger, and outputs from the adder 16 the signal from which the folding distortion due to the inter-field offset subsampling is removed and the adder 17. The signal switcher 21 is controlled so that the signal from which the aliasing distortion caused by the inter-frame offset sub-sampling is removed and the signal from which the aliasing distortion is stronger is removed.

Further, the motion detecting circuit 22 detects the motion of the image from the input signal of the field memory 13 and the output signal of the field memory 14, and the signal switch 21.
Output from the field interpolating circuit 12 and a signal in which the averaging processing between the adjacent fields before and after the correlation and the averaging processing between the frames are switched, and the field output from the field interpolation circuit 12 The signal that has been only internally processed is mixed in the mixer 23 according to the amount of movement, so that there is no deterioration in image quality (such as occurrence of double images) even in a moving image, and a signal in which aliasing distortion is almost removed is output. Obtained at terminal 24.

As described above, according to the first embodiment,
After the field interpolation corresponding to the moving picture processing of the MUSE decoder is performed in the interpolating circuit 12, the aliasing distortion occurring in the still image is subjected to the averaging processing between adjacent fields before or after the adjacent one. By performing the adder 16 from the output signal of the value selection circuit 15 and the output signal of the field memory 13, a signal in which aliasing distortion caused by inter-field offset subsampling is removed is obtained, and averaging processing between frames is performed. A signal from which the aliasing distortion due to the inter-frame offset subsampling is removed is obtained by the adder 17 using the output signal of the interpolation circuit 12 and the output signal of the field memory 14. These two signals are the fields for detecting the absolute value of the inter-field difference from the signals of the adjacent three fields as shown in FIG. 2 in order to detect the situation of the aliasing distortion between the fields and the frames from the signals of the adjacent three fields. By providing the difference detection circuit 18, the frame difference detection circuit 19 that detects the absolute value of the inter-frame difference, and the correlation determination circuit 20 that compares the two differences and detects the magnitude relationship between them, as shown in FIG. Folding distortion between fields and 15H
According to the magnitude of the aliasing distortion between frames that is the flicker component of z, the signal processed by the signal with the greater aliasing distortion is switched by the signal switcher 21 and output.
As a result, the signal switch 21 obtains a signal from which aliasing distortion caused by field and frame offset subsampling is removed.

(Embodiment 2) FIG. 4 is a block diagram of a video signal processing apparatus according to a second embodiment of the present invention. In FIG. 4, 31 is an input terminal for inputting a MUSE signal, 3
2, 33 are field memories, 34, 35, 36 are interpolation circuits for interpolating non-sample points from sample points in the field in order to restore the band-compressed MUSE signal, and 37 is a signal of three adjacent fields. Intermediate value selection circuits for extracting intermediate value signals, 38 and 39 adders for performing averaging processing, 4
0 is a field difference detection circuit that detects aliasing distortion due to inter-field offset subsampling, 41 is a frame difference detection circuit that detects aliasing distortion due to inter-frame offset subsampling, and 42 is which of the field and frame aliasing distortion is greater. A correlation discriminating circuit for discriminating, 43 is a signal switcher, 44 is a motion detecting circuit, 45 is a mixer, and 46 is an output terminal for taking out a high quality television signal from which aliasing distortion is removed.

The operation of the video signal processing device having the above components will be described. MU input to the input terminal 31
SE signal is connected to two field memories 3 connected in cascade.
2 and 33, and the band-compressed signals of the three adjacent fields stored in the interpolation circuits 34, 35 and 36 are interpolated from the sampling points in the field to the non-sampling points. Next, after the intermediate value signal of the signals between the three adjacent fields is extracted in the intermediate value selecting circuit 37 from the output signals of the interpolating circuits 34, 35, 36, the output signal of the interpolating circuit 35 A signal from which the aliasing processing has been performed and the aliasing distortion caused by the inter-field offset subsampling is removed is obtained. On the other hand, the output signal of the interpolation circuit 34 and the output signal of the interpolation circuit 36 are added by the adder 39, averaging processing between frames is performed, and aliasing distortion caused by inter-frame offset subsampling is removed. The signal is obtained. Further, the interpolated signals of three adjacent fields output from the interpolating circuits 34, 35, 36 are supplied to the field difference detecting circuit 40 and the frame difference detecting circuit 41, and the aliasing distortion due to the inter-field offset subsampling. And the aliasing distortion due to the inter-frame offset sub-sampling is detected, the correlation determining circuit 42 determines which of the field and the inter-frame aliasing distortion is larger, and the adder 38 outputs the adjacent one before and after the adjacent one. A signal from which aliasing distortion caused by inter-field offset subsampling is removed by taking an average between certain fields and a signal from which aliasing distortion caused by inter-frame offset subsampling is output from the adder 39 are removed. The signal with the stronger aliasing is removed. Signal switching unit 43 is controlled such.

Further, the motion detection circuit 44 detects the motion of the image from the output signal of the interpolation circuit 34 and the output signal of the interpolation circuit 36, and averages the fields or frames output from the signal switch 43. Since the processed signal and the signal which is output from the interpolation circuit 34 and only processed in the field are mixed according to the amount of motion detected by the mixer 45, image quality deterioration (2
A signal in which the aliasing distortion is almost eliminated is obtained at the output terminal 46 without the occurrence of a double image).

As described above, according to the second embodiment,
By providing a field memory for obtaining signals between three adjacent fields so as to store a television signal before being interpolated, aliasing distortion caused by inter-field and inter-frame offset sub-sampling of the MUSE system is almost eliminated. It can be completely removed and the field memory capacity can be cut in half.

In the first embodiment, the field difference detection circuit 18 uses all the signals of the three adjacent fields, but it may detect the difference between the preceding and following fields. Also, the motion detection circuit 22
Detects the motion of an image from a signal between 1 frames, but it is also possible to derive a signal between 1 frames and detect the motion from a signal between 2 frames. Also in the second embodiment, the motion detection circuit 44 has been subjected to interpolation processing 1
Although the motion of the image is detected from the signal between the frames, the motion of the image may be detected from the signal between the frames before being subjected to the interpolation process, and the same as in the first embodiment. It goes without saying that the motion may be detected from the signal between the two frames.

[0026]

As is apparent from the above embodiments, according to the present invention, non-sample points are interpolated from sample points in a field to restore a high-definition television signal band-compressed by offset subsampling. An interpolating means, first and second field memories connected in cascade for storing the interpolated signal, and an intermediate value signal of signals between adjacent three fields from the input / output of the two field memories. From the intermediate value selecting means to be extracted, the first adder for averaging the output of the first field memory and the output of the intermediate value selecting means, and the input of the first field memory and the output of the second field memory. A second adder that takes an average between frames and a field difference and a frame difference are detected from the inputs and outputs of two field memories, and a switching signal is output according to the magnitude of the difference. From the input of the first field memory and the output of the second field memory, the correlation discriminating means for switching, a signal switcher for switching the output of the first adder and the output of the second adder by the output of the correlation discriminating circuit, Since the motion detecting means for detecting the motion of the image and the mixer for mixing the output of the signal switching device and the output of the interpolating means according to the detected motion are provided, the image quality is not deteriorated even in the moving image. It is possible to provide a video signal processing device that removes not only aliasing distortion caused by inter-field offset subsampling but also aliasing distortion caused by visually noticeable interframe offset subsampling.

A high-definition television signal band-compressed by offset subsampling is input, and first and second cascaded field memories for storing the input television signal and two field memories are connected. First, second and third interpolation means for interpolating non-sampled points from sampled points in the field to recover the sub-sampled signals from the input and output and the outputs of the three interpolation means are adjacent An intermediate value selecting means for extracting an intermediate value signal of signals between three fields, a first adder for averaging the output of the second interpolating means and the output of the intermediate value selecting means, and a first
Second adder for taking an average between frames from the output of the interpolation means and the output of the third interpolation means, and the field difference and the frame difference are detected from the outputs of the three interpolation means, and the magnitude of the difference is detected. Correlation determining means for outputting a switching signal according to
The output of the first adder and the second
Of the image from the output of the first interpolating means and the output of the third interpolating means, or from the input of the first field memory and the output of the second field memory. Since the motion detecting means for detecting the motion and the mixer for mixing the output of the signal switching device and the output of the first interpolating means according to the detected motion are provided, the image quality does not deteriorate even in the moving image. Further, it is possible to provide a video signal processing device capable of removing aliasing distortion caused by inter-frame and inter-field offset sub-sampling and reducing the memory capacity.

[Brief description of drawings]

FIG. 1 is a block diagram of a video signal processing device according to a first embodiment of the present invention.

FIG. 2 is a detailed block diagram of a circuit for detecting a situation of aliasing distortion between a field and a frame in the embodiment.

FIG. 3 is a characteristic diagram for explaining an operation for discriminating folding distortion in the example.

FIG. 4 is a block diagram of a video signal processing device according to a second embodiment of the present invention.

FIG. 5 is a principle diagram for explaining a band compression method by MUSE offset subsampling.

FIG. 6 is a block diagram of a conventional video signal processing device.

FIG. 7 is a frequency characteristic diagram for explaining the occurrence of aliasing distortion caused by MUSE offset subsampling in field interpolation processing.

FIG. 8 is a block diagram of an intermediate value selection circuit in a video signal processing device according to the related art and the present invention.

FIG. 9 is a schematic diagram for explaining an operation of removing aliasing distortion between fields in a video signal processing device according to the related art and the present invention.

[Explanation of symbols]

 12, 34, 35, 36 Field interpolation circuit 13, 14, 32, 33 Field memory 15, 37 Intermediate value selection circuit 16, 17, 38, 39 Adder 18, 40 Field difference detection circuit 19, 41 Frame difference detection Circuit 20, 42 Correlation determination circuit 21, 43 Signal switcher 22, 44 Motion detection circuit 23, 45 Mixer

Claims (5)

[Claims] 1. Interpolation means for inputting a high-definition television signal band-compressed by offset sub-sampling and interpolating non-sampled points from sampled points in a field to restore the sub-sampled signal, First and second field memories connected in series to store the interpolated signal, and an intermediate value signal of signals between three adjacent fields from the input and output of the first and second field memories. The intermediate value selecting means for extracting, the first adder for averaging the output of the first field memory and the output of the intermediate value selecting means, the input of the first field memory and the second field memory. A second adder that averages between frames from the output and a field difference and a frame difference are detected from the inputs and outputs of the first and second field memories, and the difference is detected according to the magnitude of the difference. Correlation discriminating means for outputting a switching signal Te, the video signal processing apparatus provided with a signal switching unit for switching the output of said second adder and an output of the first adder by an output of said correlation discriminating means. 2. A first field memory input and the second field memory
2. The motion detecting means for detecting the motion of an image from the output of the field memory of 1), and the mixer for mixing the output of the signal switching device and the output of the interpolating means according to the detected motion. Video signal processing device. 3. A first and a second cascade connected for inputting a high-definition television signal band-compressed by offset subsampling and storing the input television signal.
Field memory and first, second and third interpolating non-sampled points from sampled points in the field to recover the sub-sampled signals from the inputs and outputs of the first and second field memories. Interpolating means, intermediate value selecting means for extracting an intermediate value signal of signals between adjacent three fields from outputs of the first, second and third interpolating means, and output of the second interpolating means. And a first adder for averaging the output of the intermediate value selecting means, and a second addition for averaging between frames from the outputs of the first and third interpolation means. And a correlation discriminating means for detecting a field difference and a frame difference from the outputs of the first, second and third interpolating means and outputting a switching signal according to the magnitude of the difference, and an output of the correlation discriminating means. And the output of the first adder and the Video signal processing apparatus provided with a signal switching unit for switching the output of the adder. 4. Motion detection means for detecting the motion of an image from the output of the first interpolation means and the output of the third interpolation means, and the output of the signal switcher and the output according to the detected motion. The video signal processing device according to claim 3, further comprising a mixer for mixing the outputs of the first interpolation means. 5. A first field memory input and the second field memory
4. A motion detection circuit for detecting the motion of an image from the output of the field memory of claim 1, and a mixer for mixing the output of the signal switching device and the output of the first interpolation means according to the detected motion. The described video signal processing device.