JPH08265690A - Digital video signal processor - Google Patents

Abstract

PURPOSE: To suppress the deterioration of video as much as possible, to minimize the increase of circuit scale and to lengthen recording time by performing a thinning processing for an inputted digital video signal, recording the signal by converting it into a prescribed format and converting the format into the original format by performing an interpolation processing at the time of a reproduction. CONSTITUTION: The thinning processing is performed for an inputted digital video signal 1 of a 4:4:1 format in vertical and horizontal directions by a thinning processing means 10 and the digital video signal is converted into an 8/3:2/3:0 format. Next, after a data shuffling is performed by a data shuffling means 30, a DCT processing and processings such as a variable length coding, etc., are performed by a high efficiency encoding means 40, information capacity is compressed, the information capacity is packed in a sink block and it is recorded in a magnetic tape 70. At time of a reproduction, this recorded video signal is converted into the 4:1:1 format by performing an interpolation processing for the signal after a sink block reproduction, highly efficient decoding and a deshuffling are performed for the signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital magnetic recording / reproducing apparatus for recording / reproducing a digital video signal, and more particularly to a signal processing method for recording / reproducing a digital video signal.

[0002]

2. Description of the Related Art With the recent increase in the size of home color television screens, the image quality of recording / reproducing media for video signals has been improved. Also, as a storage medium for recording and reproducing high-quality images with high image quality, a digital magnetic recording / reproducing device for home use (hereinafter referred to as a digital magnetic recording / reproducing device that digitizes a video signal and performs band compression (high efficiency coding) on the recording / reproducing) , Digital VTR) are being actively developed by each company.
In the digital VTR, a high-efficiency coding technique for reducing the data amount of an input video signal is usually used to realize recording for a long time. An example of this high-efficiency coding technique is a technique that uses orthogonal transform coding by discrete cosine transform (hereinafter referred to as DCT). This orthogonal transform coding is a technique for converting a video signal into horizontal and vertical components of a video. Encoding is performed on each frequency component obtained by frequency decomposition.

This is a method of reducing the data amount of an input video signal by using a technique such as reducing the data allocation amount for a high frequency component which is less affected by deterioration in visual characteristics, and is effective with little deterioration in image quality. It is one of the most popular data reduction technologies and is the most commonly used one. Furthermore, there is a variable-length coding technology that assigns variable-length codes with different code amounts according to the value of data obtained by orthogonal transformation. This is a method of allocating short codewords to reduce the total amount of data.

As a conventional example, a conventional digital VTR
A method of recording and reproducing the video signal in the above will be described with reference to FIG. In the figure, 1 is digital video input data, which is input in the 4: 1: 1 format described later. Thirty
Is a data shuffling means for shuffling the input data, and a high efficiency encoding means 40 in the subsequent stage.
At, pre-processing is performed to uniformly compress the entire one frame. A simple explanation is that in one video signal frame, a fine video part is locally localized, or a part where only the blue sky is reflected is biased, and the video quality is biased during high-efficiency coding. This is done to prevent this.

High efficiency coding means 4 after data shuffling
At 0, the above-mentioned DCT processing and variable length coding processing are performed to compress the information amount. The compressed information is packed by sync block packing means 50 into information units called sync blocks for recording on the tape, and is recorded on magnetic tape 70 via recording amplifier 60. The detailed operation of the sync block packing means 50 will be described later.

At the time of reproduction, the reverse processing of the above-mentioned processing is performed and the video data is restored again. First, the data read from the magnetic tape by the reproducing amplifier 160 is converted from the sync block information unit into compressed data by the sync block reproducing means 150. The video data compressed by the high-efficiency decoding means 140 is decompressed into the data before compression, then deshuffled by the data deshuffling means 130, and output again as the video data 2.

Next, the operation will be sequentially described according to the flow of signals. FIG. 12 shows the pixel configuration of one frame. The luminance signal component (hereinafter abbreviated as Y component) is composed of horizontal 720 pixels and vertical 480 pixels, and the color difference signal component is BY color difference component (hereinafter B-
It is composed of a Y component or Cb component) and an RY color difference component (hereinafter abbreviated as RY component or Cr component), and is composed of 180 horizontal pixels and 480 vertical pixels, respectively. This format is called the 4: 1: 1 format. It should be noted that the reason why the number of pixels of the color signal is small is that the human resolution ability of the color difference signal is lower than that of the luminance signal. For the NTSC signal, this number of pixels has a sufficiently high image quality.

Next, for each of the Y, Cb, and Cr components,
It is divided vertically into blocks of 8 × 8 pixels. In addition, this 8
The × 8 pixel block is called a DCT block, and is a target unit for data reduction by orthogonal transform coding which is one of the data amount reduction techniques. Further about Y component 4
A block obtained by collecting one DCT block for each of the DCT block and each Cb / Cr component and dividing it into a certain range on the screen is called a macro block. The structure of this macro block is shown in FIG. Shown in 14.

When the macroblock is constructed according to the above description, one frame is horizontal 2 as shown in FIG.
It consists of 2.5 macroblocks and 60 vertical macroblocks. The data reduced by the orthogonal transform coding, which is one of the data amount reduction techniques, is further variable length encoded and recorded on the tape. At this time, as a minimum synchronization unit for reading data at the time of reproduction, it is usual to record the data by dividing it into sync blocks of a constant data amount unit.

FIG. 16 shows the data arrangement in the sync block. SYNC in the figure means a synchronization pattern,
The reproduction data is synchronized by reading this pattern during reproduction. The ID is identification information indicating where the recorded data should be located on the screen,
DATA is video data reduced by the orthogonal transform coding and the variable length coding.

Since the data amount in the case of performing the variable length coding as described above varies depending on the design of the input image, the recording time should not be different depending on the design when recording on the magnetic tape. It is necessary to perform recording so that the amount of data becomes constant within a certain fixed range (hereinafter, referred to as compression rate control basic range, or simply compression basic range, compression basic amount).

Further, if the reading fails once for some reason during reproduction, if variable length coding is used, not only the part does not return to the original data, but also the code delimiter itself cannot be discriminated, resulting in an erroneous result. There is a problem that the data interpretation is continued after that. This situation is called error propagation, and it is better for the propagation of this error that the basic range of the compression rate control is smaller in order to reduce the amount of error propagation. However,
If the compression basic range is too small, the compression effect by the variable length coding is reduced, and the image quality is deteriorated as a result because the target data amount is reduced.

Therefore, in the method of recording / reproducing in sync block units, data is set so that a fixed data amount is obtained by using m units (m is a positive integer) of rectangular units of pixels of an input video signal subjected to orthogonal transformation as a compression unit. A method in which the amount is controlled and the compression unit is divided and recorded before and after about 5 sync blocks is usually used. FIG. 17 shows an example in which data for the m rectangular units is assigned to the 5 sync blocks. In the figure, D1, D2, ...
The area occupied by Dm indicates the area in which the data for each of the above rectangular units is recorded.

As described above, when data of one sync block is not detected and reproduced due to an error that occurs during reproduction, error propagation for the error extends to the above five sync blocks, that is, m rectangular units. The m rectangular units are usually subjected to a correction process or the like that replaces the data at the same position in the temporally adjacent front and rear screens, and the position of the data included in the wrong compression unit on the screen is the same. The correction process can be performed by replacing with 5 sync blocks for a certain compression unit. Regarding the problem of error propagation, for example, Japanese Patent Laid-Open No. 4-160990.
Although the improvement method is described in Japanese Patent Publication, etc., it is not directly related to the present invention, and the detailed description is omitted.

As described above, data for one frame is compressed on a macroblock basis and then recorded on the tape. At this time, if data compression is performed so that 5 macroblocks fit exactly into 5 sync blocks, 1350 sync blocks are required because the number of macroblocks in 1 frame is 1350. Here, when recording one frame of compressed data using ten tracks for recording on the tape as shown in FIG. 18, it is sufficient to use 135 sync blocks per track. The data format is a conventional format. Regarding the recording format on the above-mentioned tape, for example, Japanese Patent Laid-Open No. 6-24518.
Since it is described in Japanese Patent Publication No. 9, the detailed description thereof will be omitted.

[0016]

In the digital VTR using the above-mentioned conventional format, when recording for a long time such as doubling the recording time, it is necessary to simply increase the compression ratio of the image signal. There is a problem that the image deterioration is remarkable and it cannot be put to practical use. Moreover, in the conventional format, the signal processing becomes complicated because the macroblock structure of one frame is not an integer value in the horizontal direction.
There is also a problem that the circuit scale increases. The present invention solves the above-mentioned problems, and makes it possible to realize further long-term recording by a minimum change while maintaining a practically sufficient image quality, and minimizes an additional change of an actual circuit, thereby reducing the conventional format. It is an object of the present invention to obtain a digital video signal processing device having a long-time recording format that can coexist with both.

[0017]

According to a first aspect of the present invention, there is provided a digital video signal processing device, wherein an input video signal is 4: 4.
In a digital video signal processing device for recording a digital video signal of 1: 1 format on a magnetic tape and performing signal processing for reproducing the recorded digital video signal, the input digital video signal of the 4: 1: 1 format Thinning processing means for converting the signal into a predetermined format by performing thinning processing in both the horizontal and vertical directions, a shuffling means for shuffling the output signal of the thinning processing means, and information on the output signal of the shuffling means. A high-efficiency encoding means for compressing the amount, a sync-block packing means for packing the signal compressed by the high-efficiency encoding means into a sync block for each predetermined information amount unit, and a magnetic output signal from the sync-block packing means. Recording means for recording on tape, and signals recorded on magnetic tape A reproducing unit for reproducing, a sync block reproducing unit for reproducing the sync block when the signal reproduced by the reproducing unit is used as a basic unit for recording the sync block, and a high efficiency decoding for decoding the output signal of the sync block reproducing unit with high efficiency. The conversion means, the deshuffling means for deshuffling the output signal of the high efficiency decoding means, and the interpolation processing means for converting the output signal from the deshuffling means into a 4: 1: 1 format by performing interpolation processing both horizontally and vertically. Be prepared.

According to a second aspect of the present invention, there is provided a digital image signal processing device, further comprising first area filtering means provided between the thinning processing means and the shuffling means for attenuating a frequency component in an oblique direction of the signal.

According to a third aspect of the present invention, the digital video signal processing device is provided with the second area filtering means between the deshuffling means and the interpolation processing means for attenuating the frequency component in the oblique direction of the signal.

According to a fourth aspect of the present invention, there is provided a digital video signal processing device, wherein first area filtering means for attenuating a diagonal frequency component of a signal is provided between the thinning processing means and the shuffling means, and the deshuffling means is provided. The second area filtering means for attenuating the diagonal frequency components of the signal is provided between the above and the interpolation processing means.

According to a fifth aspect of the present invention, there is provided a digital video signal processing device for selecting and outputting the output signal of the thinning processing means during recording, and selecting and outputting the output signal of the deshuffling means during reproduction. A single area filtering means for attenuating a frequency component in an oblique direction with respect to the signal from the signal selecting means is provided, and an output signal of the single area filtering means is input to the shuffling means during recording, and a single area filtering means is provided during reproduction. The output signal of the area filtering means is input to the interpolation processing means.

According to the sixth aspect of the present invention, the first area filtering means of the digital video signal processing device performs at least one of the vertical direction and the horizontal direction to perform conversion to a predetermined format and diagonal direction. The filter coefficient is set so as to attenuate the frequency component of.

The second area filtering means of the digital video signal processing device according to claim 7 of the present invention performs interpolation processing of the output signal from the deshuffling means in at least one of the vertical direction and the horizontal direction: The filter coefficient is set so as to convert to the 1: 1 format and attenuate the frequency components in the oblique direction.

According to the eighth aspect of the present invention, the 4: 2: 2 format is used as the format of the video signal input to the thinning processing means during recording and the video signal output from the interpolation processing means during reproduction. Used.

According to a ninth aspect of the present invention, there is provided a digital video signal processing device for performing signal processing for recording a digital video signal of 4: 1: 1 format or 4: 2: 2 format as an input video signal on a magnetic tape. In the digital video signal processing device, the input 4: 1:
Thinning processing means for converting a digital video signal of 1 format or 4: 2: 2 format into a predetermined format by performing thinning processing in both horizontal and vertical directions, and a shuffling for shuffling the output signal of the thinning processing means. Ring means, high-efficiency coding means for compressing the information amount of the output signal of the shuffling means, and sync block packing means for packing the signal compressed by the high-efficiency coding means into a sync block for each predetermined information amount unit. And a recording means for recording the output signal from the sync block packing means on a magnetic tape.

According to a tenth aspect of the present invention, there is provided a digital video signal processing apparatus which performs signal processing for reproducing and outputting a 4: 1: 1 format or 4: 2: 2 format digital video signal as an output video signal. In a processing device, a reproducing means for reproducing a signal converted into a predetermined format and recorded on a magnetic tape, and a sync for reproducing a sync block with the sync block at the time of recording the signal reproduced by the reproducing means as a basic unit. Block reproduction means, high efficiency decoding means for highly efficient decoding of the output signal of the sync block reproduction means, deshuffling means for deshuffling the output signal of the high efficiency decoding means, and output signals from the deshuffling means both horizontally and vertically. 4: 1: 1 by performing interpolation processing
And an interpolation processing means for converting into a format or a 4: 2: 2 format.

According to the eleventh aspect of the present invention, the digital video signal processing device has a predetermined format of 8/3: 2 // which is recorded on the magnetic tape or reproduced from the magnetic tape.
Either the 3: 0 format or the 3: 1: 0 format is used.

[0028]

According to the first aspect of the present invention, the input digital video signal of the 4: 1: 1 format is thinned out to be converted into a predetermined format, which is shuffled, highly efficient coded, and sync block packed. The video signal recorded in a predetermined format is converted to a 4: 1: 1 format by performing sync block reproduction, high-efficiency decoding, deshuffling, and then interpolation processing. As a result, the compression rate can be made close to that of the conventional method, and 1
Since the number of macroblocks in a frame is an integer value in both horizontal and vertical directions, it is possible to extend the recording time while suppressing the image deterioration as much as possible and minimizing the increase in the circuit scale.

According to the second aspect of the present invention, since the frequency component in the oblique direction of the recorded video signal is suppressed by the first area filtering means, it is possible to perform a recording operation with higher image quality with the same code amount.

According to the third aspect of the present invention, since the frequency component in the oblique direction of the reproduced video signal is suppressed by the second area filtering means, image quality deterioration, particularly mosquito noise, which occurs at the time of high efficiency encoding is reduced. It becomes possible.

According to a fourth aspect of the present invention, the diagonal frequency component of the video signal recorded by the first area filter means is suppressed, and the diagonal frequency component of the video signal reproduced by the second area filter means is suppressed. Since the image is suppressed, it is possible to maintain the high image quality and reduce the reproduction distortion due to the synergistic effect.

According to the fifth aspect of the present invention, the area filtering means used at the time of recording and the area filtering means used at the time of reproduction are shared by a single area filtering means, so that the circuit scale can be reduced.

According to a sixth aspect of the present invention, the filter coefficient is set so that the first area filtering means performs the conversion into the predetermined format by the thinning processing of the recorded video signal and the attenuation of the frequency component in the oblique direction. By doing so, the thinning-out processing means and the first area filter means can be realized as one circuit block, and the circuit scale can be reduced.

According to a seventh aspect of the present invention, the output signal from the deshuffling means is interpolated by 4: 1: 1.
The second area filtering unit and the interpolation processing unit are realized as one circuit block by setting the filter coefficient so that the second area filtering unit performs the conversion into the format and the attenuation of the frequency component in the diagonal direction. It is possible to reduce the circuit scale.

In the eighth aspect of the present invention, the format of the input video signal or the output video signal is 4:
By adopting the 2: 2 format, it becomes possible to directly interface with the D1 standard signal.

According to the ninth aspect of the present invention, when the input video signal is recorded by using the predetermined format, the image deterioration is suppressed as much as possible, the circuit scale is prevented from increasing, and the recording time is long. Enables time conversion.

According to a tenth aspect of the present invention, when the video signal recorded on the magnetic tape is reproduced by using a predetermined format, the image deterioration is suppressed as much as possible.
It is possible to prevent the circuit scale from increasing and to lengthen the reproduction time.

In the eleventh aspect of the present invention, since either the 8/3: 2/3: 0 format or the 3: 1: 0 format is used as the predetermined format,
The compression rate is closest to that of the conventional method, and the number of macroblocks in one frame is an integer both horizontally and vertically.

[0039]

【Example】

Example 1. An embodiment of the present invention will be described below with reference to the drawings. In the figure, the same reference numerals as those of the conventional example indicate the same or corresponding portions as those of the conventional example. FIG. 1 is a diagram showing the configuration of a digital video signal processing apparatus according to the first embodiment of the present invention. In the figure, 1 is input 4: 1: 1
This is a digital video signal in a format, and as a result of being thinned out in the vertical and horizontal directions by the thinning-out processing means 10 which will be described later, it is converted into an 8/3: 2/3: 0 format.
Reference numeral 30 is a data shuffling means for shuffling the input data, and the high efficiency encoding means 40 in the subsequent stage performs all the processing for uniformly compressing the entire one frame.

High efficiency coding means 4 after data shuffling
At 0, the above-mentioned DCT processing and variable length coding processing are performed to compress the information amount. The compressed information is packed by sync block packing means 50 into information units called sync blocks for recording on the tape, and is recorded on magnetic tape 70 via recording amplifier 60.

At the time of reproduction, the reverse processing of the above-mentioned processing is performed to restore the video data again. First, the data read from the magnetic tape by the reproducing amplifier 160 is converted from the sync block information unit into compressed data by the sync block reproducing means 150. The video data compressed by the high-efficiency decoding means 140 is decompressed into data before compression, and then deshuffled by the data deshuffling means 130. Reference numeral 110 denotes an interpolation processing means, which results from the interpolation processing of the output signal from the deshuffling means 130, 4: 1:
It is converted to 1 format and the reproduced video signal 2 is output.

In this embodiment, the recording time is extended to 2
It is assumed that double time can be recorded. That is,
For example, it is considered that a tape that has been capable of recording for 1 hour in the past is used to enable double recording for 2 hours. In this case, considering the compatibility with the conventional format, the number of tracks that can be recorded on the tape must be the same as the conventional one, so the total number of tracks must be the same, and therefore the amount of recorded information per frame is halved. You will have to do it.

In this case, it is possible to simply double the image compression rate in the conventional format for recording, but image deterioration due to mosquito noise, block distortion, etc. is extremely impractical. Here, the mosquito noise block distortion will be briefly described. As described above, in order to perform image compression, DCT is performed to remove high-frequency components or roughly quantize, but when performing high compression such as twice the conventional compression amount, high-frequency components are Since the information is significantly compressed, a square pattern corresponding to the DCT block appears when the image is reconstructed, and in a severe case, a mosaiced image may result. This symptom is called block distortion. For the same reason, it appears as noise that is also smeared around the image contour (mosquito column-shaped mist noise), and this symptom is called mosquito noise. Both are harmful effects that appear when the image compression rate is large.

Therefore, in the present invention, first, the pixel structure of one frame is converted as shown in FIG. By the thinning-out process, the Y signal is converted into horizontal 480 pixels × vertical 480 pixels, and C
The r · Cb signal has 120 horizontal pixels × 240 vertical pixels. This pixel configuration in this embodiment is determined from the viewpoint that the image compression rate is as much as possible and the video signal actually broadcast can be faithfully reproduced. FIG. 3 shows the relationship between the pixel configuration and the image format / compression ratio / horizontal resolution. The horizontal resolution of broadcast television signals is about 340 for Y signals and about 80 for color signals. According to the figure, the 8/3: 2/3: 0 format has the closest compression rate to the conventional method, In addition, the video quality can be recorded while satisfying the current TV broadcast quality.

That is, if the pixel structure is further roughened (reduced), the resolution is lowered, the current TV broadcasting quality cannot be maintained, and the reproduced image distortion becomes large even if the compression ratio is increased, and the visual quality is improved in any case. It can be said that the image format according to the present invention is optimal because it cannot be practically used. The vertical resolution of the Cr / Cb signal is half that of the Y signal, but this has already been done in the SECAM system, which is the current television broadcasting centered in Europe, and the Cr / Cb signal is half the Y signal. It can be said that even if the vertical resolution of is not a problem in terms of visibility.

From the above, by using the pixel configuration according to the present invention, the compression rate can be increased by 10% as compared with the conventional example, and the time length can be doubled while the current broadcasting quality is sufficiently satisfied. The reason why the doubled time can be realized will be described with reference to FIG. In the conventional method, the image format 4: 1: 1 is compressed to ⅕ and recorded on the tape. Image format 8/3: 2
By setting / 3: 0, the total number of pixels is (8/3 + 2/3 +
0) / (4 + 1 + 1) = 10/18 times,
The total data amount is ½, which is half the conventional data amount. Therefore, it is possible to realize a time length twice as long as the conventional one.

Next, the process of converting from the conventional pixel configuration to the present pixel configuration, which is performed by the data thinning means 10 in FIG. 1, will be described. FIG. 4 is an internal block diagram of the data thinning means 10, and is a block diagram of a conversion process (downsampling process) performed on a pixel input in a conventional pixel configuration. In the thinning processing block 12 for the input video data, both the Y signal and the Cr / Cb signal are 2 / in the horizontal direction.
The thinning-out process is performed on No. 3, the thinning-out process is not performed on the Y signal in the vertical direction, and the thinning-out process is performed on 1/2 for the Cr / Cb signal.

Regarding the thinning processing, it is necessary to pre-attenuate the high frequency components through the low-pass filter 11 as a pre-processing in order to prevent the adverse effects of aliasing noise. In the figure, for the Y signal, the Cr / Cb signal is used. It is shown that low-pass filtering processing is performed for each of the above. As for the low-pass processing and the thinning-out processing, a so-called digital filter can be used in a digital circuit as an implementation method, and it is easy to realize in terms of technology and circuit scale.

The pixel-structured data is divided into macroblocks, compressed and stored in sync blocks. FIG. 5 shows an example of the configuration of macroblocks in this embodiment. In the configuration of this macroblock, 8 DCT blocks for the Y signal and 1 DCT block for the Cr / Cb signal are combined into 1 macroblock. With such a configuration, it is possible to hold Y, Cr, and Cb video information corresponding to the screen position in one frame of pixels.

FIG. 6 shows the structure of a macroblock on a one-frame screen. As shown in the figure, one frame is horizontal 1
5 macroblocks, vertical 30 macroblocks for a total of 45
It will consist of 0 macroblocks. As mentioned above, one frame is 5 for long time recording of 2 times.
Since it is sufficient to record on one track, the total number of sync blocks in one frame is 675 sync blocks.
It is sufficient to store 50 macroblocks. Therefore, it suffices to perform compression processing so that 2 macroblocks can be stored in 3 sync blocks, and this configuration is not a value that poses a particular problem with respect to error propagation during reproduction described above.

At the time of reproduction, it is possible to reconstruct the original video signal by performing the reverse process of recording. Pixel restoration from the sync block in DCT block units may be performed by simply performing the reverse conversion during recording. These processes themselves can be easily realized because they are not much different from the operations widely performed in the past. When the pixel configuration reproduced in this way is restored to the original video signal, the reverse process of the thinning process performed at the time of recording, that is, the interpolation process is required. This processing is performed by the data interpolating means 110 in FIG. 1, and its internal processing block is shown in FIG.

As shown in the figure, the interpolation processing block 1
11, the Y signal is 3/2 in the horizontal direction.
The interpolation process is performed twice, and the interpolation process is unnecessary in the vertical direction. The Cr / Cb signals are interpolated 3/2 times in the horizontal direction, and the interpolation processing is doubled in the vertical direction. The interpolation processing itself can be realized without any problem technically or by using a digital filter circuit as in the thinning processing.

Further, in the macro block structure, FIG.
As described above, since one frame has horizontal and vertical integer values, it is easy to control the data processing order, and there is an advantage that the circuit scale does not increase. This is clear from a comparison with the conventional macroblock configuration, for example.
As shown in, the value is not an integer in the horizontal direction, and special processing is required at the edge of the screen when the macroblock data is read and processed. Taking a specific example, considering the processing using a memory or the like, the address control circuit and the like become considerably complicated, and as a result, the circuit scale increases. On the other hand, in this embodiment, since the macroblock structure of one frame has an integer value in both horizontal and vertical directions, there is an advantage that the control becomes simple and the actual circuit can be made small.

In fact, of the formats shown in FIG.
In formats other than the 8/3: 2/3: 0 format according to the present invention, the macroblock structure of one frame does not have an integer value in the horizontal direction, and therefore, when these formats are used, they may be embodied. The problem occurs that the circuit scale increases. From the above, according to the present embodiment, by applying the above-mentioned recording format, the data recording format on the track, that is, the track pattern on the tape and the sync block format are kept compatible with the conventional one, and the image quality is deteriorated. It is possible to double the long-time recording with a minimum circuit scale without causing a significant increase in the compression rate.

Example 2. In the above-described first embodiment, an example is shown in which the long-time recording of double time is possible while maintaining the sufficient image quality while following the conventional tape recording format as much as possible. At that time, the pixel configuration was converted once from the conventional configuration and recorded, and then restored to the original pixel configuration during playback. At this time, the diagonal frequency components included in the video signal were dropped. If set, the image quality can be further improved. A block diagram of this embodiment is shown in FIG. In the figure, the same numbers as those of the first embodiment
Since the configurations are exactly the same, the description thereof will be omitted.

Area filters 20 and 1 in FIG.
Reference numeral 20 denotes a process of attenuating a signal of an oblique direction component in the input video data. Figure 9
FIG. 4 is a coefficient diagram of an area filter that attenuates a diagonal component in the present embodiment. The filtering process is performed as follows. As an example, when the coefficient of the example 1 shown in FIG. 9A is used, first, the vertical and horizontal 3 pixels centering on the pixel of interest, total 9 pixels, are multiplied by the coefficient in the same position as in FIG. The calculation is performed, the results are all added, and the result is 1/16. By performing this calculation sequentially for all pixels, it is possible to obtain a processed image in which the component in the oblique direction is attenuated, and the area filters 20 and 120 perform the above-mentioned processing on the input signal and output it.

Similarly, in the case of Example 2 shown in FIG. 9B, for each area of 5 horizontal pixels and 3 vertical pixels, each pixel value is multiplied by the corresponding coefficient and summed to 1/24. Good. In the case of Example 3 shown in FIG. 9C, each pixel value may be multiplied by the corresponding coefficient in the area of horizontal 3 pixels and vertical 3 pixels, and after adding up, it is reduced to 1/18.

At the time of recording, first, the thinning processing is performed on the input video signal as in the embodiment. After that, an area filter process is performed to remove diagonal components, and thereafter, the same process as in the above-described first embodiment is performed. At the time of reproduction, the same processing as that of the first embodiment is performed up to the data deshuffling processing, then the area filtering processing is performed, and the data interpolation processing is performed to obtain the output data.

The area filter used during recording attenuates the diagonal components, which are not so important to the visual sense, to secure the image quality in the horizontal and vertical directions, which is important to the visual sense at the time of high efficiency coding in the subsequent stage. The encoding can be performed. That is, even if the code amount after high-efficiency coding is the same, it can be visually perceived that the area filtering process is far better. This is an encoding method that makes use of human visual characteristics. The area filter performed at the time of reproduction can suppress noise components in an oblique direction such as mosquito noise generated by high-efficiency coding processing such as DCT processing, and can also obtain good visual characteristics.

In particular, since the mosquito noise tends to increase to some extent by increasing the compression rate, as in the present invention, though slightly, this area filter has a large visual noise suppression effect. Further, the area filter itself can be realized by a circuit similar to an ordinary digital filter as a method of constructing the area filter, and can be implemented without any particular problem in terms of actual circuit configuration or circuit scale. In the present embodiment, the area filter coefficient example shown in FIG. 9 has been described, but the present invention is not limited to this, and it is exactly the same as long as it shows the attenuation characteristic of the diagonal component even if another count value is used. It goes without saying that the effect of.

In the above description, it has been shown that by encoding the components in the oblique direction at the time of recording and at the time of reproducing, it is possible to perform encoding and decoding which are both visually pleasing. Of course, the same effect can be exhibited even if they are carried out individually, but when both are carried out at the same time, a better image can be produced by a synergistic effect.
That is, since the diagonal component at the time of reproduction when recording is performed by removing the diagonal component at the time of recording is obviously noise, the diagonal component removal at the time of reproduction is the noise removal operation itself. Further, although the oblique component removal operation has been described using the 8/3: 2/3: 0 format, the present invention is not limited to this, and other formats such as the 3: 1: 0 format are also described. Obviously has exactly the same effect.

Example 3. In the first embodiment, as the code amount control, the compression ratio is controlled so that 2 macroblocks fit within 3 sync blocks, but the present invention is not limited to this. For example, 4 macroblocks fit within 6 sync blocks. Needless to say, even if such control is performed, the same effect can be obtained.

Example 4. In the second embodiment, the area filter process is performed at the time of recording and at the time of reproduction, but it is obvious that the same effect can be obtained even if it is performed only at the time of recording or only at the time of reproduction. Further, by using area filters having different diagonal attenuation characteristics for recording and reproduction, respectively, the area filter 20 can be used for recording.
As an area filter characteristic that is particularly advantageous for high-efficiency encoding, and an area filter characteristic that is particularly effective for mosquito noise suppression as the characteristic of the area filter 120 during reproduction, a more visually pleasing image can be obtained. It will be possible.

Example 5. In addition, as shown in FIG.
It is also possible to share two area filters during recording and playback. In this case, the circuit scale of the filter circuit is larger than that of the input switching circuit. Is possible. The operation will be described. In the figure, at the time of recording, the input selection means 200 selects the thinning processing output and outputs it to the area filter means 300, and after the area filtering processing, outputs to the data shuffling means to perform the same operation as in the second embodiment. It is a thing.

At the time of reproduction, the input selection means 200 selects the data shuffling means output to select the area filter means 30.
The same operation as that in the second embodiment is performed by outputting the data to 0, and after performing the area filtering process, outputting to the data interpolating means. From the above, according to the present embodiment, it is possible to reduce the circuit scale and achieve the same effect as that of the second embodiment.

Example 6. In the second embodiment described above, the data thinning process and the diagonal component attenuation area filter can be realized as one area filter. This is because both processes are actually composed of digital filters as described above, and therefore it is easy to design an area filter having both characteristics. For example, both the area filter having the coefficient shown in Example 1 of FIG. 9A and the area filter having the coefficient shown in Example 3 of FIG. Instead, by changing the count value, it becomes possible for Example 3 to have a low-pass filter characteristic in the vertical direction as compared with Example 1.

That is, the area filter has both the attenuation characteristic of the oblique component and the removal characteristic of the high frequency component in the vertical direction, which shows that the low-pass filter processing in the vertical thinning can be performed at the same time. From such a fact, it is possible to realize the area filter which enables both the thinning processing and the diagonal component removal processing shown in the second embodiment, and in this case, the circuit scale is reduced and the embodiment is realized. It is possible to perform the same operation as 2.
It goes without saying that an area filter that simultaneously performs the diagonal component removal processing and the interpolation processing can be realized in the same manner.

Example 7. In the above embodiment, the input signal format or the output signal format is 4:
Although a series of description has been given using the 1: 1 format, the present invention is not limited to this, and the 4: 2: 2 format may be used, for example, and the same effect is obtained. The 4: 2: 2 format is the so-called D1.
Also called a format, it is a standard widely used in digital video equipment, and therefore 4: 2:
The use of the two format has an effect of facilitating connection with many devices.

The configuration of this embodiment can be configured by simply changing the thinning-out processing and the interpolation processing unit in the description of the first embodiment, and the operation contents and effects are the same as those of the first embodiment. It will be exactly the same. That is, Example 1
In the configuration of FIG. 4, the horizontal 2/3 decimation process of the Cb / Cr process in the decimation process in FIG. 4 can be applied to the input signal format 4: 2: 2 by changing it to 1/3 decimation process.

Further, in FIG. 7, the output signal format 4: 2: 2 can be applied only by changing the interpolation processing to triple interpolation. These changes are generally implemented by digital filters as described above, so it is only necessary to slightly change the configuration of the digital filters, and it goes without saying that realization is easy. Can be easily applied to all of the above-described embodiments, and has exactly the same effect.

[0071]

According to claim 1 of the present invention, 4: 1:
The input digital video signal of 1 format is thinned to be converted into the 8/3: 2/3: 0 format, shuffled, term efficiency coded, sync block packed, and recorded on the magnetic tape. During playback, the video signal recorded in the 8/3: 2/3: 0 format is converted to the 4: 1: 1 format by performing sync block playback, high-efficiency decoding, deshuffling, and then performing interpolation processing. , The compression rate can be made close to that of the conventional method, and the number of macroblocks in one frame is an integer both horizontally and vertically, so image deterioration is suppressed as much as possible, and increase in circuit scale is minimized. In this state, there is an effect of obtaining a digital video signal processing device capable of lengthening the recording time.

According to the second aspect of the present invention, since the frequency component in the oblique direction of the recorded video signal is suppressed by the first area filtering means, it is possible to perform a recording operation with higher image quality with the same code amount. There is.

According to the third aspect of the present invention, since the frequency component in the diagonal direction of the reproduced video signal is suppressed by the second area filtering means, image quality deterioration, particularly mosquito noise, which occurs at the time of high efficiency encoding is reduced. The effect is that it is possible to do.

According to the fourth aspect of the present invention, the diagonal frequency component of the video signal recorded by the first area filter means is suppressed, and the diagonal frequency component of the video signal reproduced by the second area filter means is suppressed. Is suppressed, there is an effect that it is possible to maintain high image quality and reduce reproduction distortion by a synergistic effect.

According to the fifth aspect of the present invention, since the area filtering means used at the time of recording and the area filtering means used at the time of reproduction are commonly used by the area filtering means, it is possible to reduce the circuit scale. There is.

According to the sixth aspect of the present invention, the conversion of the recorded video signal into the 8/3: 2/3: 0 format and the attenuation of the frequency component in the oblique direction are performed first.
By setting the filter coefficient as is done by the area filtering means, the thinning processing means and the first area filtering means can be realized as one circuit block, and there is an effect that the circuit scale can be reduced. .

According to the seventh aspect of the present invention, the second area filtering means performs the conversion of the output signal from the deshuffling means into the 4: 1: 1 format by the interpolation processing and the attenuation of the frequency components in the oblique direction. By setting the filter coefficient in this way, the second area filter means and the interpolation processing means can be realized as one circuit block, and the circuit scale can be reduced.

According to the eighth aspect of the present invention, the format of the input video signal or the output video signal is 4: 2 :.
The two format has an effect that it is possible to directly interface with the D1 standard signal.

According to the ninth aspect of the present invention, when the input video signal is recorded by using the predetermined format, the image deterioration is suppressed as much as possible, the increase of the circuit scale is prevented, and the recording time is shortened. The effect is that it can be extended.

According to the tenth aspect of the present invention, when the video signal recorded on the magnetic tape is reproduced by using the predetermined format, the video deterioration is suppressed as much as possible, and the circuit scale is prevented from increasing. Moreover, there is an effect that the reproduction time can be lengthened.

According to the eleventh aspect of the present invention, either the 8/3: 2/3: 0 format or the 3: 1: 0 format is used as the predetermined format to be recorded on or reproduced from the magnetic tape. Since the compression rate can be made as close as possible to the conventional method, and the number of macroblocks in one frame is an integer value in both horizontal and vertical directions, the deterioration of video is suppressed as much as possible and the increase in circuit scale is minimized. There is an effect that the recording time or the reproducing time can be lengthened in a limited state.

[Brief description of drawings]

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

FIG. 2 is a diagram showing a pixel configuration of one frame in the first embodiment of the present invention.

FIG. 3 is a diagram for explaining the first embodiment of the present invention.

FIG. 4 is a diagram for explaining the operation of the first exemplary embodiment of the present invention.

FIG. 5 is a diagram for explaining the operation of the first exemplary embodiment of the present invention.

FIG. 6 is a diagram for explaining the operation of the first exemplary embodiment of the present invention.

FIG. 7 is a diagram for explaining the operation of the first exemplary embodiment of the present invention.

FIG. 8 is a diagram showing a configuration of a digital video signal processing device according to a second embodiment of the present invention.

FIG. 9 is a diagram for explaining the operation of the second exemplary embodiment of the present invention.

FIG. 10 is a diagram showing a configuration of a digital video signal processing device according to a fifth embodiment of the present invention.

FIG. 11 is a diagram showing a configuration of a conventional digital video signal processing device.

FIG. 12 is a pixel configuration diagram of one frame for explaining an operation of a conventional example.

FIG. 13 is a block diagram of a macro block for explaining the operation of a conventional example.

FIG. 14 is an explanatory diagram of macroblocks for explaining the operation of the conventional example.

FIG. 15 is an explanatory diagram of macroblocks for explaining the operation of the conventional example.

FIG. 16 is an explanatory diagram of a sync block for explaining the operation of the conventional example.

FIG. 17 is an explanatory diagram of a sync block for explaining the operation of the conventional example.

FIG. 18 is an explanatory diagram of recording tracks for explaining the operation of the conventional example.

[Explanation of symbols]

1 input data, 2 output data, 10 data thinning processing means 20 area filter means, 30 data shuffling means 40 high efficiency coding means, 50 sync block packing means 110 data interpolation processing means, 120 area filter means 130 data deshuffing Ring means, 140 High-efficiency decoding means 150 Sync block reproducing means, 200 Input selecting means 300 Area filter means

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Tomohiro Ueda No. 1 Baba Institute, Nagaokakyo-shi Video System Development Laboratory, Mitsubishi Electric Corporation

Claims (11)

[Claims] 1. A digital video signal processing apparatus for recording a digital video signal of 4: 1: 1 format as an input video signal on a magnetic tape and performing signal processing for reproducing the recorded digital video signal, Thinning processing means for converting the input digital video signal of the 4: 1: 1 format into a predetermined format by performing thinning processing in both the horizontal and vertical directions, and shuffling the output signal of the thinning processing means. Shuffling means for performing, high-efficiency coding means for compressing the information amount of the output signal of the shuffling means, and the signal compressed by the high-efficiency coding means is packed in a sync block for each predetermined information amount unit The sync block packing means and the output signal from the sync block packing means are magnetically coupled. Recording means for recording on a magnetic tape, a reproducing means for reproducing a signal recorded on the magnetic tape, and a sync block for reproducing the sync block with the sync block at the time of recording the signal reproduced by the reproducing means as a basic unit. Reproducing means, high efficiency decoding means for highly efficiently decoding the output signal of the sync block reproducing means, deshuffling means for deshuffling the output signal of the high efficiency decoding means, and horizontal and horizontal output signals from the deshuffling means. An apparatus for digital video signal processing, comprising: interpolation processing means for converting into a 4: 1: 1 format by performing interpolation processing in both vertical and vertical directions. 2. A first area filtering means for attenuating a diagonal frequency component of a signal is provided between the thinning processing means and the shuffling means.
The described digital video signal processing device. 3. A second area filtering means for attenuating a diagonal frequency component of a signal is provided between the deshuffling means and the interpolation processing means.
The described digital video signal processing device. 4. A first area filtering means for attenuating a diagonal frequency component of a signal is provided between the thinning processing means and the shuffling means, and a diagonal area of the signal is provided between the deshuffling means and the interpolation processing means. 2. The digital video signal processing device according to claim 1, further comprising second area filtering means for attenuating frequency components. 5. A signal selecting means for selecting and outputting the output signal of the thinning processing means during recording and selecting and outputting the output signal of the deshuffling means during reproduction, and an oblique direction with respect to the signal from the signal selecting means. And a single area filtering means for attenuating the frequency component of the input signal, the output signal of the single area filtering means is input to the shuffling means during recording, and the output signal of the single area filtering means is interpolated during reproduction. The digital video signal processing device according to claim 1, wherein the digital video signal processing device is configured to be input to the processing means. 6. The first area filtering means converts the filter coefficient into a predetermined format by thinning out at least one of the vertical direction and the horizontal direction and sets the filter coefficient so as to attenuate the frequency components in the oblique direction. The digital video signal processing device according to claim 2, wherein the digital video signal processing device is set. 7. The second area filtering means converts the output signal from the deshuffling means into a 4: 1: 1 format by performing interpolation processing in at least one of the vertical direction and the horizontal direction, and the diagonal direction. 4. The digital video signal processing device according to claim 3, wherein the filter coefficient is set so as to attenuate the frequency component. 8. A 4: 2: 2 format is used as a format of a video signal input to the thinning processing means during recording and a video signal output from the interpolation processing means during reproduction. The digital video signal processing device according to any one of 1. 9. A digital video signal processing device for performing signal processing for recording a 4: 1: 1 format or 4: 2: 2 format digital video signal as an input video signal on a magnetic tape, The 4: 1: 1 format or 4: 2:
Decimation processing means for converting a digital video signal of two formats into a predetermined format by performing decimation processing in both the horizontal and vertical directions, a shuffling means for shuffling the output signal of the decimation processing means, and the shuffling. High-efficiency coding means for compressing the information amount of the output signal of the ring means; sync-block packing means for packing the signal compressed by the high-efficiency coding means into sync blocks in units of predetermined information amount; A digital video signal processing device, comprising: a recording means for recording an output signal from the block packing means on a magnetic tape. 10. A digital video signal processing device for performing signal processing for reproducing and outputting a 4: 1: 1 format or 4: 2: 2 format digital video signal as an output video signal, the conversion processing being performed in a predetermined format. Reproducing means for reproducing the signal recorded on the magnetic tape; sync block reproducing means for reproducing the sync block when the signal reproduced by the reproducing means is used as a basic unit for recording the sync block; High efficiency decoding means for highly efficient decoding of the output signal of the means, deshuffling means for deshuffling the output signal of the high efficiency decoding means, and by performing both horizontal and vertical interpolation processing on the output signal from the deshuffling means. Converts to 4: 1: 1 format or 4: 2: 2 format Digital video signal processing apparatus characterized by comprising an interpolation processing unit. 11. The predetermined format is 8/3: 2 /
The format according to any one of claims 1 to 6, wherein the format is either a 3: 0 format or a 3: 1: 0 format.
The digital video signal processing device according to any one of 1.