JP2000068854A - Data processing unit and its method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent deterioration in data in the case of sampling data with a sampling frequency different from a sampling frequency of input data. SOLUTION: A band limit discrimination circuit 10 discriminates whether or not each sample point before and after received data has already been band- limit-processed. In the case that each sample point before and after received data has already been band-limit-processed, a switching circuit SW2 outputs data S8 resulting from applying no band limit processing to the received data but only thinning data at an interpolation point. Furthermore, in the case that any of the sample points before and after the received data has not yet been band-limit-processed, the switching circuit SW2 outputs data S7 that are frequency-band limit processed via a delay circuit 4, a filter circuit 5, a cross product arithmetic circuit 6 and a 1/2 arithmetic circuit 7.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data processing apparatus and method for performing sampling at a sampling frequency different from the sampling frequency of input data.

[0002]

2. Description of the Related Art In recent years, with the development of digital technology, exchange of encoded data in various fields has been increasing. In this case, the transmission and recording of the encoded data may involve converting the encoded data to a different sampling standard and exchanging the data.

Here, in the field of digital television, three primary color signals of red (R), green (G) and blue (B) are converted into a luminance signal Y and two color difference signals C _R (RY), C _B (B
−Y) are used as the baseband. The sampling standard for sampling this digital component signal is defined in ITU-R
(International Telecommunication Union Radio Communications Division (formerly CCIR (International Radio Communications Advisory Committee))). In this sampling standard, for example, the ratio of the sampling frequency of each signal of the luminance signal and the color difference signal is determined. Regarding the transmission standard of digital component signals, basically, the ratio of the sampling frequency of each component signal is 4: 2: 2.
(For the sampling frequency of the luminance signal,
(A method of performing sampling by halving the sampling frequency of each color difference signal). In this method, 13.5 MHz is used as the sampling frequency of the luminance signal, and 6.75 MHz is used as the sampling frequency of each color difference signal. At this time,
The signal band of each signal is a value obtained by halving the sampling frequency according to Shannon's sampling theorem, the signal band of the luminance signal is 6.75 MHz, and the signal band of the color difference signal is 3.375 M
Hz. It should be noted that the ITU-R also defines a method in which the sampling frequency ratio of each component signal is 4: 4: 4. In this case, the sampling frequency of all signals is 13.5 MHz. It is.

On the other hand, a digital video cassette (hereinafter, referred to as a digital video cassette) for recording encoded data such as digital component signals.
DVC) and the like, a sampling standard different from the 4: 2: 2 method described above is defined. In the DVC, sampling of the 4: 2: 2 component signal is usually performed by further reducing the band of the color difference signal to 1/2. At this time, the method of limiting the band in the horizontal direction of the screen is the 4: 1: 1 method, and the method of limiting the band in the vertical direction is the 4: 2: 0 method. In this case, for example, in the 4: 1: 1 system, 13.5 MHz is used as the sampling frequency of the luminance signal, and 3.375 MHz is used as the sampling frequency of the color difference signal. At this time, the signal band of the luminance signal is 6.75 MHz, and the signal band of the color difference signal is 1.6875 MHz.

[0005] As described above, the sampling standard (4: 2: 2 system) in the transmission of digital component signals, etc.
Since the standard is different from the sampling standard in VC and the like, when recording the component signal transmitted by the interface in the DVC, it is usually 4: 2: 2.
It is necessary to perform the format conversion of the sampling standard from the system to the 4: 1: 1 system (or 4: 2: 0 system) (hereinafter, such format conversion of the sampling standard is referred to as sample conversion). . Where 4:
2: 2 scheme 4: 1: 1 in the case of performing the sample conversion to scheme the sampling frequency of the color difference signals of 4: 2: 2 scheme half the sampling frequency F _s of the (F _s / 2) Sampling is performed. At this time, in order to avoid aliasing noise caused by lowering the sampling frequency, for example, a band that limits a signal in a frequency band that is half the sampling frequency F _s / 2 for an input signal. Restriction processing is performed. By performing the band limiting process in this way, the effect of aliasing noise is reduced. By performing the sample conversion in this manner, for example, in the data of the 4: 2: 2 system,
Assuming that there are M sample points (sampled points) along with the sampling, the conversion to the 4: 1: 1 method results in M / 2
The sample points are thinned out, and the remaining M / 2 sample points are recorded in a DVC or the like as storage points after sampling (hereinafter referred to as sample storage points).

When a digital component signal recorded on a DVC or the like in the 4: 1: 1 system is reproduced,
It is necessary to perform sample conversion from the 4: 1: 1 system to the 4: 2: 2 system. In this case, once it bandlimited the sampling frequency F _s / 2 of the color difference signals of 4: 2: performing sampling is increased up to two systems of the sampling frequency F _s. At this time,
In order to interpolate the sample points thinned out by the conversion to the 4: 1: 1 method described above, an interpolation process of appropriately inserting sample interpolation points is performed. In this interpolation processing, the sample interpolation points are obtained by performing predetermined arithmetic processing on the stored points recorded by the 4: 1: 1 method.

[0007]

As described above, in the prior art, the 4: 2: 2 system is changed to the 4: 1: 1 system (or the 4: 2: 2 system).
In the sample conversion to the (2: 0 method), a frequency band limiting process for attenuating the high frequency component of the color difference signal is performed.
This band limiting process is effective in reducing aliasing noise if the input data is original data that has not been subjected to the band limiting process.For example, the input data is subjected to the band limiting process once. If the data is the same, the data will deteriorate due to the characteristics of the band limiting filter. This is because the band limiting filter can only have a finite number of taps. As a result, in the related art, the 4: 2: 2 method and the 4: 1: 1 method (or 4: 2: 2 method) have been conventionally used.
If the sample conversion between (2: 0 method) and duplication processing (data duplication processing) is repeated, there is a problem that the color difference signal is deteriorated by the band limitation processing. Less than,
The data deterioration accompanying the sample conversion will be described more specifically.

FIG. 17 is an explanatory diagram showing a change in encoded data when a plurality of sample conversions are performed.
In the following description, blue is representatively described as a color difference signal, but the same applies to red. In the sample conversion shown in this figure, an example is shown in which a three-tap filter according to the following functional equation (1) is used as a band limiting filter. Equation (1) is a digital filter having three taps and filter coefficients of 1/4, 2/4, and 1/4. In the equation (1), B indicates an original blue color difference signal sampled by, for example, the 4: 2: 2 method, and the subscript indicates the position of a sample point of the color difference signal B. I have. N in the suffix of the color difference signal B is
Indicates an arbitrary integer. In equation (1), B ′ _2N is a blue color difference signal corresponding to a sample storage point recorded after performing a 4: 1: 1 sample conversion on the original color difference signal B, for example. Is shown.

B ′ _2N = 1/4 · (B _2N−1 + 2B _2N + B _{2N + 1} ) ‥‥‥ (1)

Further, in the sample conversion shown in FIG. 17, an example is shown in which interpolation processing using the following function expression (2) of a 3-tap filter is performed. In the equation (2), B ′ _{2N + 1} is, for example, from 4: 1: 1 scheme to 4: 2:
A blue color difference signal corresponding to an interpolation point to be interpolated when resampling is performed in two systems is shown.

B ′ _{2N + 1} = 1/2 · (B ′ _2N + B ′ _{2N + 2} ) = 1/8 · (B _2N−1 + 2B _2N + 2B _{2N + 1} + 2B _{2N + 2} + B _{2N + 3} )} ‥ (2)

In FIG. 17, data D ₁₀₀ is 4:
The data of the original blue color difference signals ( _B2N-2 , _B2N-1 , _B2N , _{B2N + 1} , _{B2N + 2} , ...) sampled by the 2: 2 method are shown. The data _D101 is the data D101.
_The data obtained by performing sample conversion for ₁₀₀ is shown. In this data D _101, the data D _S11 is accompanied by a band limiting process using the equation (1) with respect to the original data D ₁₀₀ 4: 1: 1 type sample storage points are recorded after the sample conversion was carried out in , The data of the blue color difference signals (B ′ _2N−2 , B ′ _2N , B ′ _{2N + 2} ,...). The stored color difference signals are the even-numbered color difference signals (B _2N-2 , B ₂₎ in the original data D ₁₀₀ .
_2N , _{B2N + 2,.} . . ). That is, 4 in this example: 1: 1 in the conversion to scheme, the odd-numbered signal in the original data D ₁₀₀ (B _2N-1,
B _{2N + 1,.} . . ) Are skipped without being stored. In the data D _101, data D _i11
Is a blue color difference signal (corresponding to a sample interpolation point interpolated using equation (2) for data _DS11 when resampling is performed from the 4: 1: 1 scheme to the 4: 2: 2 scheme. B ′ _2N−1 , B ′ _{2N + 1,.} Color difference signal interpolated here is a signal corresponding to the odd-numbered color difference signals in the original data _{D 100 (B 2N-1,} B 2N + 1, ...). That is, it corresponds to a signal which is not stored and is decimated in the conversion to the 4: 1: 1 system.

In FIG. 17, data D ₁₀₂ is
For the data D ₁₀₁ subjected to sample conversion accompanying interpolation and bandwidth limitation process as described above, shows the data subjected to sample conversion accompanying once more band limitation processing and the interpolation processing. In this data _D102 , data D
_S12, accompanied by band limiting process using the equation (1) with respect to data D ₁₀₁ 4: 1: 1 type sample converts the color difference signals and blue corresponding to the sample storage points recorded after the (B '' _2N-2 , B'' _2N , B'' _{2N + 2} , ...). The stored color difference signals correspond to even-numbered color difference signals (B ′ _2N−2 , B ′ _2N , B ′ _{2N + 2} ,...) In the data D ₁₀₁ that has been sample-converted once. That is, the data D _S12 corresponds to the data D _S11 as sample storage point in the data D _101. In the data D _102, the data D _i12 is 4: 1:
1 Method 4: 2: 2 color difference signals (B 'blue interpolated using equation (2) with respect to data D _S12 when performing scheme _{resampling' 2N-1, B ''} 2N + ₁ , ...). Here the color difference signal interpolated in the odd-numbered color difference signals in the data D ₁₀₁ was performed once sample conversion (B '
_2N-1 , _{B'2N + 1,.} . . ). That is, data D _i11 as a sample interpolation point in data D ₁₀₁
Corresponding to

As shown in FIG. 1, when a three-tap filter is used, a plurality of sample conversions corresponding to the dubbing process are repeated, so that the values of the sample storage point and the interpolation point are changed by the characteristics of the filter. As it changes, the number of taps and coefficients of the filter function also change. Such characteristics occur as long as a filter having a finite number of taps is used.

FIG. 18 shows changes in the filter coefficient and the number of taps when sample conversion is repeated using the above-described three-tap filter. In this figure, the vertical axis shows the coefficient value of the filter, and the horizontal axis shows the tap points on the time axis. The center of the tap point is the pole of the filter function. Further, in the figure, those indicated at F ₁ is, for example, corresponding to the data D ₁₀₁ subjected to sample conversion to the original data. Further, for example, that shown by reference numeral F ₂ in figure, once for the samples converted been conducted data D _101, corresponding to the data D _102, further subjected to sample conversion. As can be seen from this figure, the number of taps of the filter function and the filter coefficient change by repeating the sample conversion. In this case, the value of the filter coefficient changes so that the distribution of the tap points becomes wider.

FIG. 19 is an explanatory diagram showing more specifically the data deterioration caused by the sample conversion described above. The example of this figure shows a state of signal degradation when a pulse signal (B _N ) having an amplitude of A ₀ is specifically input as a color difference signal. In this figure, (A) shows the original signal, corresponding to the original data D ₁₀₀ in FIG. 17.
17B shows a signal obtained by performing one sample conversion, that is, a band limiting process on the pulse signal of FIG. a signal corresponding to D _101. (C) shows a signal obtained by further performing sample conversion on the pulse signal of (B), that is, performing band limiting processing on the pulse signal of (B), and then performing interpolation processing and re-sampling. and a signal corresponding to the data D ₁₀₂ in FIG. 17. From this figure, the amplitude A ₀
It can be understood that the pulse signal is deteriorated by repeating the sample conversion. As described above, in the related art, after performing band limiting processing on input data, performing interpolation processing, and repeating sample conversion for re-sampling, data deteriorates.

The present invention has been made in view of the above problems, and has as its object to provide a data processing method capable of preventing data deterioration when sampling is performed at a sampling frequency different from the sampling frequency of input data. It is to provide an apparatus and a method.

[0018]

SUMMARY OF THE INVENTION A data processing apparatus according to the present invention determines whether or not input data subjected to sampling processing has been subjected to band limiting processing in a band lower than a predetermined frequency band. The input data according to the discrimination result of the discrimination means, performing a frequency band limiting process as needed, and performing sampling at a sampling frequency different from the sampling frequency of the input data. And a sampling means.

According to the data processing method of the present invention, it is determined whether or not the sampled input data is data subjected to band limiting processing in a band lower than a predetermined frequency band. Accordingly, the input data is subjected to frequency band limiting processing as necessary, and is sampled at a sampling frequency different from the sampling frequency of the input data.

In the data processing apparatus and method according to the present invention, it is determined whether or not the sampled input data is data subjected to band limiting processing in a band lower than a predetermined frequency band. Depending on the result, the input data is subjected to frequency band limiting processing as necessary, and is sampled at a sampling frequency different from the sampling frequency of the input data.

[0021]

Embodiments of the present invention will be described below in detail with reference to the drawings.

[First Embodiment] A data processing apparatus according to the present embodiment performs, for example, band limiting processing or the like on a digital component signal sampled in a 4: 2: 2 system, and performs DVC. Is a recording format such as 4:
In addition to the conversion into the signal of the 1: 1 system, the signal converted into the 4: 1: 1 system is subjected to interpolation processing and the like for re-sampling in order to be recorded on the DVC or the like, and the 4: 2: 2 system is again performed. To convert the signal into However, the present invention is not limited to the sample conversion between the 4: 2: 2 system and the 4: 1: 1 system.
It can be used for sample conversion between other methods.
In the following description, the sample conversion of the blue color difference signal among the digital component signals will be described, but the same applies to the red color difference signal. In addition,
In the 4: 2: 2 method and the 4: 1: 1 method, the sampling standard is the same for the luminance signal, and the conversion of the sampling frequency is not performed.

FIGS. 1 and 2 are block diagrams showing the configuration of a data processing device according to the first embodiment of the present invention. Here, FIG. 1 shows a configuration of a circuit portion that performs sample conversion accompanied by band limiting processing in the data processing device according to the present embodiment. FIG. 2 shows a configuration of a circuit portion that performs resampling processing accompanied by interpolation processing in the data processing device according to the present embodiment.

As shown in FIG. 1, the data processing apparatus according to the present embodiment includes, for example, a blue color difference signal (B _2N , B B) sampled at a sampling frequency F _S by the 4: 2: 2 method.
_{2N + 1} , _{B2N + 2} , _{B2N + 3,.} . . An image edge processing circuit 1 which encoded data is input), the horizontal and vertical directions (hereinafter, the H / V synchronizing signals S ₁ for synchronizing as H / V.) Is input, the image H for outputting the image edge detection signal S ₃ for detecting an image edge to edge processing circuit 1
And / V sync circuit 2, together with the H / V parameter signal S ₂ related to the shift amount of H / V synchronizing signals S ₁ and H / V is input, stored and interpolation point switch for switching the samples stored points and interpolation points A storage / interpolation point switching circuit 3 that outputs signals S ₄ and S ₅ , a delay circuit 4 to which an output signal from the image edge processing circuit 1 is input, and a filter circuit 5 are provided.

The data processing device according to the present embodiment also includes a product-sum operation circuit 6 to which a signal delayed by the delay circuit 4 and an output signal from the filter circuit 5 are input.
A 1/2 operation circuit 7 to which an output signal from the product-sum operation circuit 6 is input, a D flip-flop circuit 8 to which an output signal from the 1/2 operation circuit 7 is input, and a signal delayed by the delay circuit 4 D flip-flop circuit 9 to which is input
And the signal delayed by the delay circuit 4 and the output signal from the filter circuit 5 are input, and the switching signal S
₆ and output signals from the D flip-flop circuit 8 and the D flip-flop circuit 9, and are input based on the switching signal S ₆ from the band limit determination circuit 10. One of the signals
And a switching circuit SW2 for selectively outputting one of them, and a D for receiving a signal output from the switching circuit SW2.
Flip-flop circuit 18 and switching circuit SW2
And D flip-flop with the output signal from the circuit 18 is input, the switching circuit SW3 for selectively outputting on the basis of the switching signal S ₅ of one of the signal input from the store and the interpolation point switching circuit 3 And

Here, the circuit portion shown in FIG. 1 corresponds to "sampling means" in the present invention. Further, the band limitation determining circuit 10 corresponds to “determining means” in the present invention.

The image edge processing circuit 1 receives a coded data, outputs a plurality of signals obtained by delaying the input data at a plurality of timings, and outputs a plurality of signals from the delay circuit 20. When a signal is input and H
/ V image edge detection signal from the synchronization circuit 2 S ₃ and stored and
Based on the storage and interpolation point switching signal S ₄ from the interpolation point switching circuit 3, one of the signal inputted from the delay circuit 20 1
And a switching circuit SW1 for selectively outputting one of them. The delay circuit 20 includes the D flip-flop circuit 2
1, a D flip-flop circuit 22, a D flip-flop circuit 23, and a D flip-flop circuit 24, thereby outputting a plurality of delayed signals.

The band limit discriminating circuit 10 includes a rear interpolation point comparing circuit 11 to which an output signal from the delay circuit 4 and an output signal from the filter circuit 5 are input, and a signal from the output signal from the delay circuit 4 and the filter circuit 5. , A D flip-flop circuit 13 to which an output signal from the D flip-flop circuit 12 is input, and an output signal from the delay circuit 4 through the D flip-flop circuit 9. The output signal from the delay circuit 4 is input via the D flip-flop circuit 14 and the D flip-flop circuit 9 and the D flip-flop circuit 14 to be input, and the D flip-flop circuit 12 and the D flip-flop circuit 13 Filter circuit 5
And an AND circuit 16 to which an output signal is input. AND circuit 16, and outputs a switching signal S ₆ to the switching circuit SW2.

In the circuit having the above configuration, for example, 4:
The color difference signals B _2N , B _{2N + 1} , B _{2N + 2} , B _{2N + 3} , _{... Of} the sampling frequency F _S sampled by the 2: 2 method. . . Is subjected to band limiting processing as necessary, and the color difference signals B ′ _2N , B ′ _2N , B ′ _{2N + 2} , B ′ _{2N + 2,.} . . Is output, and the data of the sampling frequency F _S / 2 in the 4: 1: 1 system is output.

The input color difference signals B _2N , B _{2N + 1} ,
B _{2N + 2} , B _{2N + 3,.} . . In the data of the above, which data is the data corresponding to the sample storage point and the sample interpolation point is determined by the sampling standard of the 4: 1: 1 method and the storage / interpolation point before the actual sampling process. is determined on the basis of the the H / V parameter signal S ₂ input in the switching circuit 3. Further, the input color difference signals B _2N ,
_{B2N + 1} , _{B2N + 2} , _{B2N + 3,.} . . In the above data, the data serving as the sample interpolation point corresponds to “first data” in the present invention, and the data serving as the sample storage point corresponds to “second data” in the present invention.

As shown in FIG. 2, the data processing apparatus according to the present embodiment includes, for example, a sampling frequency F _S by a 4: 1: 1 method as a circuit element for performing re-sampling processing accompanied by interpolation processing. / 2 color difference signal (B _2N , B
_{2N + 2,.} . . With encoded data is input), the input data first-in-first-out (hereinafter, FIFO called.) And FIFO circuit 31 to be output in a manner, H / V sync signal S ₁₁ for synchronizing H / V input together are an H / V sync circuit 32 for outputting an image edge detection signal S ₁₃ for detecting an image edge, H / V synchronizing signals S ₁₁ and H / V relates to a shift amount of H / V parameter signal S ₁₂ Are input, and a storage / interpolation point switching circuit 33 that outputs storage / interpolation point switching signals S ₁₄ and S ₁₅ for switching between a sample storage point and an interpolation point, and an output signal from the FIFO circuit 31 are input. together, the image edge processing circuit 34 to store and interpolation point switching signal S ₁₄ from the image edge detection signal S ₁₃ and stored and the interpolation point switching circuit 33 from the H / V sync circuit 32 is inputted, image edge processing circuit 34 From A delay circuit 35 and the filter circuit 36 the output signal is input, the output signal from the delay circuit 35 and the filter circuit 36 is input, based on the switching signal S ₁₅ from the storage and interpolation point switching circuit 33, A switching circuit SW5 for selectively outputting any one of the input signals.

The image edge processing circuit 34 includes a FIFO circuit 31
And a delay circuit 40 that outputs a plurality of signals obtained by delaying the input data at a plurality of timings, and a plurality of output signals from the delay circuit 40, based on the storage and interpolation point switching signal S ₁₄ from the image edge detection signal S ₁₃ and stored and the interpolation point switching circuit 33 from the H / V sync circuit 3, one of the signal input from the delay circuit 40 And a switching circuit SW4 for selectively outputting. Delay circuit 40
Includes a D flip-flop circuit 41, a D flip-flop circuit 42, a D flip-flop circuit 43, and a D flip-flop circuit 44, thereby outputting a plurality of delayed signals.

In the circuit having the above configuration, for example, 4:
The color difference signals B _2N , B _{2N + 2} , _{... Of} the sampling frequency F _S / 2 sampled in the 1: 1 system. . . Interpolation processing is performed on the data of the color difference signals B _2N , B ′ _{2N + 1} , B
_{2N + 2,.} . . Is output and the sampling frequency F _S 4: 2:
Two types of data are output.

Next, the band limiting process and the interpolation process performed in the data processing apparatus having the above configuration will be described more specifically.

Conventionally, as described above, the signal value at the sample storage point changes every time a sample conversion process involving band limitation such as dubbing is performed using, for example, a three-tap filter represented by equation (1). It began to. Here, when the first sample conversion process is performed in a case where the input data is the original data that has not been subjected to the band limiting process, the band limiting process of the frequency with respect to the data at the sample storage point is performed by the sampling theorem. is necessary. However, after the second sample conversion process, the data at the sample storage point has already been subjected to the band limiting process, and thus it is not necessary to perform the band limiting process. In the related art, after the second sample conversion process, the same band limiting process is performed on the data at the sample storage points as in the first sample conversion, so that the signal value of the data at the sample storage points deteriorates. Therefore, in the present embodiment, the second and subsequent band limiting processes are performed by only thinning out the data of the sample interpolation points generated by the first sample conversion process so as to prevent data deterioration. I have. That is, in the present embodiment, the value of the sample storage point recorded after the second and subsequent band limiting processing is:
It is set to be the same as the sample storage point recorded in the first sample conversion process.

FIG. 3 is an explanatory diagram showing a change in encoded data when a plurality of sample conversions are performed in the data processing apparatus of the present embodiment. In this figure,
The data D _ORG is, for example, an original blue color difference signal (B _2N-2 , B _2N-1 , B _2N , B _2N) sampled by the 4: 2: 2 method.
_{2N + 1} , _{B2N + 2,.} . . ) Shows the data. Also,
Data D _1ST indicates data obtained by _subjecting the data D _ORG to sample conversion. In the data D _1ST , the data D _S1 is obtained by _subjecting the original data D _ORG to a band limiting process using Expression (1), for example, 4: 1:
Blue color difference signals (B ′ _2N-2 , B ′ _2N , B ′ _2N-2 , B ′ _2N) corresponding to the sample storage points recorded after performing the one-system sample conversion.
B ' _{2N + 2,.} . . ) Shows the data. The stored color difference signals are even-numbered color difference signals (B _2N-2 , B _2N , B _{2N + 2} ,...) In the original data D _ORG .
Corresponding to That is, in the conversion to the 4: 1: 1 system in this example, signals corresponding to the odd-numbered signals (B _2N−1 , B _{2N + 1} ,...) In the original data D _ORG are not stored. It has been thinned out. In the data D _1ST , the data D _i1 is changed from the 4: 1: 1 format to the 4: 2: 2 format.
The color difference signal (B blue corresponding to the sample interpolation points to be interpolated using equation (3) below with respect to data D _S1 when performing resampling in scheme _{'2N-1, B' 2N} + 1, ..). Here the color difference signal interpolated in the odd-numbered color difference signals in the original data D _ORG (B _2N-1,
B _{2N + 1,.} . . ). That is,
In the conversion to the 4: 1: 1 system, this corresponds to a signal that has been thinned out without being stored.

B ′ _{2N + 1} = 1/2 · (B ′ _2N + B ′ _{2N + 2} ) ‥‥‥ (3)

In FIG. 3, data D _2ND is data obtained by performing another sample conversion on the data D _1ST obtained by performing the sample conversion including the band limiting process and the interpolation process as described above. I have. This data D
_{In 2ND} , the data D _S2 is a blue color difference signal (B ″) corresponding to a sample storage point recorded after performing, for example, a 4: 1: 1 sample conversion on the data D _1ST .
_2N-2 , B " _2N , B" _{2N + 2,.} . . ) Shows the data. In the present embodiment, data D _S2 corresponding to the sample storage point for the data D _1ST, without performing band limiting process using the equation (1), simply samples interpolation point from data D _1ST This is obtained by thinning out the corresponding data _Di1 . That is, data D corresponding to the sample storage point in data D _2ND
S2, is equal to the data D _S1 as sample storage point in the data D _1ST. In the data D _2ND , the data _{Di 2} is, for example, from the 4: 1: 1 format to the 4: 2 format.
2: 2 color difference signals and blue to be interpolated using equation (3) to the data D _S2 when resampling is carried out in method _{(B '' 2N-1,} B '' 2N + 1, ... ). The data D _i2 is calculated from the data D _S2 as the sample storage point. In the present embodiment, the data D _S2 is
Data D Data D as sample storage point in _1ST
Since this is equal to _S1 , this data _Di2 is also
It is equal to the data D _i1 which is the data interpolation point in _1ST .

As described above, in the present embodiment, the data subjected to the band limiting process is subjected to the second and subsequent band limiting processes to determine the sample interpolation points generated by the first sample conversion process. By performing a process of thinning out the data, the data is prevented from deteriorating.

Here, it is determined whether or not the data to be subjected to the sample conversion is data that has been subjected to the band limiting process once, when the sample interpolation point of the data to be subjected to the band limiting process satisfies Expression (3). It can be done depending on whether or not there is.
This is because, as can be seen from FIG. 3, the sample interpolation points are calculated only at the band-limited sample storage points. Therefore, when the sample interpolation point of the data to be band-limited satisfies Expression (3), it can be determined that the sample point is data for which the band-limiting processing of the frequency has already been performed. That is, when Expression (3) is satisfied, it is understood that the data originally has only the frequency band of the sample storage point. By simply performing the band limiting process on such data by thinning out the sample interpolation points without using a filter as in equation (1), the data can be restored without deteriorating the data during resampling. It is possible to do.

Such a process in which the filtering process is not performed is also applied to original 4: 2: 2 data that has not been sample-converted. Even in the case of original 4: 2: 2 format data, data having a frequency band only at the same sample storage point as the 4: 1: 1 format or 4: 2: 0 format (for example, CG (computer graphics) In the case of artificial image data such as :), the expression (1) is used without using
By simply performing the process of thinning out the sample interpolation points, the data can be restored without deteriorating the data.

FIG. 4 shows a 4: 2: 2 system and a 4: 1: 1 system.
FIG. 4 is an explanatory diagram for describing data frequency band limitation in a system. In this figure, black circles indicate sample storage points, and white circles indicate sample interpolation points. Also, in this figure, (A) shows the distribution of sample points in the 4: 2: 2 system on the signal waveform of the original data. Also, in this figure, (B) shows the distribution of sample points in the 4: 1: 1 system on a straight line corresponding to (A), and (C) shows the distribution in the 4: 2: 2 system. The distribution of sample points is shown on a straight line corresponding to (A).

In the data of the 4: 2: 2 system, since 6.75 MHz is used as the sampling frequency F _S ,
By the sampling theorem, there should be a signal band up to 3.375 MHz. However, as shown in FIG. 3 (C), the sample points of the sampled data are
If it is composed of the sample interpolation points interpolated between the sample storage points, there is no signal band up to 3.375 MHz. That is, when the sample storage points and the sample interpolation points are alternately arranged as shown in FIG. 3C, the signal band is 1.6, which is lower than 3.375 MHz.
This means that the band is limited to 875 MHz. This is equal to the signal band by the 4: 1: 1 system as shown in FIG. In the data of the 4: 1: 1 method, since 3.375 MHz is used as the sampling frequency F _S , 1.6875 MHz is used according to the sampling theorem.
There is a signal band up to z. The data processing device of the present embodiment can prevent data degradation when performing a sample change on such band-limited data.

Next, a filter function used in the data processing apparatus of this embodiment shown in FIGS. 1 and 2 will be described.

The three-tap filter function of the above equations (1) and (2) can be expressed by the following equations (5) and (6). Equation (5) is replaced by equation (1)
Equation (6) corresponds to Equation (2). In these equations, the same function equation (4)
Contains. By using the equations (5) and (6) including the same function equation (4) in this way, in the band limiting process, the frequency band limiting process at the sample storage point and the frequency band limiting at the sample interpolation point are performed. The same circuit element (the filter circuit 5 in FIG. 1) can be shared when performing the data discrimination processing.
The two processes can be performed collectively on the time axis.

F (B _N−1 , B _{N + 1} ) = １ ／ · (B _N−1 + B _{N + 1} ) ‥‥‥ (4) B ′ _2N = １ ／ · [B _2N + f (B _2N-1 , B _{2N + 1} )] ‥‥‥ (5) B ' _{2N + 1} = １ ／ · f (B _2N , B _{2N + 2} ) ‥‥‥ (6)

Here, in the circuit shown in FIG. 1, an arithmetic processing corresponding to the equation (4) is performed in the filter circuit 5. Further, the delay circuit 4, the filter circuit 5, the product-sum operation circuit 6, and the 演算 operation circuit 7 perform an operation corresponding to the equation (5). Further, the rear-side interpolation point comparison circuit 11 and the front-side interpolation point comparison circuit 15 in the frequency discrimination circuit 10 perform an arithmetic process corresponding to the equation (6). In the rear-side interpolation point comparison circuit 11, a rear-side interpolation point (for example, B _2N ) is
_{2N + 1} ) is performed, and the front interpolation point comparison circuit 15
In, the process related to the _preceding interpolation point (for example, B _2N−1 ) is performed on the saved point (for example, B _2N ). In the circuit shown in FIG. 2, the filter circuit 36 performs an arithmetic operation corresponding to the equation (6).

FIG. 5 is an explanatory diagram for explaining the frequency band limiting process and the process of determining the frequency band limited data in the present embodiment. In this figure, the data D ₁₀ represent respectively the particle diameters for example, 4: 2: color difference signals and blue as the sampled input data by 2 mode _{(B 2N-2, B 2N} -1, B 2N,
B _{2N + 1} , B _{2N + 2,.} . . ) Shows the data. Also, in this figure, data D ₁₁ is accompanied by band limiting processing using equation (5) on data D ₁₀ , for example, 4:
A blue color difference signal (B ′ _2N−2 , B ′ _2N−2) corresponding to a sample storage point recorded after performing 1: 1 sample conversion.
B ′ _2N , B ′ _{2N + 2,.} . . ) Shows the data. As shown in this figure, in this embodiment, with respect to data D ₁₀ as input data, performs the equation (5) the frequency band limitation process in the sample storage point as required, sample interpolation point (B _2N-1 ,
B _{2N + 1,.} . . The processing of discriminating the data whose frequency band is limited in the step (1) is performed collectively on the time axis by the equation (6).

Next, with reference to FIG. 6 and FIG. 7, a description will be given of the determination processing of the frequency band limitation in the present embodiment. In these figures, the distribution of sample points in 4: 2: 2 data is shown on a straight line. In these figures, black circles indicate sample storage points, and white circles indicate sample interpolation points. However, in FIG. 6, white circles indicate sample interpolation points that match the filter function represented by equation (6), and crosses indicate equation (6).
Shows sample interpolation points that do not match the filter function represented by.

As shown in FIG. 6, with respect to data in a section where the sample interpolation point matches the filter function represented by equation (6), it can be said that the section is band-limited data. In FIG. 6, B _2N-2 ~B _2N data section included sample interpolation points that match the filter function corresponds to the band-limited data section. That is, B
_In the data section of _{2N−2 to} B _2N , the band is 1.6875 in the normal band of 3.375 MHz according to the 4: 2: 2 system.
The band is limited to MHz.

[0051] Here, for example, the value determination of whether or not in the interval of sample storage point B _2N is a frequency band limitation, as shown in FIG. 7, the samples before and after the interpolation point B of the sample storage point B _2N It can be determined based on whether or not _2N-1 and B2 _{2N + 1} match the filter function of Expression (6). Thereby, it is possible to determine the frequency band limitation for each sample storage point. That is, in the present embodiment, it is possible to perform the determination process of the frequency band limitation in units of one pixel.

Next, with reference to FIG. 8 to FIG. 10, data processing at the data end in the data processing apparatus of the present embodiment will be described. 8 to FIG.
The portions indicated by the symbols D _E1 and D _E2 indicate end portions of the data. In these figures, black circles indicate sample storage points, and white circles indicate sample interpolation points. In FIGS. 8 and 10,
The distribution of sample points in 4: 2: 2 data is shown on a straight line. In FIG. 9, the distribution of the sample points in the data of the 4: 1: 1 system or the 4: 2: 0 system is shown on a straight line.

At the end of an image (hereinafter referred to as a sample point end), sufficient data for the number of taps required for the filter processing cannot be obtained. In some cases, a process for reducing the noise peculiar to the end of the sample point may be performed by extending the sample point on the virtual data from the end by the number of taps of the filter. In FIG. 7, the sample point ends D _E1 and D _E1
The data B _0, B m _{+ 1} in the _E2 to the point closest to the sample point end D _E1, D _E2, sample point end D _E1, D
An example is shown in which data is extended on virtual data from _E2 .

FIG. 9 shows an example in which 4: 2: 2 data is sampled in 4: 2: 1 format or 4: 2: 0 format with respect to 4: 2: 2 data in which data is extended at the sample point end as shown in FIG. This shows data when conversion is performed.
Even in the case where the frequency band limitation processing is performed in accordance with the sample conversion, the sample point end is extended by the filter tap similarly to FIG. Here, as shown in FIG. 9, when the data of the sample storage point in which the band is limited by the 4: 1: 1 system or the 4: 2: 0 system is resampled into the data of the 4: 2: 2 system, the data at the sample point end is obtained. There are two cases of the data of the storage point and the interpolation point.

FIG. 10 shows a case where the data at the sample storage point whose band is limited by the 4: 1: 1 system or the 4: 2: 0 system as shown in FIG. 9 is resampled into 4: 2: 2 system data. FIG. In the case of this figure, the data at the sample point end _DE1 is the sample storage point _B0.
And the data at the sample point end _DE2 is the sample interpolation point _{Bm + 1} . Here, as in the case of FIG. 8, when data is extended using only one point at the end of the sample point, the relationship between the sample storage point and the interpolation point on the extended data is broken, and the sample in this embodiment is broken. Band limit discrimination processing at the point end becomes impossible.

FIG. 11 is an explanatory diagram showing data processing at the data end performed in the data processing apparatus of the present embodiment. In this figure, black circles indicate sample storage points, and white circles indicate sample interpolation points. In these figures, the portions indicated by the symbols D _E1 and D _E2 indicate the end portions of the data.

In the present embodiment, as shown in the example of FIG.
In order to avoid that the relationship between the sample storage point and the interpolation point on the extended data is broken and the band limit discrimination processing at the sample point end in the present embodiment becomes impossible, the two sample points closest to the sample point end Points are used alternately as extension data. As a result, the relationship between the sample storage point and the sample interpolation point is established at the sample end, and the band limitation determination process can be performed. In the example of FIG. 11, the sample point end D
_E1 data by use alternately is extended data B _0, B ₁ from. Also, data B from the sample point end D _E2
The data is extended using _m and B _{m + 1} alternately. In addition, in order to perform the band limitation discrimination process at the storage point in the effective area of the data, it is necessary to supplement the data up to the interpolation point adjacent to the storage point at the sample point end. That is, it is necessary to extend the data for the number of taps required for the filter function from the sample interpolation points at both ends to the virtual data.

Next, the operation of the data processing apparatus according to the present embodiment will be described with reference to FIGS. The following description also serves as a description of the data processing method according to the present embodiment.

FIG. 12 is a flowchart showing a band limiting process in the data processing device according to the present embodiment. In the circuit shown in FIG. 1, for example, the image edge processing circuit 1
4: 2: 2 format by the sampling frequency F _S in the sampled blue color difference signal _{(B 2N, B 2N + 1} , B 2N + 2,
B _{2N + 3,.} . . When coded data of) is input (step S101), based on the H / V parameter signal S ₂ inputted in storage and interpolation point switching circuit 3, the sample storage points and sample interpolation points of the input data The position is determined (Step S102). Next, based on the H / V synchronization signal S ₁ input in the H / V synchronization circuit 2,
It is determined whether or not the input data is at the end of the sample point (step S103). If the input data is at the sample point end (Y), the image end processing circuit 1 performs the sample point end processing as shown in FIG. 11 (step S104).

If the input data is not at the sample point end (step S103; N), and if the sample point end processing has been performed, then the band limit discriminating circuit 10 precedes the input data before and after the input data. It is determined whether or not each sample point has already been subjected to the band limiting process (step S105). If each of the preceding and succeeding sample points has already been subjected to band limiting processing (Y), data obtained by performing processing for merely thinning out interpolation point data without performing band limiting processing on input data S ₈ is outputted from the switching circuit SW2 (step S106). If any one of the preceding and succeeding sample points has not been subjected to the band limiting process (step S105; N), the delay circuit 4, the filter circuit 5,
Data S ₇ to band limitation processing frequency is performed through the product-sum operation circuit 6 and the half operation circuit 7 is outputted from the switching circuit SW2 (step S107).

Next, based on the storage and interpolation point switching signal S ₅ from the storage and interpolation point switching circuit 3, whether the data is sample storage point is determined (step S108).
If it is the sample storage point (Y), the switching circuit SW3 outputs, for example, the sampling frequency F _S / 2 of 4:
Data corresponding to the 1: 1 system (color difference signals B ′ _2N ,
B ′ _2N , B ′ _{2N + 2} , B ′ _{2N + 2,.} . . ) Is output (step S109). If it is not the sample storage point (step S108; N), the data is truncated (step S110).

FIG. 13 is a flowchart showing the resampling interpolation processing in the data processing apparatus according to the present embodiment. In the circuit shown in FIG. 2, the FIFO circuit 31 supplies, for example, a blue color difference signal (B _2N , B _{2N + 2} ,...) Encoded at a sampling frequency F _S / 2 by a 4: 1: 1 method. ) Is input (step S201), the data is stored and stored.
Based on the H / V parameter signal S ₁₂ inputted in the interpolation point switching circuit 33, the position of the sample storage points and sample interpolation points of the input data is determined (step S
202). Then, based on the H / V sync signal S ₁₁ which is input in the H / V sync circuit 32, the inputted data whether a sample point end is judged (step S203). Here, if the input data is the sample point end (Y), the image end processing circuit 34 performs the sample point end processing as shown in FIG. 11 (step S204).

If the input data is not at the sample point end (step S203; N), and if the sample point end processing has been performed, then the storage / interpolation point switching circuit 33 converts the data to sample interpolation. It is determined whether it is a point (step S205). If it is the sample interpolation point (Y), the data S _{2N + 1 that} has been subjected to the interpolation processing using Expression (6) in the filter circuit 36 is output (Step S206). Next, when the interpolation processing is performed and when the data is not the sample interpolation point (step S206; N), the switching circuit SW
5, on the basis of the stored and the interpolation point switching signal S ₁₅ from the storage and interpolation point switching circuit 33, data from the delay circuit 35 S
_2N and the data S _{2N + 1} from the filter circuit 36 are output alternately, for example, at a sampling frequency F _S / 2 of 4: 1: 1.
The data subjected to the sump conversion is output (step S207).

As described above, according to the present embodiment, two-time
As the band limiting processing for the second time and thereafter, processing for thinning out the data of the sample interpolation points generated by the first sample conversion processing is performed. For example, the 4: 2: 2 method and the 4: 1: 1 method ( Or 4: 2: 0 method), there is no data deterioration due to multiple dubbing processes in which sample conversion is repeated, and even if theoretically infinite dubbing is performed, data that is substantially the same as the data subjected to the first sample conversion Can be obtained. Also, even if the original data has not been sample-converted yet, the image having the frequency band only at the sample saving point (artificial image such as CC) is not subjected to the band limitation processing. Since only sample interpolation point thinning processing is performed, sample conversion can be performed without deteriorating data as compared to uniform band limiting processing on original data, similar to data before sample conversion Data can be restored.

Further, according to the present embodiment, it is possible to perform the determination process of the frequency band limitation in units of one pixel at each sample storage point. Therefore, for example, the data base in the 4: 2: 2 system can be used. On the band, special effects, character insertion or image editing such as A and B rolls are partially applied to the image, and the 4: 2: 2 method and the 4: 1: 1 method (or 4: 1: 2 method) are performed.
Even when the sample conversion is performed between (2: 0 method) and the pixel point where the image editing is not performed, data deterioration due to the image editing does not occur. Accordingly, for example, even in the case of editing the data of the 4: 2: 2 system in the non-linear editor or the VTR device of the 4: 1: 1 system and the 4: 2: 0 system requiring the band limiting process, the editing is performed. Can minimize data degradation.

Further, according to the present embodiment, the processing of extending the sample point end is performed so that the relationship between the sample storage point and the interpolation point does not collapse, so that the band can be effectively used even at the sample point end. It is possible to perform a restriction determination process.

As described above, according to the present embodiment, it is possible to prevent data deterioration when sampling is performed at a sampling frequency different from the sampling frequency of input data.

[Second Embodiment] Next, a second embodiment of the present invention will be described.
An embodiment will be described. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted.

The frequency band limiting process in the sample conversion from the 4: 2: 2 system to the 4: 1: 1 system (or 4: 2: 0 system) is called an adaptive filter that adjusts according to the input data value. There is something. This is used to reduce ringing of data to be band-limited and to improve an image more. However, the adaptive filter is more likely to be degraded by repeated dubbing than a fixed filter such as the three-tap filter described above. Here, the thinning processing of the sample interpolation points according to the present invention is possible if the relationship between the sample storage points and the sample interpolation points is established by a certain filter function that has been subjected to the interpolation processing by resampling. It can be effectively used even with an adaptive filter. In the present embodiment, an example in which the adaptive filter is applied will be described.

FIGS. 14 and 15 are block diagrams showing the configuration of a data processing device according to the second embodiment of the present invention. Here, FIG. 14 shows a configuration of a circuit portion that performs sample conversion accompanied by band limiting processing in the data processing device according to the present embodiment. FIG. 15 shows a configuration of a circuit portion that performs resampling processing accompanied by interpolation processing in the data processing apparatus according to the present embodiment.

As shown in FIG. 14, the data processing apparatus according to the present embodiment receives data from the image edge processing circuit 1 and performs filtering on the input data using an adaptive filter function G. And a filter circuit 51 for outputting to the sum-of-products arithmetic circuit 6 and data from the image edge processing circuit 1. And a filter circuit 52 for outputting to the circuit 10. In the configuration shown in FIG. 14, the configuration other than the filter circuits 51 and 52 is the same as that in FIG.

As shown in FIG. 15, the data processing apparatus according to the present embodiment receives data from the image edge processing circuit 34 as a circuit element for performing resampling processing accompanied by interpolation processing. The input data is subjected to a filtering process using a predetermined function H so that the switching circuit SW
5 is provided. In the configuration shown in FIG. 15, the configuration other than the filter circuit 61 is as follows.
This is the same as the case of FIG. 2 and the description is omitted.

Here, in the present embodiment, the following equation (7) is used as a function equation corresponding to the equation (5) in the first embodiment. Further, in the present embodiment, the following equation (8) is used as a function equation corresponding to the equation (6) in the first embodiment. Here, the function G represents an adaptive filter function, and the function H represents an interpolation processing function corresponding to the equation (4) in the first embodiment.

B ′ _2N = G (B _2N−1 , B _2N , B _{2N + 1} ) {(7) B ′ _{2N + 1} = １ ／ · H (B _2N , B _{2N + 2} )} ‥ (8)

FIG. 16 is an explanatory diagram for explaining the frequency band limiting process and the process of determining the frequency band limited data in the present embodiment. In this figure, data D ₂₀ is, for example, a blue color difference signal (B _2N-2 , B _2N-1 , B _2N-1 , B _2N-2) as input data sampled by the 4: 2: 2 method.
B _2N , B _{2N + 1} , B _{2N + 2,.} . . ) Shows the data. Also, in this figure, data D ₂₁ is data D ₂₀
Is accompanied by a band limiting process using equation (7), for example, a blue color difference signal (B ′ _2N−2 , B ′ _2N−2 ,
B ′ _2N , B ′ _{2N + 2,.} . . ) Shows the data. As shown in this figure, in this embodiment, with respect to data D ₂₀ as input data, performs the equation (7) the frequency band limitation process in the sample storage point as required, sample interpolation point (B _2N-1 ,
B _{2N + 1,.} . . The processing of discriminating the data whose frequency band is limited in the step ()) is collectively performed on the time axis by the equation (8).

As described above, according to the present embodiment, even when an adaptive filter function is used, it is possible to prevent data deterioration due to sampling processing.

The other configurations, operations, and effects of this embodiment are the same as those of the first embodiment.

Note that the present invention is not limited to the above-described embodiment, and various modifications can be made. For example, in each of the above embodiments, the sample conversion between the 4: 2: 2 method and the 4: 1: 1 method (4: 2: 0) for image data has been mainly described. If the data to be band-limited has a relationship between the sample storage point and the sample interpolation point, and if a synchronous tap filter function is used for the band-limiting process and the interpolation process associated with resampling, sample conversion using another method It is also applicable to image data and audio data in. In this case, the tap filter function is such that the number of taps is 3
Other than the three-tap filter described above.

Even if the sample-converted storage point has deteriorated due to other processing (data compression / decompression processing, etc.), the relationship between the deteriorated sample storage point and the sample interpolation point interpolated from the storage point. Is satisfied in the interpolation processing at the time of resampling, the thinning processing of the sample interpolation points can be performed in the next band limitation processing. Therefore, the present invention can be applied to general compression / expansion processing involving sample conversion.

[0080]

As described above, according to the data processing device of the present invention or the data processing method of the present invention, the input data sampled is processed at a predetermined frequency. It is determined whether or not the data is band-limited in a band lower than the band, and based on the result of the determination, the input data is subjected to frequency band limiting as necessary, and Since the sampling is performed at a sampling frequency different from the sampling frequency of the input data, the effect of preventing data deterioration when sampling at a sampling frequency different from the sampling frequency of the input data can be prevented. Play.

According to the data processing device of the present invention, the input data is determined for each unit data,
Since sampling is performed by performing frequency band limiting processing as necessary for each unit data, for example, when input data is image data, processing is performed for each pixel as unit data. Therefore, even when sampling is performed by partially performing image editing on an image, data deterioration due to image editing can be prevented from occurring at pixel points in a portion where image editing is not performed. To play.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a circuit portion that performs sample conversion accompanied by band limiting processing in a data processing device according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a circuit portion that performs a resampling process including an interpolation process in the data processing device according to the first embodiment of the present invention.

FIG. 3 is an explanatory diagram showing a change in encoded data when a plurality of sample conversions are performed in the data processing device according to the first embodiment of the present invention.

FIG. 4 is an explanatory diagram for describing frequency band limitation of data in the 4: 2: 2 system and the 4: 1: 1 system.

FIG. 5 is an explanatory diagram for explaining a frequency band limiting process and a process of determining frequency band limited data in the data processing device according to the first embodiment of the present invention.

FIG. 6 is an explanatory diagram illustrating a frequency band limitation determination process in the data processing device according to the first embodiment of the present invention.

FIG. 7 is another explanatory diagram illustrating the determination processing of the frequency band limitation in the data processing device according to the first embodiment of the present invention.

FIG. 8 is an explanatory diagram for explaining sample point end processing in the data processing device according to the first embodiment of the present invention.

FIG. 9 is another explanatory diagram for explaining sample point end processing in the data processing device according to the first embodiment of the present invention.

FIG. 10 is still another explanatory diagram for describing sample point end processing in the data processing device according to the first embodiment of the present invention.

FIG. 11 is an explanatory diagram showing sample point end processing in the data processing device according to the first embodiment of the present invention.

FIG. 12 is a flowchart illustrating band limiting processing in a data processing device according to a second embodiment of the present invention.

FIG. 13 is a flowchart showing resampling interpolation processing in the data processing device according to the second embodiment of the present invention.

FIG. 14 is a block diagram illustrating a configuration of a circuit portion that performs a sample conversion accompanied by a band limiting process in the data processing device according to the second embodiment of the present invention.

FIG. 15 is a block diagram illustrating a configuration of a circuit portion that performs a resampling process including an interpolation process in the data processing device according to the second embodiment of the present invention.

FIG. 16 is an explanatory diagram for explaining a frequency band limiting process and a process of determining data subjected to frequency band limiting in the data processing device according to the second embodiment of the present invention.

FIG. 17 is an explanatory diagram illustrating a change in encoded data when a plurality of sample conversions are performed.

FIG. 18 is an explanatory diagram illustrating changes in filter coefficients and the number of taps when sample conversion is repeated using a three-tap filter.

FIG. 19 is an explanatory diagram more specifically showing data deterioration accompanying sample conversion.

[Explanation of symbols]

1: image end processing circuit, 2, 32: H / V processing circuit, 3,
34: storage / interpolation point switching circuit, 4, 35: delay circuit,
5, 36: filter circuit, 10: band-limit determining circuit, 1
1 ... rear interpolation point comparison circuit, 15 ... front interpolation point comparison circuit.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04N 11/04 F term (Reference) 5C055 AA01 BA01 CA15 EA02 EA04 FA22 HA01 HA27 HA31 5C057 AA08 BB01 CC04 CE03 CE10 EA02 EA07 EA17 EF00 EH01 EJ02 EK02 EK03 EL01 FA01 GC01 GC04 GF08 GK02 5J064 AA01 BB04 BC08 BC11 BC14 BC18 BC24 BD01

Claims (8)

[Claims] A determining means for determining whether or not the input data sampled is data subjected to band limiting processing in a band lower than a predetermined frequency band; and a determination result of the determining means. In response to the input data, if necessary, perform a frequency band limiting process,
A data processing apparatus comprising: sampling means for performing sampling at a sampling frequency different from the sampling frequency of the input data. 2. The method according to claim 2, wherein the sampling unit determines that the input data is data that has already been subjected to band limiting processing in a band lower than a predetermined frequency band. 2. The data processing apparatus according to claim 1, wherein the input data is sampled without performing the restriction processing. 3. The method according to claim 1, wherein the discriminating unit is configured to sample, by the sampling unit, first data corresponding to a part of the input data which is sampled by the sampling unit and thinned out without being stored. If the input data is equal to a value calculated by a predetermined function based on the second data corresponding to the portion stored after the data is input, it is determined that the input data is data that has already been subjected to the band limiting process. The data processing device according to claim 1, wherein: 4. The method according to claim 1, wherein the first data and the second data are data whose data positions are determined in advance according to a predetermined sampling standard in a stage prior to the sampling processing by the sampling means. Claim 3
The data processing device according to claim 1. 5. The data processing apparatus according to claim 1, wherein the sampling unit performs the restriction process on the frequency band using the same function formula as a predetermined function formula used in the determination unit. 6. The discriminating means performs discrimination of the input data for each unit data, and the sampling means responds to the discrimination result of the discriminating means.
2. The data processing apparatus according to claim 1, wherein sampling is performed for each unit data by performing frequency band limiting processing as necessary. 7. The data processing apparatus according to claim 1, wherein the determination unit performs the determination by extending data at a data end of the input data. 8. It is determined whether or not the sampled input data is data that has been subjected to a band limiting process in a band lower than a predetermined frequency band. On the other hand, a data processing method characterized by performing frequency band limiting processing as necessary and performing sampling at a sampling frequency different from the sampling frequency of the input data.