WO2002058244A1 - Compression method and device, decompression method and device, compression/decompression system, recording medium - Google Patents
Compression method and device, decompression method and device, compression/decompression system, recording medium Download PDFInfo
- Publication number
- WO2002058244A1 WO2002058244A1 PCT/JP2002/000277 JP0200277W WO02058244A1 WO 2002058244 A1 WO2002058244 A1 WO 2002058244A1 JP 0200277 W JP0200277 W JP 0200277W WO 02058244 A1 WO02058244 A1 WO 02058244A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- differential
- value
- points
- sample
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
Definitions
- the present invention relates to a compression method and apparatus, a decompression method and apparatus, a compression / decompression system, and a recording medium, and more particularly to a compression and decompression method for continuous analog or digital signals.
- the signals are compressed to reduce the amount of transmitted information and extend the time that can be stored in storage media.
- Stretching is being done.
- the analog signal is first sampled and digitized according to a predetermined sampling frequency, and the obtained digital data is subjected to compression processing.
- DPCM Pulse Code Modulation
- a method for performing time-frequency conversion there is a method using a sub-band filter or MDCT (Modified Discrete Cosine Transiorm), and an encoding method using such a method is MPEG (Moving Picture IM). age C ding ng E xperts G ro up) Audio power.
- MDCT Modified Discrete Cosine Transiorm
- the most widely used image compression systems are also commonly known as the MPEG standard.
- Decompression processing of data compressed according to the above-mentioned compression method is basically performed by the reverse operation of compression processing of the same compression method.
- the compressed digital data is converted from a frequency domain signal to a time domain signal by frequency / time conversion processing, and then subjected to a predetermined decompression processing, whereby the original digital data is reproduced. You. Then, the original data obtained in this way is digital-to-analog converted as necessary, and output as an analog signal.
- the compression side of the present invention differentiates the data to be compressed for each sampling point, and sequentially adds the absolute values of the differentiation, thereby differentiating for each sampling point. Find the total data. Then, the difference between the differential sum data obtained at each sampling point and the original data when performing linear interpolation between the data at the two sampling points is less than the desired value. Processing is performed to sequentially detect points as sample points of compressed data.
- the differential sum data at each sample point included in the compressed data, the timing data indicating the time interval between the sample points, and the polarity data of the differential value at each sampling point are obtained. Based on this, the amplitude data at each sampling point is obtained. Then, decompressed data is obtained by performing an interpolation operation for interpolating between the obtained amplitude data at each sampling point.
- each data value on a straight line connecting data of two sampling points and each differential value at the same sampling point as each data value on the straight line are provided.
- the sampling point where the error from the total data value is less than the desired value and the time interval between the above two sampling points is the longest in the predetermined range is the sampling point of the compressed data.
- two sampling ports are provided on the compression side.
- a sampling point in which an error between a data value on a straight line connecting data of the input and a differential sum data value at the same sampling point as the data value on the straight line is equal to or less than a desired value;
- the sampling point immediately before the sampling point whose error exceeds the above-mentioned desired value is sequentially detected as a compressed sampling point.
- the compressed data includes differential total data at each sample point, timing data representing a time interval between the sample points, and polarity data of a differential value at each sampling point.
- the compressed data includes timing representing a time interval between each sampling point, data of an average differential value per unit time between the sampling points, and polarity data of the differential value at each sampling point. Including evening.
- normal sample data at several points in each sampling point is adopted as a part of the compressed data. This regular sample data is used when performing interpolation processing on the decompression side.
- normal data points can be arbitrarily detected. For example, it is possible to detect normal data points at regular or irregular intervals of sampling points.
- the sampling point in which the value of the differential total data exceeds a predetermined threshold value or the data value of the sampling point in which the previous regular sample data was used is used.
- regular sample data may be used as part of the constrained data. .
- sampling points at which an error from original data becomes equal to or less than a desired value when linear interpolation is performed between two sample data included in data to be compressed are sequentially set as sampling points. Detected and obtain a set of the amplitude data of each sample point and the timing data indicating the time interval between each sample point as linear compressed data, and the amplitude data of each sample point included in the linear compressed data and the timing data between them And decompressing data by linearly interpolating between amplitude data having a time interval indicated by the timing data—interpolated data to obtain decompressed data.
- the processing described in any one of the items is performed.
- the present invention comprises the above technical means, even if the data of each sampling point is reproduced from the average differential value data between sample points and the polarity data of the differential value during the decompression process, an error from the original data is obtained.
- Sampling points at which the sample value does not become larger than the desired value are detected as sample points, and the differential sum data at the discrete sample points detected in this way, the average differential value data per unit time between sample points, and the sample point
- Only the timing data representing the time interval of the data, the polarity data of the differential value of each sampling point, etc. are generated as compressed data, and the quality of the data reproduced by decompression is significantly improved while achieving a high compression ratio. It becomes possible.
- the sample data is differentiated and the absolute values are sequentially added.
- error judgment processing on the differential total data generated by the above to compress high-frequency signals that is, signals whose sample data values change relatively significantly even at nearby sampling points
- the number of sample points to be detected can be reduced as much as possible, and a higher compression ratio can be realized.
- the present invention when compressing a signal on the time axis, it is possible to perform processing on the time axis without performing time / frequency conversion and performing processing on the frequency axis. Become. Also, when decompressing data compressed in this way, it is possible to perform processing while keeping the time axis.
- the original data before compression can be obtained simply by performing extremely simple processing, such as adding the polarities to the averaged differential values and sequentially adding them, or interpolation (which may be a simple one such as linear interpolation). It is possible to reproduce high-precision decompressed data, which is almost the same as in the above.
- each data value on a straight line connecting data of two sampling points, and each differential total data at the same sampling point as each data value on the straight line Sampling points whose error from the values are all less than or equal to the desired value and whose time interval between the above two sampling points is the longest in the predetermined range are sequentially detected as sample points of the compressed data.
- a data value on a straight line connecting data of two sampling points, and a differential total data value at the same sampling point as the data value on the straight line Is the sampling point at which the error from the sampling point is less than or equal to the desired value, and the sampling point immediately before the sampling point at which the above error exceeds the desired value.
- the average differential value data per unit time between sample points as compressed data, compared with the case where compressed differential data at each sample point itself is used as compressed data, thus, the amount of individual data can be reduced, and the compression ratio can be further increased. Further, it is not necessary to perform the processing of calculating the average differential value data from the differential total data and the timing data at each sample point on the decompression side, and the processing load can be reduced.
- the normal sample data is converted to a part of the compressed data at the sampling point where the value of the differential total data exceeds a predetermined threshold.
- data compression is performed by obtaining differential total data as described above.
- data compression can be performed after removing unnecessary high-frequency components that cause noise in advance.
- the compression ratio can be further improved, and the quality of data reproduced by decompression based on the compressed data can be further improved.
- FIG. 1 is a diagram for explaining the basic principle of the compression method according to the first embodiment.
- FIG. 2 is a diagram for explaining the basic principle of the compression method according to the first embodiment.
- FIG. 3 is a diagram for explaining the basic principle of the decompression method according to the first embodiment.
- FIG. 4 is a block diagram illustrating an example of a functional configuration of the compression device according to the first embodiment.
- FIG. 5 is a block diagram illustrating a functional configuration example of the decompression device according to the first embodiment.
- FIG. 6 is a diagram for explaining the basic principle of the compression method according to the second embodiment.
- FIG. 7 is a diagram for explaining the basic principle of the decompression method according to the second embodiment.
- FIG. 8 is a block diagram illustrating a functional configuration example of a compression device according to the second embodiment.
- FIG. 9 is a block diagram illustrating a functional configuration example of a compression device according to the third embodiment.
- FIG. 10 is a diagram showing waveforms of an original analog signal (input data) before compression and a reproduced analog signal (output data) obtained by compressing and expanding the analog signal.
- FIG. 11 is a partially enlarged view of the waveform shown in FIG.
- FIG. 12 is a characteristic diagram showing a correlation between an original analog signal (input data) before compression and a reproduced analog signal (output data) obtained by compressing and expanding the analog signal.
- the compression method of the present embodiment first, when an analog signal is input as a signal to be compressed, the input analog signal is AZD-converted to digital data. Then, the following compression processing is performed on the obtained digital data. When digital data is input as a signal to be compressed, the following compression processing is directly performed on the digital data.
- FIG. 1 and FIG. 2 are diagrams for explaining the basic principle of the compression processing according to the first embodiment.
- the horizontal axis represents time, and the vertical axis represents data amplitude.
- the solid line waveform shown in Fig. 1 shows an example of an analog signal to be compressed.
- S 1 to S 20 are a part of digital data obtained by sampling the analog signal to be compressed for each clock CLK based on a predetermined sampling frequency.
- the sample data S 1 is This is the data of the sample points of the standard used first.
- digital data (sample data S 1 to S 20) to be compressed is differentiated for each sampling point, and their absolute values are sequentially added.
- the differential sum data D 3 at the third sampling point is a value obtained by adding the absolute value of the differential between the sample data S 2 and S 3 to the immediately preceding differential sum data D 2.
- the differential total data D 4 at the fourth sampling point is a value obtained by adding the absolute differential value between the sample data S 3 and S 4 to the immediately preceding differential total data D 3.
- the differential absolute values at each sampling point are sequentially added to obtain differential total data D 1 to D 20 for each sampling point.
- the sampling point exceeding the threshold value is Use regular sample data instead of differential sum data (hereinafter, this sampling point is referred to as a normal data point).
- this sampling point is referred to as a normal data point.
- the differential total data D 12 exceeds the threshold value at the 12th sampling point, so that the regular sample data S 12 is adopted at this sampling point.
- the differential absolute values of the sampling points are sequentially added starting from the regular sample data S12. And the 18th At the sampling point, the differential total data D 18 again exceeds the threshold value, so that the regular sample data S 18 is also used at this sampling point. Then, with the sample data S 18 as a starting point, the same processing is repeated.
- the following linear compression processing is performed on the differential total data D1 to D20 at each sampling point obtained in this manner. That is, a data value on a straight line connecting between the data of two sampling points (between differential total data or normal sample data and differential total data), and a data value on the straight line Sampling points whose error from the differential sum data value at the same sampling point is equal to or smaller than a desired value are sequentially detected as sampling points. Then, the discrete differential sum data at each detected sample point and timing data (the number of clocks) representing the time interval between each sample point or between the sample point and the normal data point are obtained. Transmit or record as part of the compressed data.
- the process for detecting the sample points will be described more specifically as follows. That is, from the differential total data at each sampling point or the regular sample data at each normal data point, the reference data and the other data whose time interval is within a predetermined range from there. And. Then, the sampler in which the error between each data value on a straight line connecting the two data and each differential total data value at the same sampling point as each data value on the straight line is less than or equal to a desired value is obtained. A sampling point, which is a sampling point and has a longest time interval in the predetermined range, is detected as a sample point.
- FIG. 2 is a diagram for explaining this operation principle.
- the horizontal axis represents time
- the vertical axis represents the amplitude of differential sum data.
- D1 to D9 shown in FIG. 2 are a part of the differential total data obtained by the processing of FIG.
- the values of the differential total data D 1 to D 9 shown in FIG. 2 do not exactly match those of FIG. 1, but actually, the values of the differential total data D 1 to D 9 shown in FIG. The following processing is executed.
- the time interval between the two data selected when detecting discrete sampling points is within the range of 6 clocks at the maximum.
- the time interval between the differential sum data can be 7 clocks or 15 clocks at the maximum.
- the differential total data D 1 of the reference and the differential total data D 7 whose time interval is maximum within a predetermined range are selected.
- the data values D 2,, D 3 ′, D 4 ′, D 5 ′, and D 6 ′ of each sampling point on the straight line connecting the two differential total data, and each data value on the straight line Determine whether each error of each differential sum data value D 2, D 3, D 4, D 5, D 6 at the same sampling point as D 2 'to D 6' is less than or equal to the desired value I do.
- the time interval from the reference differential total data D1 is one clock CLK shorter than the differential total data D7. Choose D6 in the evening. Then, the data values D2 ", D3", D4 ", D5" of each sampling point on the straight line connecting the two differential sum data D1 to D6, and the data values on the straight line Determine whether each error with each differential sum data value D 2, D 3, D 4, D 5 at the same sampling point as D 2 "to D 5" is less than or equal to the desired value .
- the sampling point of the differential total data D6 is detected as a sample point.
- the error between each data value D 2 ", D 3", D 4 ", D 5" on the straight line and each differential total data value D 2, D 3, D 4, D 5 is a desired value. Since the following is obtained, the sampling point of this differential total data D 6 is detected as a sample point.
- the error condition that all errors are less than the desired value If neither is satisfied, the sampling point of the differential total data D2 is detected as a sample point.
- the sample point When one sample point is detected, the sample point is newly used as the differential sum data of the reference, and the same processing as above is performed within a range of six clocks from there.
- the sampling point at which all errors are less than or equal to the desired value within the range of 6 clocks from the differential total data D6 and the time interval from the differential total data D6 is the longest is the next sampling point. Detected as
- a plurality of sample points are sequentially detected in the same manner.
- the selection of the two data that form the straight line is based on one normal data point and the next normal data.
- the interval until the evening point is set as one break.
- the sample data of the regular data points (S12 and S18 in Fig. 1) are always used as the reference-side data.
- the amplitude value of the differential sum data at each discrete sample point detected in this manner and the time interval between the normal data point and the sample point between each sample point are represented by the number of clocks CLK.
- a pair with the compression data value is obtained as a part of the compression data.
- the pair (D 1, 5), (D 6, *) of the differential total data value (D 1, D 6,-) and the timing data value (5, *,...) at each sampling point is It is obtained as part of the compressed data (* indicates that it is undecided in this example)
- the sample data S12 and S18 of the normal data points also form part of the compressed data.
- the sampling point (the differential total data D 1 and D 7 in the example of FIG. 2) in which the time interval between the two data is the maximum within the predetermined range is selected, and the error judgment is started.
- the direction of the sample point search is not limited to this.
- the sampling point where the time interval between the two differential sum data is the minimum within a predetermined range (in the example of FIG. 2, the differential sum data D 1 and D 3) is selected, and the error judgment is started.
- the processing may be performed in a direction in which the interval is gradually increased.
- error sampling is started by selecting a sampling point (for example, differential total data D 1 and D 4 in the example of FIG. 2) where the time interval between the two differential total data is near the center within a predetermined range. Is also good.
- the differential sum data at each sample point detected in each section in this manner, the timing data representing the time interval between the sample points or between the normal data point and the sample point, and the Regular sample data and data representing the polarity of the differential data at each sampling point are obtained as compressed data and transmitted to a transmission medium or recorded on a recording medium.
- the differential sum data at discrete sample points extracted from each sampling point in the data to be compressed the timing representing the time interval between sample points, etc.
- Data, sample data of discrete normal data points, and obtain only the polarity data of each differential value that can be simply represented by either "0" or "1" as compressed data Therefore, a high compression ratio can be realized.
- the sampling point with the longest time interval from the evening of the reference data is sampled. Point To detect.
- the value of the timing data can be kept within a predetermined bit, the number of sample points to be detected can be reduced as much as possible, and a high compression ratio can be realized.
- the compression method of the present embodiment instead of performing the linear compression processing as shown in FIG.
- each sample data S 1 to S 2 on each sample data S 1 to S 20 itself, each sample data S 1 to S 2 Linear differential processing is performed on the differential sum data D 1 to D 20 generated by differentiating 0 and sequentially adding their absolute values.Therefore, each sample data S 1 to S The compression ratio can be further increased as compared with the case of performing linear compression processing on 20.
- sampling points are discretely taken even in high frequency parts such as the first and second sample data S 12 and thereafter.
- the number of sample points to be detected can be reduced as much as possible. Therefore, the number of differential sum data at the sampling points that should be held as compressed data can be reduced as much as possible, and the compression ratio can be increased.
- the decompression method of the present embodiment for decompressing the compressed data generated as described above will be described.
- the amplitude data values of sampling points that may exist between discrete sample points detected on the compression side are input. It is determined based on the differential total data at each sample point included in the compressed data, the timing data indicating the time interval between sample points, etc., and the polarity data of each differential value.
- the average differential value data per unit time is obtained from the difference between the data values at the two sampling points and the evening data, and the polarity data of the differential value at each sampling point is added to the average differential value data obtained.
- the amplitude data value at each sampling point is obtained by sequentially adding the value obtained by multiplying the value to the immediately preceding amplitude data value.
- the interpolation data interpolating between the individual data is obtained. Generate. Further, the generated interpolation data is converted into analog signals by DZA conversion as required, and output.
- FIG. 3 is a diagram for explaining the principle of extension.
- Q 1 to Q 20 indicate the amplitude data values at each expanded sampling point.
- Q 1, Q 12, and Q 18 are normal sample data at the normal data point.
- five points of differential total data D 2, D 6, D 11, D 13, and D 17 are detected as sampling points.
- the normal sample data S 1 1 and the differential sum data D 2 are used as they are as the amplitude data values Q 1 and Q 2 of the expanded data.
- the amplitude data Q 3 to Q 6 at these four sampling points are calculated as the differential sum data D 2 and D 6 at the two sample points included in the Shochu data. Then, it is obtained based on the timing between sampling points (4 CLK) and the polarity data (1, +, +, 1) of the differential value of each sampling point.
- the average differential data value per clock ( D 6-D 2) No 4) is calculated. Then, the value obtained by multiplying the average differential data value by the polarity data (1, +, +,-) of the differential value at each sampling point is sequentially added to the immediately preceding amplitude data value Q 2. Thus, the amplitude data values Q3 to Q6 at the four sampling points are obtained.
- the amplitude data values Q7 to Q11 at the sampling points are obtained by dividing the differential sum data D6 and D11 at the two sampling points, the timing between the sampling points (5 CLK), and each sampling point. It is determined based on the polarity data (1, 1, 10, +, +) of the differential value at the point.
- interpolation data having a waveform as shown in FIG. 3 is obtained. Furthermore, the generated interpolation data is subjected to D / A conversion processing, converted to an analog signal, and output.
- the average differential value data per clock is obtained from the differential total data and the timing data at each sample point included in the compressed data generated according to the compression method of the present embodiment. , And the polarity data of the differential value at each sampling point, and the amplitude data values Q 1 to Q 20 of each sampling point are obtained.
- the individual sections for which error determination is performed on the compression side are microscopically observed, the amount of error from the original data is naturally smaller, but as the plurality of sections are processed, the slight error It can be considered that the difference from the original data gradually increases when the whole is viewed macroscopically.
- the normality at several points in each sampling point is Since the sampled data is used as part of the compressed data, the accumulated error can be cut off by the regular sampled data inserted in some places. Reproducibility can be improved.
- FIG. 4 is a block diagram illustrating a functional configuration example of a compression device according to the first embodiment that implements the above-described compression method.
- the compression device shown in FIG. 4 is applicable, for example, when an analog audio signal is input and compressed.
- the first-stage single-pass filter (LPF) 1 and the A / D converter 2 are unnecessary.
- the compression device of the present embodiment includes an LPF 1, an A / D converter 2, a D-type flip-flop 3, a differentiator 4, a differential total data calculator 5, a linear It comprises a compression unit 6 and a blocking unit 7
- LPF 1 removes high frequency component noise by performing a filtering process on an analog signal input as a compression target in order to facilitate detection of a sample point.
- the AZD converter 2 converts an analog signal output from LPF1 into digital data.
- the AZD conversion unit 2 performs A / D conversion processing according to an input clock of a predetermined frequency f ck (for example, 44.1 KHz in the case of an audio signal) as a reference.
- the D-type flip-flop 3 holds the digital data at each sampling point output from the A / D conversion unit 2 according to the input clock of the reference frequency fck.
- the differentiator 4 differentiates the sample data output from the D-type flip-flop 3. At this time, the differentiator 4 differentiates the sample data every time an input clock of the reference frequency fck is supplied, that is, for each sampling point based on the reference frequency fck.
- the derivative value is, for example, This is obtained by subtracting the current data captured at the clock timing from the data captured at the previous clock timing.
- the differential sum data calculating unit 5 calculates the absolute value of the differential value calculated for each sampling point by the differentiating unit 4, and sequentially adds each of the differential values for each sampling point. At this time, if the sum of the differential sum data exceeds a predetermined threshold value, normal sample data is used for sampling points that exceed the threshold value. Through such processing, the differential total data calculation unit 5 generates differential total data D 1 to D 20 having a waveform as indicated by a dashed line in FIG.
- the linear compression section 6 performs the linear compression processing described with reference to FIG. 2 on the differential total data D 1 to D 20 generated by the differential total data calculation section 5. As a result, the linear compression unit 6 detects discrete sample points from among the sampling points based on the reference frequency fck, and calculates the amplitude data value of the differential total data at each sample point and the difference between each sample point. And a timing value representing the time interval of.
- the blocking unit 7 includes data representing the polarity of the differential value at each sampling point calculated by the differentiating unit 4, the sample data of the normal data points obtained by the differential total data calculating unit 5, and the linear compression unit.
- the differential sum data at each sample point and the timing data indicating the time interval between each sample point, etc., obtained in step 6 are appropriately divided into blocks and output as compressed data.
- the output compressed data is transmitted to a transmission medium, for example, or recorded on a recording medium such as a nonvolatile memory.
- the polarity data of the differential value in each sampling point represented by the binary value of "0" and "1" is divided into separate fields for each sample point division and divided into blocks.
- the time interval (sample) between sample points depends on the number of polarity data included in one field. Lock number). Therefore, in this case, the timing data can be made unnecessary as the compressed data.
- FIG. 5 is a block diagram showing an example of a functional configuration of the decompression device according to the present embodiment.
- the decompression device according to the present embodiment includes a linear decompression unit 11, a D-type flip-flop 12, an interpolation processing unit 13, a DA conversion unit 14, and an LPF 15. It is configured.
- the linear decompression unit 11 performs the linear decompression processing as described in FIG. 3 on the input compressed data, so that the amplitude data Q 1 to Q 20 are obtained for each sampling point based on the reference frequency fck. To reproduce.
- the D-type flip-flop 12 holds the amplitude data Q 1 to Q 20 at each sampling point output from the linear extension unit 11 in accordance with the clock of 6 times frequency 6 fck. As a result, the digital data Q 1 to Q 20 at each sampling point are oversampled by a factor of six.
- the interpolation processing unit 13 uses the data sampled by the D-type flip-flop 12 to perform, for example, a straight line between the amplitude data Q1 to Q20 at each sampling point of the reference frequency fck. Performs the interpolation operation to generate interpolation data with a waveform as shown in Fig. 3.
- the DZA conversion unit 14 performs D / A conversion on the interpolation data generated in this manner to obtain an analog signal.
- the LPF 15 removes high frequency component noise by filtering the analog signal converted by the DZA converter 14, and outputs the signal as a reproduced analog signal.
- the average differential value per clock between sample points is obtained from the differential total data at each sample point, and the amplitude data at each sampling point is obtained from this. And its amplitude Interpolation data for interpolating the data for one night is output as decompressed data.
- high-precision decompression data that is almost the same as the original data before compression can be reproduced only by performing extremely simple processing such as linear decompression processing and linear interpolation processing.
- the compression device and the decompression device according to the present embodiment configured as described above are configured by, for example, a CPU or a computer system including an MPU, a R ⁇ M, a RAM, and the like.
- the device's differentiator 4, differential total data calculator 5, linear compressor 6, blocker 7, expander linear expander 11, interpolation processor 13, etc.) are stored in the above-mentioned ROM or RAM. This is realized by the operation of the program.
- the compression device and the decompression device according to the present embodiment configured as described above can be configured as hardware by combining logic circuits.
- the hardware configuration for realizing the function of the linear compression section 6 of the compression device and the function of the linear extension section 11 of the decompression device is described in Japanese Patent Application No. 2000-0001 filed earlier by the present applicant. This is described in detail in 6 826 5.
- the configuration described in detail in Japanese Patent Application No. 2000-166680 can be applied to the present embodiment.
- the interpolation data generated by the linear interpolation not only has a small amplitude error but also a very small phase shift compared to the original data before compression. Can be suppressed.
- the phase shift greatly affects the timbre, but in the present embodiment, since the phase shift hardly occurs, the timbre of the original data can be faithfully reproduced.
- the differential sum generated by differentiating each sample data and sequentially adding the absolute values thereof is obtained.
- Linear compression processing is performed on the data.
- the analog signal or digital data to be compressed can be directly compressed and expanded on the time axis without performing time-Z frequency conversion, so that processing is not complicated. Also, the configuration can be simplified.
- the compressed data is transmitted from the compression side and played back on the decompression side, the input compressed data must be sequentially processed and played back by a very simple linear interpolation operation on the time axis. Therefore, real-time operation can be realized.
- the normal sampling data is used for the sampling point. Is used as part of the compression data.
- the differential values at all the sampling points are sequentially added without performing such processing, and the differential total data thus obtained is added to the predetermined clock at a predetermined clock.
- An error determination may be performed for each range to sequentially detect sample points.
- by using regular sample data every time a predetermined threshold is exceeded accumulated errors can be cut off each time, and the reproducibility of analog signals reproduced by expansion from compressed data can be improved.
- the differential total data at each sample point, the timing data indicating the time interval between the sample points, the normal sample data at each normal data point, and the The polarity data of the differential value at is calculated as compressed data, and the average differential value data per clock is calculated on the decompression side from the differential total data and the timing data at each sample point included in the compressed data.
- the differential sum data at each sampling point is not required, and instead, it is sufficient to have the average differential data per clock between sampling points.
- This makes it possible to compress the amount of individual data as compared with the case where the differential total data itself is stored as a compressed data, thereby further increasing the overall compression ratio. Also, on the decompression side, the calculation for obtaining the average differential value data becomes unnecessary, and the calculation load can be reduced and the reproduction time can be shortened.
- the normal sample data when the differential total data exceeds a predetermined threshold value, the normal sample data is adopted, and thereafter, the differential total data is calculated starting from the normal sample data.
- the present invention is not limited to this example.
- regular sample data may be adopted at the sampling point.
- the differential total data value at the sampling point increases as it goes to the subsequent stage, and the compression ratio decreases.However, if the average differential value data is used as the compressed data, the high compression ratio is maintained. can do.
- the interpolation processing unit 13 performs linear interpolation between digital data Q 1 to Q 20, but the interpolation operation is not limited to this example.
- a curve interpolation process using a predetermined sampling function may be performed.
- the interpolation processing described in Japanese Patent Application No. 11-173732, filed earlier by the present applicant may be performed. In this case, a very analog waveform can be obtained by interpolation itself, so that the DZA converter 14 and LPF 15 at the subsequent stage of the interpolation processor 13 can be omitted.
- FIG. 6 is a view for explaining the basic principle of compression processing according to the second embodiment.
- the solid line waveform shows an example of an analog signal to be compressed
- S1 to S20 denote digital data obtained by sampling the analog signal to be compressed for each clock CLK based on a predetermined sampling frequency. Department.
- the analog waveform and the sample data S 1 to S 20 are exactly the same as those shown in FIG.
- the second embodiment is similar to the first embodiment shown in FIG. 1 in that the digital data to be compressed is differentiated for each sampling point and the absolute values are sequentially added. is there.
- the main difference from the first embodiment is the method of taking regular data points. That is, in the second embodiment, in the process of adding the differential absolute values of each sampling point, an error determination as shown in FIG. 2 is performed, and a normal data point is extracted according to the result.
- a data value on a straight line that connects data of two sampling points (between differential total data or between normal sample data and differential total data) and a sampling point that is the same as the data value on the straight line
- the sampling points in which the error from the differential sum data value obtained is less than or equal to a desired value are sequentially detected as sampling points.
- regular sampling data is used for the sampling point immediately after the detected sample point.
- the differential total data is obtained for each sampling point based on the first sample data S1. Then, when a sampling point at which the above error is equal to or less than a desired value and a time interval is the longest within a predetermined clock range from the first sample data S 1 is obtained, in this case, This is the second sampling point. Therefore, the sample data S 2 at this point is detected as the amplitude data at the sampling point, and the normal sampling at the next sampling point is performed. Rude night S 3 is adopted as sample data of normal data points.
- the differential total data is obtained for each subsequent sampling point based on the regular sample data S 3. Then, a sampling point where the above error is equal to or less than a desired value, and a sampling point having the longest time interval within a predetermined clock range from the sample data S3 is obtained. This is the second sampling point. Therefore, the sample data S 6 at this point is detected as the amplitude data of the sample point, and the normal sample data S 7 at the next sampling point is adopted as the sample data of the normal data point.
- the same processing is repeated.
- the sampling point immediately after a certain sample point is always a normal data point, and the next sample point always exists after the normal data point.
- the differential total data of the sampling points adopted as the compressed data and the regular sample data appear alternately.
- the timing data represents the time interval between a normal data point and a subsequent sample point.
- the discrete differential sum data at each sample point obtained by such processing, the normal sample data at each normal data point, the timing data representing the time interval therebetween, Polarity data of the differential value at the sampling point is obtained as compressed data and transmitted to a transmission medium or recorded on a recording medium.
- the decompression method according to the second embodiment for decompressing the compressed data generated as described above is almost the same as the decompression method described in the first embodiment. That is, the amplitude data values at the sampling points that may exist between the sample points adopted as the compressed data are converted into the differential sum data, the timing data, and the respective data at the respective sample points included in the input compressed data. Determined based on the polarity data of the differential value at the sampling point. Then, by performing an interpolation operation for interpolating between the obtained amplitude data at each sampling point and regular sample data included in the compressed data, interpolation data for interpolating between individual data is generated. I do. Furthermore, the generated interpolation data is converted into an analog signal by DZA conversion as required, and output.
- FIG. 7 shows reproduced data when the compressed data generated according to the compression method according to the second embodiment is decompressed.
- Q1 to Q20 indicate data values at each expanded sampling point.
- Ql, Q3, Q7, Q12, Q14, and Q20 are normal sample data values at normal data points.
- five points D2, D6, Dll, D13, and D19 are detected as sample points.
- the normal sample data S 1 and The differential sum data D 2 is used as it is as the amplitude data values Q 1 and Q 2 of the decompressed data.
- the normal sample data S3 is used as it is. Adopted as the amplitude data Q 3 of the decompressed data.
- the amplitude data at these three sampling points The values Q 4 to Q 6 are converted to the normal sample data S 3 and differential total data D 6 included in the compressed data, the timing data (3 CLK) between them, and each sample data. Determined based on the polarity data (+, 10, 1) of the differential value at the ring point.
- the average differential data value per clock (D 6 — Calculate S 3) / 3). Then, a value obtained by multiplying the average differential data value by the polarity (+, +, —) of the differential value at each sampling point is sequentially added to the normal sample data S 3, thereby obtaining The amplitude data values Q4 to Q6 at the three sampling points are obtained. Then, the normal sample data S7 at the next sampling point is directly used as the amplitude data Q7 of the decompressed data.
- the average differential data value per clock ( D 1 1 — S 7) / 4) is calculated. Then, a value obtained by multiplying the average differential data value by the polarity of the differential value at each sampling point ( ⁇ , +, tens, +) is sequentially added to the regular sample J data S 7. Then, the amplitude data values Q 8 to Q 11 at the four sampling points are obtained. Then, the normal sample data S 12 at the next sampling point is left as it is. Adopted as the amplitude data Ql2 of the decompressed data.
- the amplitude data values Ql3 to Q20 at each sampling point are obtained. Then, by interpolating between the amplitude data values Q1 to Q20 at each sampling point obtained as described above, interpolation data having a waveform as shown in FIG. 7 is obtained. Further, the generated interpolation data is subjected to a DZA conversion process, converted into an analog signal and output.
- FIG. 8 is a block diagram illustrating a functional configuration example of a compression device according to the second embodiment that implements the above-described compression method.
- components denoted by the same reference numerals as those shown in FIG. 4 have the same functions, and thus redundant description will be omitted here.
- the linear compression section 21 calculates the absolute value of the differential value calculated for each sampling point by the differentiating section 4, and sequentially adds each of the absolute values for each sampling point. At this time, an error determination as shown in Fig. 2 is performed during the addition process, and the data value on a straight line connecting the data of the two sampling points and the data value on the same sampling point as the data value on the straight line are determined. Sampling points at which the error from the differential total data value is equal to or less than a desired value are sequentially detected as sampling points. Furthermore, for the sampling point immediately after the detected sample point, processing is performed to use regular sample data.
- the differential total data calculation / linear compression section 21 generates data D 1 to D 20 having waveforms as indicated by the dashed line in FIG. this Then, the differential sum data at each sample point constituting the compressed data, the normal sample data at the normal data points, the timing data representing the time interval therebetween, and the data representing the polarity of the differential value at each sampling point are obtained.
- the compression device and the decompression device according to the second embodiment configured as described above are also configured by, for example, a computer system having a CPU or an MPU, a ROM, a RAM, and the like, and all or part of the functions thereof (for example, The differentiator 4 of the compressor, the differential total data calculation, the linear compressor 21 and the blocking unit 7, the linear decompressor 11 of the decompressor, and the interpolation processor 13) are stored in the ROM and RAM described above. It is realized by running a program. Further, the compression device and the decompression device according to the present embodiment configured as described above can be configured as hardware by combining logic circuits.
- the second embodiment As described above in detail, in the second embodiment, as in the first embodiment, it is possible to improve the quality of data reproduced by decompression while achieving a high compression rate. In addition, even when compressing a high-frequency signal, the number of detected sample points can be reduced as much as possible, and a high compression ratio can be realized. In addition, since compression and decompression can be performed as they are on the time axis, processing is not complicated, and the configuration can be simplified. Also, real-time operation of compression and decompression can be realized.
- a sampling point at which the error from the original data does not become larger than a desired value even when linear interpolation is performed on the differential total data is detected as a sampling point, and the sampling point (original data) immediately after the sampling point is detected. -The point where the error from the evening is larger than the desired value) It uses pull data.
- the accumulated error can be reduced as compared with the case where the differential absolute value is unconditionally added to a predetermined threshold value as in the first embodiment, and the data reproduced by decompression can be reduced. Quality can be further improved.
- the polarity data of the differential value at the point is obtained as compressed data
- the average differential value data per clock is obtained on the decompression side from the differential total data and the timing data at each sample point included in the compressed data.
- the average differential value data may be obtained on the compression side to be a part of the compressed data.
- the data amount of each individual can be compressed as compared with a case where the differential total data itself is used as compressed data, and the overall compression ratio can be further increased. Also, on the decompression side, the calculation for obtaining the average differential value data is not required, and the calculation load can be reduced and the reproduction time can be shortened.
- the sampling point next to the sample point detected by the error determination performed for each predetermined clock range is extracted as the normal data sunset point. It is not limited to. For example, processing is performed without imposing a restriction that the time interval between two data selected when detecting discrete sample points is within a predetermined range. Then, sampling points immediately before the sampling point where the error exceeds a desired value are sequentially detected as sampling points, and sampling points where the error exceeds a desired value are extracted as normal data points. May be. In this case, the number of sampling points can be further reduced, and the compression ratio can be further increased, while suppressing the cumulative error from increasing due to the input of regular sample data.
- subsequent differential total data is calculated starting from the certain normal sample data (for example, the normal sample data S 3 Then, the differential total data D4 to D6 are obtained from the starting point, the normal sample data S7 is obtained again, and then the differential total data D8 to D11 are obtained from the starting point.
- the present invention is not limited to such an algorithm.
- the differential sum data at all the sampling points are added, and the differential sum data is simply obtained.
- the above-described error determination may be performed on the differential total data obtained in this way, and the sample points and the normal sample data may be obtained in order.
- the algorithm on the compression side can be simplified to reduce the processing load.
- the average differential value data as a part of the compressed data instead of the differential total data at each sample point, a high compression ratio can be maintained.
- the interpolation processing unit 13 performs linear interpolation between digital data Q1 and Q20, but the interpolation calculation is not limited to this example.
- a curve interpolation process using a predetermined sampling function may be performed.
- the interpolation processing described in Japanese Patent Application No. 11-173732, filed earlier by the present applicant may be performed. In this case, a very analog waveform can be obtained by interpolation itself, so the D / A converter 14 and LPF 15 at the subsequent stage of the interpolation processor 13 are used. It can be unnecessary.
- FIG. 9 is a block diagram illustrating a functional configuration example of a compression device according to the third embodiment. Note that in FIG. 9, components denoted by the same reference numerals as those illustrated in FIGS. 4 and 8 have the same functions, and thus redundant description will be omitted.
- an A / D converter 32 and a D-type flip-flop are used instead of the AD converter 2 and the D-type flip-flop 3 shown in FIGS. 33 and a linear compression / expansion processing section 41 and a downsampling section 42.
- the AZD converter 32 and the D-type flip-flop 33 are functionally the same as the A / D converter 2 and the D-type flip-flop 3 shown in FIGS. However, it differs in that it operates according to a clock with a frequency of 6 fck, which is six times the reference frequency. That is, in the third embodiment, the data to be compressed is oversampled six times by using the A / D converter 32 and the D-type flip-flop 33.
- the linear compression / decompression processing unit 41 performs linear compression processing on the sample data of each oversampled sampling point output from the D-type flip-flop 33 according to the algorithm shown in FIG. At the same time, the original data is reproduced by performing linear expansion processing on the compressed data obtained thereby.
- the compressed data obtained in this case consists only of the amplitude data at each sample point and the timing data representing the time interval between the sample points.
- Sampling point Is a sampling point where the error between each data value on a straight line connecting two sample data and each sample data value at the same sampling point as the data value on the straight line is less than or equal to a desired value.
- the sampling point having the longest time interval within the range of the predetermined clock is detected from the reference sample data.
- such linear expansion processing of compressed data simply involves linear interpolation between amplitude data at each sample point of the compressed data at a time interval indicated by timing data. That is, based on the input compressed data (a set of the amplitude data and the timing data), an interpolation operation for linearly interpolating between the amplitude data of successive sample points is sequentially performed, so that the amplitude data between the individual amplitude data is obtained. Generate interpolation data for interpolating.
- the down-sampling section 42 down-samples the data output from the linear compression / expansion processing section 41 in accordance with the original reference frequency fck clock. In this way, before performing the linear compression-expansion processing before the differentiation processing by the differentiator 4, the linear compression / expansion processing is performed on the data oversampled to 6 times the frequency, and the result is converted to the original frequency. By downsampling, the waveform of the original data before compression is hardly distorted. Therefore, only unnecessary high frequency components can be removed.
- a silence processing unit 43 is provided next to the downsampling unit 42.
- the silence processing unit 43 regards the sample data as silence, and sets the data value to “ Perform the process of replacing with "0" and outputting. As a result, the compression ratio is further improved.
- the differential sum data calculation unit 35 sequentially adds the absolute differential values at all sampling points without returning to the normal sample data as shown in FIG. Calculate the differential total data at the running point.
- the linear compression unit 36 performs an error determination on the differential total data obtained in this manner according to the algorithm shown in FIG. 2 and sequentially obtains sample points and normal data points.
- the normal sample data at each normal data point, the timing data representing the time interval between the normal data point and the sample point, and the average differential value data per clock between the normal data point and the sample point are obtained. Are obtained as part of the compressed data.
- the blocking section 37 includes the normal sample data, timing data, and average differential value data generated by the linear compression section 36, and the polarity data of the differential value at each sampling point obtained by the differential section 4. One night is appropriately blocked and output as compressed data.
- the output compressed data is transmitted to a transmission medium, for example, or recorded on a recording medium such as a nonvolatile memory.
- the configuration of the decompression device corresponding to the compression device according to the third embodiment described above is the same as that shown in FIG. However, the operation contents of the linear extension unit 11 are different from those of the first and second embodiments. That is, the first and In the second embodiment, the average differential value data is calculated by the linear decompression unit 11 using the differential total data included in the compressed data, whereas in the third embodiment, the average differential value is calculated. Since the data is calculated on the compression device side and output as compressed data, this calculation is unnecessary in the linear decompression unit 11.
- the compression device and the decompression device according to the third embodiment configured as described above are also configured by, for example, a computer system having a CPU or an MPU, a ROM, a RAM, and the like, and all or some of the functions thereof (for example, Differentiator 4, Differential total data calculator 35, Linear compressor 36, Linear compression / expansion processor 41, Silence processor 43, Linear expander 11, Expander 1 and interpolation processor 1 3) are realized by the operation of the programs stored in the ROM and RAM described above. Further, the compression device and the decompression device according to the present embodiment configured as described above can be configured as hardware by combining logic circuits. FIGS.
- FIG. 10 to 12 are diagrams showing waveforms and characteristics of a certain analog signal (human voice) and a reproduced analog signal reproduced by applying the compression / expansion processing according to the third embodiment.
- FIG. 10 is a diagram showing waveforms of an original analog signal (input data) before compression and a reproduced analog signal (output data) obtained by compressing and expanding the analog signal.
- FIG. 11 is a partially enlarged view of the waveform shown in FIG.
- FIG. 12 is a diagram showing the correlation between input data and output data.
- the accumulated error can be reduced as compared with the case where the differential absolute value is unconditionally added to a predetermined threshold value as in the first embodiment.
- the quality of data reproduced by decompression can be further improved.
- each sample data itself is subjected to linear compression as shown in FIG. 2, and the compressed data is expanded by linear interpolation.
- compressed data can be obtained by performing the same processing as in the first embodiment or the second embodiment, and based on the compressed data. The quality of data reproduced by decompression can be further improved.
- the average differential value data is obtained on the compression side and is used as a part of the compressed data.
- the average differential value data may be obtained on the decompression side.
- the differential absolute values at all sampling points are simply added without returning to the normal sample points at the normal data points. To data You may make it return.
- the number of bits of the timing data is set to 3 bits, and the error determination is performed by drawing a straight line within 6 clocks from the reference sample data.
- the present invention is not limited to this example.
- the predetermined range for performing the error determination may be 7 clocks.
- the number of bits of the timing data may be set to 4 bits or more, and the predetermined range for performing error determination by drawing a straight line from the reference sample data may be set to 8 clocks or more. By doing so, it is possible to further increase the compression ratio.
- the number of bits of the timing data or a predetermined range for performing the error determination may be arbitrarily set as a parameter.
- the allowable value of the error for example, 64, 128, 256, 384, 512 can be used. If the tolerance of the error is reduced, compression and decompression can be realized with emphasis on the reproducibility of the reproduced analog signal. Also, if the tolerance of the error is increased, compression / decompression with emphasis on the compression ratio can be realized. Note that the error tolerance may be arbitrarily set as a parameter.
- the error tolerance may be a function of the data amplitude. For example, the error tolerance may be increased when the amplitude is large, and the error tolerance may be decreased when the amplitude is small. Where the amplitude is large, the error is not noticeable even if the error becomes large to some extent, and does not significantly affect the sound quality. Therefore, by dynamically changing the error tolerance value as a function of the data amplitude in this way, it is possible to further increase the compression ratio while maintaining extremely good sound quality of the reproduced data.
- the error tolerance is made a function of frequency, for example, increasing the error tolerance at a high frequency and decreasing the error tolerance at a low frequency You may do it.
- the sample points detected as having a small error tolerance are detected. May increase, and a high compression ratio may not be achieved.
- dynamically increasing the error allowance in the high frequency part it is possible to further increase the compression ratio while maintaining the sound quality of the reproduced data as a whole extremely good.
- the error tolerance may be dynamically changed as a function of both the data amplitude and the frequency.
- the deinterlacing value is oversampled by a factor of 6 in the interpolation process on the decompression side.
- the oversampling is not limited to the factor of 6, but may be an arbitrary factor. It is possible.
- the compression / expansion methods according to the first to third embodiments described above can be realized by any of the hardware configuration, the DSP, and the software, as described above.
- the compression device and the decompression device according to the present embodiment are actually configured by a combination CPU or MPU, RAM, ROM, etc., and are stored in RAM or ROM. It can be realized by the operation of the stored program.
- the present invention can be realized by recording a program that causes a computer to perform the functions of the present embodiment on a recording medium such as a CD-ROM, and reading the program into the computer.
- a recording medium for recording the above program besides CD-ROM, use a floppy disk, hard disk, magnetic tape, optical disk, magneto-optical disk, DVD, nonvolatile memory card, etc. Can be.
- the computer executes the supplied program to Not only the functions of the above-described embodiment are realized, but also the program is executed in cooperation with an operating system (OS) running on a computer or other application software.
- OS operating system
- the function of the form is realized, or all or a part of the processing of the supplied program is performed by the function expansion port or the function expansion unit of the computer, thereby realizing the function of the above-described embodiment.
- Such a program is also included in the embodiment of the present invention.
- the present invention even when the data of each sampling point is reproduced from the average differential value data between sample points and the polarity data of the differential value during the decompression process, an error from the original data is desired.
- the sampling points that do not become larger than the value are detected as sampling points, and the differential sum data at the discrete sampling points detected in this way, the average differential value data per unit time between the sampling points, and the time interval between the sampling points are calculated. Since only the represented timing data, the polarity data of the differential value of each sampling point, and the like are obtained as compressed data, the quality of data reproduced by decompression can be significantly improved while achieving a high compression ratio.
- the sample data is differentiated and the absolute values are sequentially added.
- the present invention when compressing a signal on the time axis, it is possible to perform processing on the time axis without performing time / frequency conversion and performing processing on the frequency axis. Also, when decompressing the data compressed in this way, the processing can be performed on the time axis. In particular, on the decompression side, high-precision decompression data that is almost the same as the original data before compression is reproduced only by performing extremely simple processing such as adding the polarities to the averaged differential values and sequentially adding them and interpolation processing. can do.
- each data value on a straight line connecting data of two sampling points, and each differential sum at the same sampling point as each data value on the straight line are sequentially set as the sampling points of the compressed data.
- a data value on a straight line connecting data of two sampling points, and a differential total data value at the same sampling point as the data value on the straight line Is the sampling point at which the error from the sampling point is equal to or less than the desired value, and the sampling point immediately before the sampling point at which the above error exceeds the desired value.
- the differential total data at each sample point itself can be used as compressed data.
- the amount of individual data can be reduced, and the compression ratio can be further increased.
- regular sample data is compressed at a sampling point where the value of the differential total data exceeds a predetermined threshold.
- a predetermined threshold By adopting it as a part of the data, it is possible to prevent the value of the differential total data included as a part of the compressed data from becoming larger than a predetermined threshold, and to reduce the amount of individual data.
- the linear interpolation is performed between two pieces of differential total data, the difference from the original data exceeds a desired value.
- regular sample data as part of the compressed data at the printing point, insert the regular sample data for each part where a cumulative error may occur and cut off the occurrence of the cumulative error.
- data compression is performed by obtaining differential total data as described above.
- data compression can be performed after removing unnecessary high-frequency components that cause noise in advance.
- the compression ratio can be further improved, and the quality of data reproduced by decompression based on the compressed data can be further improved.
- the present invention is useful for providing an entirely new compression / decompression method that achieves both an improvement in compression rate and an improvement in the quality of reproduced data.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Image Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/239,080 US20030133505A1 (en) | 2001-01-19 | 2002-01-17 | Compression method and device, decompression method and device, compression/ decompression system, recording medium |
EP02715777A EP1365516A4 (en) | 2001-01-19 | 2002-01-17 | "COMPRESSION METHOD AND DEVICE, DECOMPRESSION METHOD AND DEVICE, COMPRESSION / DECOMPRESSION SYSTEM, RECORD MEDIUM" |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2001-10892 | 2001-01-19 | ||
JP2001010892A JP2002217740A (ja) | 2001-01-19 | 2001-01-19 | 圧縮方法及び装置、伸長方法及び装置、圧縮伸長システム、記録媒体 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2002058244A1 true WO2002058244A1 (en) | 2002-07-25 |
Family
ID=18878118
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2002/000277 WO2002058244A1 (en) | 2001-01-19 | 2002-01-17 | Compression method and device, decompression method and device, compression/decompression system, recording medium |
Country Status (7)
Country | Link |
---|---|
US (1) | US20030133505A1 (ja) |
EP (1) | EP1365516A4 (ja) |
JP (1) | JP2002217740A (ja) |
KR (1) | KR20030005231A (ja) |
CN (1) | CN1455987A (ja) |
TW (1) | TW522660B (ja) |
WO (1) | WO2002058244A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1420346A3 (de) * | 2002-10-14 | 2017-09-13 | Deutsche Telekom AG | Verfahren zur zweidimensionalen Darstellung, Interpolation und zur Kompression von Daten |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001095496A1 (fr) * | 2000-06-06 | 2001-12-13 | Sakai, Yasue | Procede et appareil de compression, procede et appareil d'expansion, systeme de compression expansion |
JP2002312000A (ja) | 2001-04-16 | 2002-10-25 | Sakai Yasue | 圧縮方法及び装置、伸長方法及び装置、圧縮伸長システム、ピーク検出方法、プログラム、記録媒体 |
US8442441B2 (en) | 2004-12-23 | 2013-05-14 | Qualcomm Incorporated | Traffic interference cancellation |
JP4692388B2 (ja) * | 2006-05-24 | 2011-06-01 | ソニー株式会社 | データ処理装置およびデータ処理方法 |
US8787454B1 (en) | 2011-07-13 | 2014-07-22 | Google Inc. | Method and apparatus for data compression using content-based features |
CN102961132B (zh) * | 2012-11-26 | 2015-02-18 | 重庆德领科技有限公司 | 一种无线表面肌电信号拾取装置及前端压缩方法 |
WO2020004497A1 (ja) | 2018-06-28 | 2020-01-02 | 富士フイルム株式会社 | 光学素子及び光学素子の製造方法 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5996513A (ja) * | 1982-11-24 | 1984-06-04 | Nippon Gakki Seizo Kk | 波形の記録及び再生方法 |
JPH03192400A (ja) * | 1989-12-22 | 1991-08-22 | Gakken Co Ltd | 波形情報処理装置 |
JP2001136073A (ja) * | 1999-11-02 | 2001-05-18 | Sakai Yasue | 圧縮方法及び装置、圧縮伸長システム、記録媒体 |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2491703B1 (fr) * | 1980-10-03 | 1988-04-29 | Thomson Csf | Dispositif de compression et dispositif de decompression temporelle de donnees et systeme de transmission comportant au moins l'un de ces dispositifs |
JP2884163B2 (ja) * | 1987-02-20 | 1999-04-19 | 富士通株式会社 | 符号化伝送装置 |
US4897855A (en) * | 1987-12-01 | 1990-01-30 | General Electric Company | DPCM system with adaptive quantizer having unchanging bin number ensemble |
US5222189A (en) * | 1989-01-27 | 1993-06-22 | Dolby Laboratories Licensing Corporation | Low time-delay transform coder, decoder, and encoder/decoder for high-quality audio |
US5175769A (en) * | 1991-07-23 | 1992-12-29 | Rolm Systems | Method for time-scale modification of signals |
US5285498A (en) * | 1992-03-02 | 1994-02-08 | At&T Bell Laboratories | Method and apparatus for coding audio signals based on perceptual model |
CA2090052C (en) * | 1992-03-02 | 1998-11-24 | Anibal Joao De Sousa Ferreira | Method and apparatus for the perceptual coding of audio signals |
JP2900719B2 (ja) * | 1992-08-24 | 1999-06-02 | 日本電気株式会社 | 音声コーデック処理方法 |
DE69328399T2 (de) * | 1992-09-30 | 2000-10-19 | Hudson Soft Co Ltd | Sprachdaten-Verarbeitung |
JPH06232826A (ja) * | 1993-02-08 | 1994-08-19 | Hitachi Ltd | 音声差分pcmデータ伸長方法 |
US5920842A (en) * | 1994-10-12 | 1999-07-06 | Pixel Instruments | Signal synchronization |
US6360022B1 (en) * | 1997-04-04 | 2002-03-19 | Sarnoff Corporation | Method and apparatus for assessing the visibility of differences between two signal sequences |
US6249766B1 (en) * | 1998-03-10 | 2001-06-19 | Siemens Corporate Research, Inc. | Real-time down-sampling system for digital audio waveform data |
US6122619A (en) * | 1998-06-17 | 2000-09-19 | Lsi Logic Corporation | Audio decoder with programmable downmixing of MPEG/AC-3 and method therefor |
US6112170A (en) * | 1998-06-26 | 2000-08-29 | Lsi Logic Corporation | Method for decompressing linear PCM and AC3 encoded audio gain value |
JP2002312000A (ja) * | 2001-04-16 | 2002-10-25 | Sakai Yasue | 圧縮方法及び装置、伸長方法及び装置、圧縮伸長システム、ピーク検出方法、プログラム、記録媒体 |
-
2001
- 2001-01-19 JP JP2001010892A patent/JP2002217740A/ja active Pending
-
2002
- 2002-01-17 KR KR1020027012318A patent/KR20030005231A/ko not_active Application Discontinuation
- 2002-01-17 US US10/239,080 patent/US20030133505A1/en not_active Abandoned
- 2002-01-17 CN CN02800068A patent/CN1455987A/zh active Pending
- 2002-01-17 WO PCT/JP2002/000277 patent/WO2002058244A1/ja not_active Application Discontinuation
- 2002-01-17 EP EP02715777A patent/EP1365516A4/en not_active Withdrawn
- 2002-01-18 TW TW91100790A patent/TW522660B/zh not_active IP Right Cessation
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5996513A (ja) * | 1982-11-24 | 1984-06-04 | Nippon Gakki Seizo Kk | 波形の記録及び再生方法 |
JPH03192400A (ja) * | 1989-12-22 | 1991-08-22 | Gakken Co Ltd | 波形情報処理装置 |
JP2001136073A (ja) * | 1999-11-02 | 2001-05-18 | Sakai Yasue | 圧縮方法及び装置、圧縮伸長システム、記録媒体 |
Non-Patent Citations (1)
Title |
---|
See also references of EP1365516A4 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1420346A3 (de) * | 2002-10-14 | 2017-09-13 | Deutsche Telekom AG | Verfahren zur zweidimensionalen Darstellung, Interpolation und zur Kompression von Daten |
Also Published As
Publication number | Publication date |
---|---|
US20030133505A1 (en) | 2003-07-17 |
TW522660B (en) | 2003-03-01 |
EP1365516A4 (en) | 2006-10-18 |
JP2002217740A (ja) | 2002-08-02 |
CN1455987A (zh) | 2003-11-12 |
KR20030005231A (ko) | 2003-01-17 |
EP1365516A1 (en) | 2003-11-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6785644B2 (en) | Alternate window compression/decompression method, apparatus, and system | |
US6657567B2 (en) | Compressing method and device, decompression method and device, compression/decompression system, and recorded medium | |
US20050143981A1 (en) | Compressing method and apparatus, expanding method and apparatus, compression and expansion system, recorded medium, program | |
US5594443A (en) | D/A converter noise reduction system | |
US20040027260A1 (en) | Method and apparatus for compression, method and apparatus for decompression, compression/ decompression system, record medium | |
JPH10313251A (ja) | オーディオ信号変換装置及び方法、予測係数生成装置及び方法、予測係数格納媒体 | |
WO2002058244A1 (en) | Compression method and device, decompression method and device, compression/decompression system, recording medium | |
US7224294B2 (en) | Compressing device and method, decompressing device and method, compressing/decompressing system, program, record medium | |
US20050075743A1 (en) | Digital signal processing method, learning method, apparatuses for them, and program storage medium | |
JP5471603B2 (ja) | 量子化ビット数拡張方法および量子化ビット数拡張装置 | |
JPH11144382A (ja) | 符号情報処理方法及び装置、符号情報の記録媒体への記録方法 | |
JP2001069010A (ja) | 圧縮方法及び装置、伸長方法及び装置、圧縮伸長システム、記録媒体 | |
US5384763A (en) | Signal compression recording and reproducing method | |
US20070096961A1 (en) | Signal processing device | |
JPH08172363A (ja) | 信号伸張装置及び方法 | |
JPH0823262A (ja) | サンプリング周波数変換装置 | |
JPS62217730A (ja) | 周波数圧伸方式 | |
JPH08314498A (ja) | 適応変換符号化オーディオ信号復号器 | |
JPH05291859A (ja) | ディジタル・ノイズ・リダクション装置 | |
JP2001223595A (ja) | 圧縮方法及び装置、伸長方法及び装置、記録媒体 | |
JP2001265393A (ja) | 音声録音再生装置 | |
JPH11501785A (ja) | ディジタル・アナログ・コンバータ | |
JPH08172362A (ja) | 信号伸張装置及び方法 | |
JPH06124100A (ja) | 音声録音再生装置 | |
JPH09153771A (ja) | 信号処理装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CN KR US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 028000684 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1020027012318 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10239080 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 1020027012318 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2002715777 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2002715777 Country of ref document: EP |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: 2002715777 Country of ref document: EP |