JP5640305B1 - Data volume reduction device - Google Patents

Abstract

The present invention provides a data amount reduction device capable of reducing the data amount of data that can be compressed to obtain high compression efficiency and restoring the original data with high restoreability. A data amount reduction device 1 converts pre-conversion data 10 having a first value and a second value into sum data as a sum of a first value and a second value, and a first value. And a data format conversion unit 2 for converting into format conversion data composed of difference data as a difference between the second values. In addition, the data amount reduction device 1 includes a correlation content management unit 3 in which processing information based on correlation information of sum data and difference data constituting the format conversion data is recorded, and the sum data and difference data based on the processing information. A data deletion / addition unit 4 is provided that deletes at least one of all or some of the values to generate reduced data in which the total data amount is reduced. [Selection] Figure 1

Description

  The present invention relates to a technique for reducing the data amount of various data including audio data and image data.

  A Hadamard transform is a technique for compressing the data amount of various data. This is a kind of orthogonal transform and is used in lossless compression of data (see, for example, Patent Document 1). As an application of this Hadamard transform, in the pre-conversion data consisting of two values (for example, audio data consisting of a right channel value and a left channel value), sum data that is the sum of one value and the other value And the difference data value, which is the difference between one value and the other value, can statistically narrow the distribution region of the converted data values and obtain a high compression rate. In this method, in particular, when the correlation between the original two values is high, the range of data distribution is narrowed, and high compression efficiency can be obtained.

JP 2011-257806 A

  However, in the invention described in Patent Document 1, depending on the mode of conversion, the range of values that can be taken after conversion is larger than that before conversion. For example, when 8-bit data from 0 to 255 is converted into sum data and difference data as described above, there is a possibility that after conversion, 9-bit data including the code is obtained. For this reason, there is a problem that the data after conversion becomes larger than that before conversion, and the compression efficiency of the compressed data after compression compared to the data before conversion cannot be sufficiently high.

  The present invention has been made in view of such problems, and can reduce the data amount of data that can be compressed to obtain high compression efficiency, and can also reduce the amount of data with high recoverability. It is an object of the present invention to provide a data amount reduction device that can restore the original data.

  In order to solve such a problem, the invention according to claim 1 is characterized in that pre-conversion data having a first value and a second value is a sum as a sum of the first value and the second value. A data amount reduction device comprising a data format conversion means for converting data into format conversion data consisting of difference data as a difference between the first value and the second value, and comprising the format conversion data Processing content management means in which processing information for determining the processing content for the sum data or the difference data is recorded based on the correlation information of the sum data and the difference data, and the sum based on the processing information. Data reduction means is provided for generating reduced data in which all or a part of at least one of the data and the difference data is deleted to reduce the total data amount.

  The invention according to claim 2 is the configuration according to claim 1, wherein the sum data and the difference data constituting the format conversion data are configured by binary numbers, and the processing information is the sum data and the If one of the difference data is even, the other is even, and the sum data and the difference data match based on the correlation information that one of the sum data and the difference data is odd and the other is odd. It is processing information for deleting the bottom digit, and the data reduction means deletes the bottom digit for one of the sum data and the difference data based on the processing information. It is characterized by.

  According to a third aspect of the present invention, in addition to the configuration of the first aspect, the processing information is a maximum value that can be taken by the sum data or a minimum value that can be taken, and the difference data is 0 in a specific case. The sum data is processing information for deleting the difference data when the sum data is in the specific case, and the data reduction means is based on the processing information when the sum data is in the specific case. The difference data is deleted.

  According to a fourth aspect of the present invention, in addition to the configuration of the first aspect, the processing information is a maximum value that can be taken by the difference data or a minimum value that can be taken, and the sum data is 0 in a specific case. Based on the correlation information, the sum data is information on processing for deleting the sum data when the specific data is in the specific case, and the data reduction means is configured to determine whether the difference data is based on the processing information. The sum data is deleted.

In addition to the configuration described in any one of claims 1 to 4, the invention described in claim 5 includes a condition for generating reduced data in part or all of sum data or difference data constituting reduced data. A data adding means for adding a specific value based on the above and restoring the original format conversion data is provided.

  According to a sixth aspect of the present invention, in addition to the configuration according to any one of the first to fifth aspects, the pre-conversion data is image data or audio data to be subjected to lossless compression. .

  According to the first aspect of the present invention, reduced data in which the entire data amount is reduced is generated based on the processing information recorded in the processing content management means. Since this processing information is based on the correlation information of the sum data and the difference data constituting the format conversion data, the sum data or the difference data is deleted from the processing information for the highly correlated portion of the sum data and the difference data. , Can reduce the amount of data. If the data with the reduced data amount is compressed, compressed data with high compression efficiency can be generated for the pre-conversion data. Further, the deleted part of the sum data or difference data can be easily restored based on the data remaining after the deletion and the correlation between the sum data and the difference data. As a result, the amount of data that can be compressed can be reduced to obtain high compression efficiency, and the data with the reduced amount of data can be restored to the original data with high recoverability.

  According to the second aspect of the present invention, in the binary sum data and difference data, the bottom digit is deleted from the sum data and difference data composed of binary numbers based on the processing information. Since the highly correlated part is deleted, an excessive increase in the amount of data that can be compressed can be suppressed, and the data from which the value has been deleted can be restored to the original data with a higher degree of restoration.

  According to the third aspect of the invention, based on the processing information, the difference data that is uniquely determined when the sum data is the maximum value or the minimum value that can be taken is deleted. While suppressing an excessive increase in the amount of data, it is possible to restore the data from which the value has been deleted to the original data with a higher degree of restoration.

  According to the fourth aspect of the present invention, the sum data that is uniquely determined when the difference data is the maximum value or the minimum value that can be taken is deleted based on the processing information. While suppressing an excessive increase in the amount of data, it is possible to restore the data from which the value has been deleted to the original data with a higher degree of restoration.

According to the invention described in claim 5, the reduction data generated by the data amount reduction device according to claims 1 to 4 is added with a specific value to a part or all of sum data or difference data, By restoring the original format conversion data, the original format conversion data is restored from the reduced data, and the present invention can be used as a decoder.

  According to the invention described in claim 6, the sum data and the difference data obtained by converting the first value and the second value of the image data and the sound data are based on the correlation information of the sum data and the difference data. By generating correlation data, it is possible to generate reduction data that can be compressed using image data and audio data that can easily obtain highly correlated data as the first and second values. Or can be reconverted to the original pre-conversion data without degrading the data. Thereby, high compression efficiency can be obtained for image data and audio data, and data with a reduced data amount can be restored to the original image data and audio data with high recoverability.

It is a functional block diagram which shows the whole structure of the data amount reduction apparatus which concerns on embodiment of this invention. It is a conceptual diagram of the correlation information and process information which were recorded on the correlation content management part of the data amount reduction apparatus same as the above. It is a conceptual diagram which shows the 1st process of a data amount reduction apparatus same as the above. It is a conceptual diagram which shows the 2nd process of a data amount reduction apparatus same as the above. It is a flowchart which shows the procedure of the reduction process of the data amount reduction apparatus same as the above, and a compression process. It is a flowchart which shows the procedure of the decoding process of the data amount reduction apparatus same as the above, and an addition process.

[Basic configuration]
1 to 6 show an embodiment of the present invention.

  FIG. 1 is a functional block diagram showing the overall configuration of the data amount reducing apparatus according to this embodiment.

  The data amount reduction apparatus 1 shown in FIG. 1 is an apparatus for reducing the data amount of various data represented by, for example, image data and audio data, adding the reduced data, and compressing and decoding the data.

  The data amount reduction device 1 shown in FIG. 1 includes at least one CPU (not shown), and the operation of the data amount reduction device 1 is controlled by processing of instructions and data in the CPU (not shown). Furthermore, the data amount reduction device 1 includes a RAM, a ROM, an EEPROM, a magnetic disk (all not shown), and the like. The data amount reduction device 1 may be provided with various configurations for sequence control instead of a CPU (not shown), and the operation may be controlled by the operation of the configuration.

  The data amount reduction device 1 includes a data format conversion unit 2 as a “data format conversion unit”, a correlation content management unit 3 as a “processing content management unit”, a data reduction unit as a “data reduction unit”, and a “data addition unit”. An addition unit 4 and a compression / decoding unit 5 are included. The data format conversion unit 2, the correlation content management unit 3, the data reduction / addition unit 4, and the compression / decoding unit 5 may be configured as hardware or realized by executing various programs in a CPU (not shown). Functional means may be used.

  The data format conversion unit 2 performs data format conversion. Specifically, conversion from pre-conversion data 10 (described later) to format conversion data (described later) shown in FIG. 1 or reverse conversion from format conversion data to pre-conversion data is performed.

  The correlation content management unit 3 includes data of correlation information (described later) indicating the correlation between sum data (described later) and difference data (described later) constituting the format conversion data, and each sum data, difference data, and reduction data. Processing information indicating processing contents for (described later) is recorded. The correlation content management unit 3 may be configured to be recorded on a recording medium that can be read by a command such as a CPU, or may be configured as a functional unit or hardware logic based on a program.

  The data reduction / addition unit 4 deletes all or part of the sum data and difference data based on the processing information of the correlation content management unit 3 to reduce the total amount of format conversion data Data (described later) is generated. Also, based on the processing information, data is added to the reduced data to generate sum data or difference data.

  The compression / decoding unit 5 performs compression processing of the format conversion data and the reduced data to generate the compressed data 20 illustrated in FIG. Further, the compressed data 20 is decoded (restored) to generate format conversion data and reduced data before compression. The compression / decoding unit 5 compresses and decodes data by lossless compression.

[Operating Principle 1: First Process]
2 and 3 show the theoretical basis and operation principle of the first processing of the data amount reduction apparatus 1 according to this embodiment.

  FIG. 2 is a conceptual diagram of correlation information and processing information recorded in the correlation content management unit 3 of the data amount reduction device 1. The correlation information shown in the figure indicates the theoretical basis for data reduction by the data amount reduction device 1. Further, the processing information shown in the figure determines specific contents of processing performed in the data amount reduction apparatus 1 based on the correlation information.

  For example, the pre-conversion data 10 is audio data, and the left data as data of the signal value (hereinafter referred to as “left signal value”) of the left channel as the “first value” (hereinafter referred to as “left signal value”). And right data as data of a right channel signal value (at the same specific time point) (hereinafter referred to as “right signal value”) as “second value”.

In this case, the sum data as the sum of the left signal value and the right signal value, and the value obtained by subtracting the right channel signal value from the left channel signal value (the right channel signal value minus the left channel signal value). As shown in the first correlation information 31 in FIG. 2, there is the following relationship with the difference data as the value of
(Principle 1) If the left signal value and the right signal value are odd numbers, the sum data value and the difference data value are even numbers.
(Principle 2) If the left signal value and the right signal value are even numbers, the sum data value and the difference data value are even numbers.
(Principle 3) If the left signal value is odd and the right signal value is even, both the sum data value and the difference data value are odd numbers.
(Principle 4) If the left signal value is even and the right signal value is odd, the sum data value and the difference data value are odd numbers. Fourth Principle 314
That is, if the sum data value is an even number, the difference data value is also an even number. If the sum data value is an odd number, the difference data value is also an odd number.

  Part of this principle is specifically shown in the schematic diagram of FIG. The pre-conversion data 10 shown to (a) of the figure has the left data 10a and the right data 10b. The left data 10a and the right data 10b are 8-bit unsigned integer data, respectively. The value of the left data 10a is “208”, which is an even number. The value of the right data 10b is “96”, which is an even number.

  The sum of the left data 10a and the right data 10b is the sum data 11a of the format conversion data 11 shown in FIG. A difference obtained by subtracting the right data 10b from the left data 10a is difference data 11b of the format conversion data 11 shown in FIG. As shown in the figure, in the sum data 11a and the difference data 11b, the values of the lowest digits 11c and 11d are both “0” and both are even numbers. This is in line with the second principle 312.

  In the format conversion data 11, in the coordinate space (for example, the coordinate space in which the sum data is plotted on the horizontal axis and the difference data is plotted on the vertical axis), the value range distributed on the difference data 11b side is smaller than the sum data 11a side. (Narrow). Then, compared to the pre-conversion data 10, the compression performance when forming the compressed data 20 using entropy coding is improved and the compression efficiency is increased by the amount of the range where the approximate values are distributed.

  On the other hand, the possible values of the sum data 11a are 0 to 510. Nine digits are required to express this in binary. Therefore, as shown in FIG. 3B, the sum data 11a is composed of 9 bits. On the other hand, the possible values of the difference data 11b are −255 to 255. To represent this value, plus and minus signs are required. That is, in addition to the 8 bits necessary to indicate 0 to 255, for example, 1 bit is required to indicate a sign such that “0” is positive and “1” is negative. Therefore, as shown in FIG. 3B, the difference data 11b is composed of 9 bits. Thereby, the format conversion data 11 has a total of 18 bits, and the number of bits is larger than the total 16 bits of the pre-conversion data 10.

  In this embodiment, the following procedure is used to reduce the number of bits of the format conversion data 11 while the original format conversion data 11 can be easily restored.

  As shown in (b) of FIG. 3 and the second principle 312 of FIG. 2, the values of the bottom digits 11c and 11d of the sum data 11a and the difference data 11b of the format conversion data 11 are both “0”. . If this is known, even if one of the lowest digits 11c, 11d (eg, the lowest digit 11d of the difference data 11b) is deleted, the other value (eg, the first digit of the sum data 11a) is deleted. By copying and adding the lower digit 11c) together with the value, the original data (original difference data 11b) can be easily restored.

Therefore, in this embodiment, as shown in FIG. 2, the correlation content management unit 3 provides the first processing information 31a as the lowest digit of one of the sum data 11a or the difference data 11b ( For example, the processing content of deleting or adding the bottom digit 11d) of the difference data 11b is recorded. Based on this processing content, the data reduction / addition unit 4 performs processing according to the following (processing guideline 1) or (processing guideline 2).
(Processing guideline 1): When the lowest digit 11c, 11d exists in both the sum data 11a and the difference data 11b of the format conversion data 11, one of the lowest digits (for example, the first digit of the difference data 11b) Delete the lower digit 11d).
(Processing guideline 2): When only one of the sum data 11a and difference data 11b of the format conversion data 11 (for example, sum data 11a) has the bottom digit (for example, the bottom digit 11c), The value of the bottom digit 11c is copied and added to the bottom digit 11d of the other (for example, difference data 11b). Note that this processing is actually performed on the reduction data 12 described later.

  Then, the reduced data 12 shown in FIG. 3C is formed by deleting the lowest digit 11d of the difference data 11b of the format conversion data 11. In the reduction data 12, the sum data 12a is the same as the sum data 11a of the format conversion data 11 (9 bits), but the difference data 12b is deleted from the bottom value 11d of the difference data 11b of the format conversion data 11. (8 bits). Therefore, the reduction data 12 has a smaller number of bits than the format conversion data 11. In addition, the original format conversion data 11 can be easily restored by copying the lower digit 12c of the sum data 12a and adding the reduced data 12 to the lower digit 12d of the difference data 12b.

  The operating principle 1 can be applied not only to unsigned integer data but also to signed integer data. The operating principle 1 can be applied not only to 8-bit data but also to data of any number of bits such as 4 bits and 16 bits.

[Operation Principle 2: Second Process]
2 and 4 show the operation principle of the second process of the data amount reduction apparatus 1 according to this embodiment.

  For example, as in [Operation Principle 1], it is assumed that the pre-conversion data 10 is audio data and has left data and right data. Also in this case, the left data 10a and the right data 10b of the pre-conversion data 10 are assumed to be 8-bit unsigned integer data, respectively.

In this case, when the sum data is a specific value, the combination in which the difference data value is always “0” (hereinafter, the format conversion data 11 of such a combination is referred to as “specific data”). There are the following.
(Principle 5) If the left signal value and the right signal value are the maximum values that can be taken, the sum data value becomes the maximum value that the sum data can take, and the difference data value becomes “0”.
(Principle 6) If both the left signal value and the right signal value are “0”, the sum data is “0” and the difference data value is also “0”. Sixth principle 322
Part of this principle is specifically shown in the schematic diagram of FIG. The format conversion data 11 shown in FIG. 6A is generated based on the left data 10a and the right data 10b of the pre-conversion data 10 being “255” which is the maximum value of 8-bit data. It is a thing. That is, in the format conversion data 11, the sum data 11e is “510” which is the maximum value that the sum data 11e can take, and the difference data 11f is “0”. This is in line with the fifth principle 321 shown in FIG.

In this case, if it is known that the value “510” of the sum data 11e is the maximum value that the sum data 11e can take, even if the difference data 11f is deleted, as the difference data 11f based on the sum data 11e. If a value of “0” is added, the original data (original difference data 11f) can be easily restored.

Therefore, in this embodiment, as shown in FIG. 2, the correlation content management unit 3 records the processing content that the difference data 11f is added or deleted as the second processing information 32a. Based on this processing content, the data reduction / addition unit 4 performs processing based on the following (processing guideline 3) or (processing guideline 4).
(Processing guideline 3): When the sum data 11e and the difference data 11f exist in the format conversion data 11, and the sum data 11e is specific data of the maximum value “510” or “0”, the difference data 11f is delete.
(Processing guideline 4): When only the sum data 11e exists in the format conversion data 11 and the sum data 11e is specific data of the maximum value “510” or “0”, the difference data 11f is “0”. Is added. Note that this processing is actually performed on the reduction data 12 described later.

Then, the reduction data 12 shown in FIG. 4B is formed by deleting the difference data 11 f of the difference data 11 f of the format conversion data 11. In the reduction data 12, the sum data 12e is the same as the sum data 11e of the format conversion data 11 (9 bits), but the data corresponding to the difference data 12f, that is, the difference data 11f of the format conversion data 11 is deleted. Therefore, the reduction data 12 is 9 bits in total, and the number of bits is smaller than the format conversion data 11 (18 bits). Moreover, the reduced data 12 is obtained by adding “0” as the difference data 12f and copying the bottom digit 12c in consideration of the sum data 12e being the maximum value that the sum data 12e can take. If it is added to the lowest digit 12d of 12b, the original format conversion data 11 can be easily generated.

The operating principle 2 can be applied not only to unsigned integer data but also to signed integer data. The operation principle 2 can be applied not only to 8-bit data but also to data of any number of bits such as 4 bits and 16 bits. Here, for example, when the operating principle 2 is applied to 16-bit signed integer data, the specific data is “65534” and “0” which are the maximum values that can be taken.

[Processing procedure]
FIG. 5 is a flowchart showing the data reduction processing and the processing procedure of the data amount reduction apparatus 1 according to this embodiment. Hereinafter, the contents of the processing of the data amount reduction device 1 will be described with reference to FIG.

  First, a processing procedure for performing data reduction and data compression will be described.

  The data amount reduction device 1 performs data format conversion in the data format conversion unit 2 on the pre-conversion data 10 that is data to be processed (step S1). Thereby, as shown in FIGS. 3A and 3B, the data format of the pre-conversion data 10 is converted, and the format conversion data 11 is generated. That is, here, the pre-conversion data 10 composed of the left data 10a and the right data 10b is converted into the format conversion data 11 composed of the sum data 11a and the difference data 11b.

  After converting all the pre-conversion data 10 into the format conversion data 11, the data reduction / addition unit 4 performs the data reduction processing using the first processing information 31 a and the second processing information 32 a of the correlation content management unit 3. Do.

  First, the data reduction / addition unit 4 confirms whether the format conversion data 11 corresponds to specific data. Specifically, the data reduction / addition unit 4 checks whether the sum data of the format conversion data 11 corresponds to the second correlation information 32 of the correlation content management unit 3. For example, as shown in FIG. 4A, when the sum data 11e of the format conversion data 11 matches the second correlation information 32 (in this case, matches the fifth principle 321) (“ Yes "), the data reduction / addition unit 4 performs the second process (step S3). Specifically, the data reduction / addition unit 4 uses the (processing guideline 3) of the second processing information 32a, deletes the difference data 11f of the format conversion data 11, and performs the reduction shown in (b) of FIG. Data 12 is generated.

  On the other hand, when the format conversion data 11 is not specific data (“No” in step S2), the data reduction / addition unit 4 performs the first process (step S4). Specifically, the data deletion / addition unit 4 uses (processing guideline 1) of the first processing information 31a, and as shown in (a) and (b) of FIG. One bottom digit (here, the bottom digit 11d of the difference data 11b) is deleted, and the reduced data 12 shown in FIG. 3C is generated.

  The second process for the specific data ("Yes" in step S2 to step S3) and the other first process for the format conversion data 11 ("No" in step S2 to step S4) are all format conversion data. 11 (“No” in step S5). When the second process and the first process are performed on all the format conversion data 11 (“Yes” in step S5), the compression / decoding unit 5 performs the compression process on the reduced data 12 (step S6). Thereby, the compressed data 20 is generated.

  Thus, the processing procedure for performing data reduction and data compression is completed. In this procedure, by performing compression processing on the reduced data 12 in which the data amount of the format conversion data 11 is reduced, the data amount of the compressed data 20 becomes smaller than when the compression processing of the format conversion data 11 is simply performed.

  On the other hand, when the original pre-conversion data 10 is generated from the compressed data 20 compressed in this way, the reverse procedure of steps S1 to S6 is performed.

  FIG. 6 is a flowchart showing the procedure of the data decoding process and the addition process of the data amount reducing apparatus 1 according to this embodiment. As shown in the figure, first, the compression / decoding unit 5 of the data amount reduction device 1 restores the compressed data 20 to generate reduced data 12 (step S11).

  Of the reduction data 12, as shown in (c) of FIG. 3, the first process is performed for the sum data 12a and the difference data 12b, that is, the data that is not specific data ("No" in step S12). (Step S13). Specifically, based on (processing guideline 2) of the first processing information 31a of the correlation content management unit 3, one of the data having the lowest digit 12c (sum data 12a in FIG. 3C). ) Is copied and added as the bottom value 12d of the other data (difference data 12b in FIG. 3C) to generate the format conversion data 11.

  On the other hand, of the reduction data 12, the specific data having only the sum data 12e as shown in FIG. 4B ("Yes" in step S12) is subjected to the second process (step S13). Specifically, the format conversion data 11 is generated by adding the difference data 12f based on (processing guide 4) of the second processing information 32a of the correlation content management unit 3. This process is performed for all the reduced data 12 (“Yes” in step S15).

  When all the reduced data 12 is converted into the format conversion data 11 (“No” in step S15), the data format conversion unit 2 converts the format conversion data 11 into the pre-conversion data 10 (step S16). Thus, in this embodiment, based on the compressed data 20, the original data before conversion 10 can be reconverted without degrading the data.

  As described above, in this embodiment, the reduced data 12 in which the entire data amount is reduced based on the first processing information 31 a and the second processing information 32 a recorded in the correlation content management unit 3. Is generated. The first processing information 31a is formed based on the first correlation information 31, and the second processing information 32a is formed based on the second correlation information 32, whereby the sum data 11a and the difference data 11b are formed. The bottom digit 11c (or the bottom digit 11d), which is a highly relevant part of the data, and the difference data 11f, which is a highly relevant part of the sum data 11e and the difference data 11f, are deleted, and format conversion is performed. Reduction data 12 having a data amount reduced from that of data 11 can be generated. In addition, the lower digit 11c (or the lowest digit 11d), which is data deleted from the format conversion data 11, and the difference data 11f are added to the reduction data 12, so that the original format conversion data 11 can be easily obtained. Or the pre-conversion data 10. As a result, the data amount of the format conversion data 11 that can be compressed can be reduced to obtain high compression efficiency, and the reduced data 12 with the reduced data amount can be converted into the original format conversion data 11 or the like with high recoverability. It is possible to return to the pre-conversion data 10.

  In this embodiment, the lowest digit (for example, the lowest digit 11d) is deleted from the sum data 11a and the difference data b composed of binary numbers based on the first processing information 31a. Thus, the highly correlated portion of the binary sum data 11a and difference data 11b is deleted, so that an excessive increase in the data amount of the format conversion data 11 that can be compressed is suppressed, and the reduction data 12 is further reduced. It is possible to restore the original format conversion data 11 and the pre-conversion data 10 with high recoverability.

  In this embodiment, based on the second processing information 32a, the difference data 11f in the case where the value of the sum data 11e is “510” as the maximum possible value or “0” as the minimum possible value. By deleting the value, the value of the difference data 11f that is uniquely determined based on the value of the sum data 11e is deleted, so that an excessive increase in the data amount of the format conversion data 11 that can be compressed is suppressed. The reduced data 12 can be restored to the original format conversion data 11 and the pre-conversion data 10 with higher reconstructability.

  In this embodiment, the reduction data is based on the first processing information 31 a formed based on the first correlation information 31 and the second processing information 32 a formed based on the second correlation information 32. 12, the original format conversion data 11 can be reliably configured. As a result, the reduced data 12 formed by deleting data from the format conversion data 11 that can be compressed can be returned to the original format conversion data 11 and the pre-conversion data 10 with higher recoverability.

  In this embodiment, the first processing information 31a for the sum data 11a and difference data 11b and sum data 11e and 11f of the format conversion data 11 obtained by converting the left data 10a and the right data 10b of the audio data. By generating the reduced data 12 based on the second processing information 32a and the left data 10a and the right data 10b of highly correlated audio data, the reduced data 12 that can be compressed is obtained using the correlation. It can be generated or reconverted to the original pre-conversion data 10 without degrading the data. Thereby, high compression efficiency can be obtained for the pre-conversion data 10 as audio data, and the reduced data 12 with the reduced data amount can be returned to the pre-conversion data 10 of the original audio data with high recoverability.

  In the data amount reduction apparatus 1 of this embodiment, the data reduction processing by the data reduction / addition unit 4 increases the data reduction amount in [second processing]. Therefore, by recursively repeating the processes shown in steps S1 to S6 for data that is expected to be filled with the maximum value or the minimum value, such as the “α plane”, the data amount can be further increased. Can be reduced.

  In addition, although the said embodiment illustrated and demonstrated the case where the left data 10a and the right data 10b of the data 10 before conversion were 8-bit unsigned integer data, respectively, without limiting to this, It is applicable even to numerical data with such a bit number. For example, the pre-conversion data 10 may be 4 bits or 16 bits, respectively, or may be integer data with a sign.

Here, for example, the left data 10a and the right data 10b of the pre-conversion data 10 are each 8-bit signed integer data (the upper 1 bit indicates a positive / negative sign, the lower 8 bits indicate a numerical value, -255 to 255) In addition to (Principle 5) and (Principle 6) of [Operation Principle 2], processing based on the following (Principle 7) and (Principle 8) is performed.
(Principle 7) If the maximum value (255) that the left signal value can take and the minimum value (-255) that the right signal value can take, the sum data value becomes “0” and the difference data value can take the maximum value ( 510).
(Principle 8) If the left signal value can be the minimum value (−255) and the right signal value can be the maximum value (255), the sum data value is “0” and the difference data value can be the minimum value ( -510).
In this case, the data reduction / addition unit 4 uses (Principle 7) (Principle 8) to perform processing based on the following (Processing Guideline 5) or (Processing Guideline 6).
(Processing guideline 5): The sum data 11e and the difference data 11f exist in the format conversion data 11, and the sum data 11e is “0” specific data, and the maximum value (510) or the difference data 11f can take. If it is the minimum value (−510), the sum data 11e is deleted.
(Processing guideline 6): When only the difference data 11f exists in the format conversion data 11 and the difference data 11e is specific data of the maximum value “510” or the minimum value “−510” that can be taken, the sum data “0” is added as 11e.
By such processing, the reduced data 12 having a smaller number of bits than the format conversion data 11 can be formed and the original format conversion data 11 can be easily generated from the reduction data 12 as in the case of the above embodiment.

  In the above-described embodiment, the configuration and processing procedure of the data amount reduction device 1 have been described using audio data as an example. However, the present invention is not limited to this, and image data and various types of data other than image data are not limited to this. The data amount reduction device 1 can be applied.

  In the above embodiment, in the [second processing], only the sum data 12e is left in the reduction data 12, but the present invention is not limited to this, and only the difference data 12f is left in the reduction data 12. You can also For example, if the difference data 12f is “−255”, based on the correlation information, the sum data 11e of the original format conversion data 11 is “255” (that is, the original pre-conversion data 10 The left data 10a is “0” and the right data 10b is “255”), and the format conversion data 11 can be generated and reconverted to the pre-conversion data 10.

  In the above embodiment, the first correlation information 31, the second correlation information 32, the first processing information 31a, and the second processing information 32a are recorded in the correlation content management unit 3. Alternatively, the first correlation information 31 and the second correlation information 32 may not be recorded, and the first processing information 31a and the second processing information 32a may be recorded.

  In the above embodiment, the data format conversion unit 2 and the data reduction / addition unit 4 are configured as separate functional blocks, but may be configured as the same module.

  The above embodiment is an exemplification of the present invention, and it is needless to say that the present invention is not limited to the above embodiment.

DESCRIPTION OF SYMBOLS 1 ... Data amount reduction apparatus 2 ... Data format conversion part (data format conversion means)
3 ... Correlation content management unit (processing content management means)
4. Data reduction / addition part (data reduction means)
10: Pre-conversion data 10a: Left data (first value)
10b ... Right data (second value)
11... Format conversion data 11a, 11e, 12a, 12e... Sum data 11b, 11f, 12b, 12f... Difference data 11c, 11d, 12c, 12d. Reduction data 31 ... first correlation information (correlation information)
31a: First processing information (processing information)
32 ... second correlation information (correlation information)
32a ... Second processing information (processing information)

Claims (6)

The pre-conversion data having the first value and the second value, the sum data as the sum of the first value and the second value, and the difference between the first value and the second value A data amount reduction device comprising a data format conversion means for converting to format conversion data consisting of difference data of
Processing content management means in which processing information for determining the processing content for the sum data or the difference data is recorded based on the correlation information of the sum data and the difference data constituting the format conversion data;
Data reduction means for generating reduced data in which the total data amount is reduced by deleting all or part of at least one of the sum data and the difference data based on the processing information A data amount reduction device characterized by that. The sum data and the difference data constituting the format conversion data are composed of binary numbers,
The processing information is based on the correlation information that if one of the sum data and the difference data is even, the other is even, and if one of the sum data and the difference data is odd, the other is odd. Information on processing to delete the lowest digit that matches in the sum data and the difference data,
2. The data amount reduction device according to claim 1, wherein the data reduction unit deletes the bottom digit of one of the sum data and the difference data based on the processing information. . The processing information is based on the correlation information that the difference data is 0 when the sum data is the maximum value or the minimum value that the sum data can take, and the difference data when the sum data is the specific case. Is the process information to delete
2. The data amount reduction device according to claim 1, wherein the data reduction unit deletes the difference data when the sum data is the specific case based on the processing information. The processing information is based on the correlation information that the sum data is 0 in a specific case where the difference data is a maximum value or a minimum value that the difference data can take, and the sum data when the sum data is the specific case. Is the process information to delete
2. The data amount reduction device according to claim 1, wherein the data reduction unit deletes the sum data when the difference data is the specific case based on the processing information. A specific value is added to a part or all of the sum data or the difference data constituting the reduction data generated by the data reduction means based on conditions at the time of generation of the reduction data, and the original data 5. The data amount reducing device according to claim 1, further comprising data adding means for restoring the format conversion data.   6. The data amount reduction device according to claim 1, wherein the pre-conversion data is image data or audio data to be subjected to lossless compression.

Images