Compression processing device, decompression processing device, compression processing program, decompression processing program
By employing modified logistic mapping to divide and calculate compression intervals based on probability, the method enhances data compression efficiency and achieves higher compression ratios through optimized initial value intervals and adaptable control parameters.
Patent Information
- Application Number
- JP2023059298
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-03-31
- Publication Date
- 2024-11-27
- Estimated Expiration
- 2043-03-31
AI Technical Summary
Existing data compression methods lack emphasis on achieving high compression efficiency and do not provide effective decompression of data with high compression ratios.
The method involves dividing the horizontal axis interval into two sections based on the appearance probability value P, using modified logistic mapping formulas to calculate upper and lower limits of compression values, and generating compression codes from these ranges, with the ability to vary the control parameter P for enhanced compression.
This approach achieves higher compression ratios by optimizing the initial value intervals and reducing the number of bits required for compression, outperforming conventional methods by adapting to data patterns and varying the control parameter.
Smart Images

Figure 0007590667000030 
Figure 0007590667000031 
Figure 0007590667000032
Abstract
Description
[Technical field]
[0001] The present invention relates to a compression processing device, a decompression processing device, a compression processing program, and a decompression processing program that perform lossless compression and decompression. [Background technology]
[0002] Patent Document 1 discloses a compression processing device, a decompression processing device, a compression processing program, and a decompression processing program. Patent Document 1 introduces a compression method that traces back a modified Bernoulli shift map. This method uses the occurrence probability of data as the gradient coefficient of the map, and introduces that it is possible to obtain a compression rate equivalent to that of conventional arithmetic coding.
[0003] In addition, Patent Document 2 discloses a random number generator, which generates an initial value X by referencing random numbers output by a logistic map that is in a topological conjugate relationship with a tent map and tracing back. 0 Furthermore, in this patent document 2, the initial value X of the Gray code generated from the tent map and the logistic map is 0 The interval is shown.
[0004] Patent Document 3 discloses a method for compressing natural image data (still images). This patent document 3 shows that bitmap natural image intensity data is expanded into bitmap gradation value data, an exclusive OR operation is performed between each layer, code conversion based on distribution is performed, and compression using a chaos compression / expansion table is performed to obtain a compressed code. Furthermore, it also shows that expansion is performed by following the reverse process to the above.
[0005] Furthermore, Patent Document 4 discloses that the encoder of the terminal device sequentially applies a chaos map that generates chaos to the original data of the domain, and obtains a compression result that satisfies the criterion and the number of times the chaos map is applied. Meanwhile, it is disclosed that another terminal device obtains the compression result and the number of times the chaos map is applied from the terminal device 22 by communication, and the other terminal device applies the chaos inverse map to the compression result by the decoder the same number of times as the number of applications to obtain data that is the decompression result. [Prior art documents] [Patent documents]
[0006] [Patent Document 1] JP 2019-047450 A [Patent Document 2] JP 2019-168963 A [Patent Document 3] JP 2004-129179 A [Patent Document 4] JP 2009-33474 A Summary of the Invention [Problem to be solved by the invention]
[0007] In the data compression described above, there is no particular focus on obtaining high compression efficiency, nor is there any disclosure of decompressing compressed data with such a high compression ratio.
[0008] The present invention has been made in consideration of the current state of conventional data compression processing and data decompression processing, and has an object to provide a compression processing device, a decompression processing device, a compression processing program, and a decompression processing program that are capable of obtaining a high compression ratio. [Means for solving the problem]
[0009] The compression processing device according to the present invention includes a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting uncompressed decompressed code data to be compressed, and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function having an occurrence probability value equal to or greater than P; The decompressed code data and a modified logistic map compression processing means for iteratively calculating an upper limit and a lower limit of a compressed value for each unit data of the data to obtain a range of the compressed values, and a modified logistic map compression code generating means for generating a compressed code from the range of the compressed values finally obtained by the modified logistic map compression processing means.
[0010] In the compression processing device according to the present invention, the modified logistic map compression processing means calculates the upper limit and the lower limit of the compression value for each unit data by using a quadratic equation aX 2 Formula for finding the solution to +bX+c=0
number
[0011] In the compression processing device according to the present invention, the modified logistic map formula generation means prepares a plurality of the occurrence probability values P, and generates a plurality of modified logistic map formulas based on the plurality of occurrence probability values P, and uses the plurality of modified logistic map formulas generated by the modified logistic map formula generation means to perform the following: The decompressed code dataand a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the modified logistic map equations obtained by the modified logistic map initial value calculation means to obtain the modified logistic map equation which results in the initial value with the smallest number of bits, wherein the modified logistic map compression processing means performs calculation using the modified logistic map equation obtained by the modified logistic map optimal equation acquisition means.
[0012] The compression processing device according to the present invention includes a code conversion means for converting decompressed code data into a Gray code, dividing the decompressed code data represented by the converted Gray code into a plurality of ranges according to size, and changing the decompressed code data at the ends of the ranges into decompressed code data at the center of the ranges, and further includes the modified logistic map formula generation means, the modified logistic map compression processing means, The modified logistic map initial value calculation means, and the modified logistic map optimal equation acquisition means The method is characterized in that the processing is carried out using the code after the code conversion means has converted the code.
[0013] The compression processing device according to the present invention includes a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting uncompressed decompressed code data to be compressed, and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function having an occurrence probability value equal to or greater than P; The decompressed code dataa modified logistic side processing unit including a modified logistic map compression processing unit which calculates an upper limit and a lower limit of a compressed value for each unit data of the data, and repeats the calculation to obtain a range of the compressed values; and a modified logistic map compressed code generating unit which generates a compressed code from the range of compressed values finally obtained by the modified logistic map compression processing unit; and a modified tent mapping equation generating unit which divides a horizontal axis interval from 0 to 1 into two intervals based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generates an equation of a modified tent mapping consisting of a first linear function from 0 to less than the occurrence probability value P, and a second linear function equal to or greater than the occurrence probability value P. The decompressed code data and a modified tent side processing unit comprising a modified tent map compression processing means for calculating an upper limit and a lower limit of a compressed value for each unit data of the data to obtain a range of the compressed values by iteratively performing an operation to obtain a range of the compressed values, and a modified tent map compressed code generation means for generating a compressed code from the range of compressed values finally obtained by the modified tent map compression processing means, wherein the modified logistic map formula generation means uses a plurality of prepared occurrence probability values P to generate a plurality of modified logistic map formulas based on the plurality of the occurrence probability values P, and the modified tent map formula generation means uses a plurality of prepared occurrence probability values P to generate a plurality of modified tent map formulas based on the plurality of the occurrence probability values P, and the modified logistic side processing unit comprises: The decompressed code data and a modified logistic map optimum formula acquisition means for comparing the numbers of bits of initial values corresponding to the modified logistic map formulas obtained by the modified logistic map initial value calculation means to obtain the modified logistic map formula that results in the initial value with the smallest number of bits. The modified tent side processing means performs the following operations using the multiple modified tent map formulas generated by the modified tent map formula generation means: The decompressed code dataand a modified tent mapping optimal equation acquisition means for comparing the number of bits of initial values corresponding to the modified tent mapping equation obtained by the modified tent mapping initial value calculation means to acquire the modified tent mapping equation which yields the smallest initial value with respect to the number of bits. The modified tent mapping initial value calculation means further comprises an optimal mapping equation comparison result acquisition means for comparing an initial value having the smallest number of bits among the number of bits of initial values corresponding to the modified logistic map equation obtained by the modified logistic map initial value calculation means with an initial value having the smallest number of bits among the initial values corresponding to the modified tent mapping equation obtained by the modified tent mapping initial value calculation means to acquire the mapping equation which yielded the smaller initial value as a result of the comparison, and for acquiring the mapping equation acquired by the modified logistic map optimal equation acquisition means or the modified tent mapping optimal equation acquisition means which corresponds to the mapping equation. The modified logistic map equation or the modified tent mapping equation acquired by the optimal mapping equation acquisition means is characterized in that the modified logistic map equation or the modified tent mapping equation acquired by the optimal mapping equation comparison result acquisition means is used for calculation.
[0014] The decompression processing device according to the present invention includes a modified logistic map formula generating means for dividing an occurrence probability value P used in a compression process into two intervals and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function for the occurrence probability value P or more; and performing an operation using the modified logistic map formula on a compressed code, Decompressed Code Data a modified logistic map decompression processing means for performing an iterative calculation in which a one-unit data acquisition calculation is a calculation for obtaining a value corresponding to one unit data of the modified logistic map, and a one-unit data acquisition calculation is used to obtain the next one unit data by using the obtained value in an operation using the modified logistic map formula, until a predetermined number of times is reached; and The decompressed code data and a modified logistic map decompression code generating means for converting the data into one unit data of the above.
[0015] The decompression processing device according to the present invention is characterized by comprising a means for extracting P included in the compressed code and sending it to the modified logistic map formula generating means.
[0016] The compression processing program according to the present invention includes a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting decompressed code data before compression, and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function equal to or greater than the occurrence probability value P. The decompressed code data and a modified logistic map compression code generating means that generates a compressed code from the range of compressed values finally obtained by the modified logistic map compression processing means.
[0017] In the compression processing program according to the present invention, when the computer, as the modified logistic map compression processing means, calculates the upper limit and the lower limit of the compression value for each unit data, the computer calculates a quadratic equation aX 2 Formula for finding the solution to +bX+c=0
number
[0018] In the compression processing program according to the present invention, the computer functions as the modified logistic map equation generating means to prepare a plurality of the occurrence probability values P and generate a plurality of equations of the modified logistic map based on the plurality of the occurrence probability values P, and the computer further functions to perform the following calculation using the plurality of equations of the modified logistic map generated by the modified logistic map equation generating means: The decompressed code dataand a modified logistic map initial value calculation means for calculating an upper limit and a lower limit of a compression value for a first unit data to obtain an initial value, and a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the modified logistic map equation obtained by the modified logistic map initial value calculation means to obtain the modified logistic map equation which results in an initial value with the smallest number of bits, and the computer functions as the modified logistic map compression processing means to perform a calculation using the modified logistic map equation obtained by the modified logistic map optimal equation acquisition means.
[0019] In the compression processing program according to the present invention, the computer functions as a code conversion means for converting decompressed code data into a gray code, dividing the decompressed code data represented by the converted gray code into a plurality of ranges according to size, and changing the decompressed code data at the ends of the ranges into decompressed code data at the center of the ranges, and the computer functions as the modified logistic map formula generation means, the modified logistic map compression processing means, The modified logistic map initial value calculation means, and the modified logistic map optimal equation acquisition means The code conversion means functions to perform processing using the code converted by the code conversion means.
[0020] The compression processing program according to the present invention includes a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting decompressed code data before compression, and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function equal to or greater than the occurrence probability value P. The decompressed code dataa modified logistic map compression processing means for calculating an upper limit and a lower limit of a compressed value for each unit data of the data to obtain a range of the compressed values, and a modified logistic map compression code generating means for generating a compressed code from the range of compressed values finally obtained by the modified logistic map compression processing means, thereby realizing a modified logistic side processing unit; and a modified tent mapping equation generating means for dividing a horizontal axis interval from 0 to 1 into two intervals based on an occurrence probability value P of a code constituting the decompressed code data before compression, and for generating an equation of a modified tent map composed of a first linear function from 0 to less than the occurrence probability value P and a second linear function equal to or greater than the occurrence probability value P. The decompressed code data and a modified tent map compressed code generation means that generates a compressed code from the range of compressed values finally obtained by the modified tent map compression processing means, so as to realize a modified tent side processing unit. The computer causes the modified logistic map formula generation means to function to generate a plurality of modified logistic map formulas based on the plurality of occurrence probability values P using the plurality of prepared occurrence probability values P, and the computer causes the modified tent map formula generation means to function to generate a plurality of modified tent map formulas based on the plurality of prepared occurrence probability values P, and the computer causes the modified logistic side processing unit to The decompressed code data a modified logistic map initial value calculation means for calculating an upper limit and a lower limit of a compressed value for a first unit data of the data set to obtain an initial value, and a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the modified logistic map equations obtained by the modified logistic map initial value calculation means to obtain the modified logistic map equation which results in an initial value with the smallest number of bits, and the computer is caused to function to have the following configuration: The decompressed code dataa modified tent mapping initial value calculation means for calculating an upper limit and a lower limit of a compressed value for a first unit data to obtain an initial value, and a modified tent mapping optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the equation of the modified tent mapping obtained by the modified tent mapping initial value calculation means to obtain the equation of the modified tent mapping which has resulted in an initial value with the smallest number of bits, and the computer is caused to function as follows: the computer functions as an optimal mapping formula comparison result acquisition means which compares an initial value of the smallest number of bits among the bits of the initial value corresponding to the tent mapping formula with an initial value of the smallest number of bits among the bits of the initial value corresponding to the tent mapping formula, to obtain a mapping formula which yields the smaller initial value as a result of the comparison, and which obtains the mapping formula obtained by the modified logistic map optimal formula acquisition means or the modified tent mapping optimal formula acquisition means corresponding to the mapping formula, and the computer causes each means of the modified logistic side processing unit or each means of the modified tent side processing unit to function to perform calculations using the modified logistic map formula or the modified tent mapping formula obtained by the optimal mapping formula comparison result acquisition means.
[0021] The decompression program according to the present invention includes a modified logistic map formula generating means for dividing a computer, which decompresses compressed data, into two intervals based on an occurrence probability value P used in the compression process, and generating a formula for a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P, and a second quadratic function for an occurrence probability value equal to or greater than P; and a calculation means for performing an operation using the modified logistic map formula on the compressed code, Decompressed Code Data a modified logistic map decompression processing means for performing an iterative calculation in which a one-unit data acquisition calculation is a calculation for obtaining a value corresponding to one unit data of the above, and a one-unit data acquisition calculation is performed by using the obtained value in an operation using the formula of the modified logistic map to obtain the next one unit data, until a predetermined number of times is reached; The decompressed code dataThe present invention is characterized in that the modified logistic map decompression code generating means functions as a modified logistic map decompression code generating means for converting the data into one unit of data.
[0022] The decompression processing program according to the present invention is characterized in that it causes the computer to function as a means for extracting P included in the compressed code and sending it to the modified logistic map formula generating means. [Brief description of the drawings]
[0023] [Figure 1A] FIG. 1 is a configuration diagram of a compression processing device and a decompression processing device according to a first embodiment of the present invention. [Figure 1B] FIG. 11 is a configuration diagram of a compression processing device and a decompression processing device according to a second embodiment of the present invention. [Figure 1C] FIG. 11 is a configuration diagram of a compression processing device and a decompression processing device according to a third embodiment of the present invention. [Figure 1D] FIG. 11 is a configuration diagram of a compression processing device and a decompression processing device according to a fourth embodiment of the present invention. [Diagram 2] FIG. 13 is a diagram showing a logistic map. [Diagram 3] A graph showing the change in time series values due to the logistic map. [Figure 4] FIG. 13 is a diagram showing a tent map. [Diagram 5] A graph showing the change in time series values due to tent mapping. [Figure 6] FIG. 13 shows a tent map for P=0.75. [Figure 7] FIG. 13 is a diagram showing examples of intervals of the initial value X0 of the tent map and the logistic map, and notation of the Gray code and binary code generated corresponding to each interval. [Figure 8] FIG. 13 is a flowchart showing a process for performing compression according to equation (4) of the tent map. [Figure 9] FIG. 13 is a diagram showing the transition of determining the initial value X0 interval when compressing the tent mapping formula by going back through it. [Figure 10] FIG. 13 is a flowchart showing a process for performing decompression according to the tent mapping formula. [Figure 11] A diagram showing the process of expanding the compressed code (0.1011)2 according to the tent mapping equation (3). [Figure 12] A diagram showing the transition of determining the initial value X0 interval when compressing by going back through the formula for the logistic map with P=0.5. [Figure 13] A diagram showing a modified logistic map that is similar to the modified tent map when P=0.75. [Figure 14] A diagram showing the right half of the P=1 / 4 logistic map moved by plus 1 / 2 along the X-axis. [Figure 15] A diagram showing modified logistic maps for P=1 / 8, 2 / 8, 3 / 8, 4 / 8, 5 / 8, 6 / 8, and 7 / 8. [Figure 16] 4 is a flowchart showing the operation of the compression processing device according to the first embodiment of the present invention. [Figure 17] FIG. 13 is a diagram showing an example of compression processing for tracing back the modified logistic map when P=0.75. [Figure 18] 4 is a flowchart showing the operation of the decompression processing device according to the first embodiment of the present invention. [Figure 19] FIG. 13 shows the results of compressing data with an occurrence probability value of P=0.75 using two methods, the modified tent map and the modified logistic map, to find the initial value X0. [Figure 20] This is a graph in which P is plotted on the horizontal axis, P is changed to obtain a deformed tent map, and each map is used to obtain an initial value X0 range, which is assigned to the vertical axis. [Figure 21] This is a graph in which P is plotted on the horizontal axis, P is changed to obtain a modified logistic map, and each map is used to obtain an initial value X0 range, which is assigned to the vertical axis. [Figure 22] 10 is a flowchart showing the operation of a compression processing device according to a second embodiment of the present invention. [Diagram 23] A diagram showing the extension process using the modified logistic map with P=5 / 8(0.625) and initial value X0=0.75. [Figure 24]FIG. 13 shows the results of comparing the results of compressing P in the first embodiment as the probability of occurrence of "0" in 254 patterns excluding data with all 8 bits being "0" and all being "1" and compressing P by varying it in the second embodiment. [Diagram 25] FIG. 13 is a flowchart showing an example of processing related to code conversion (bit inversion) according to the third embodiment. [Figure 26] FIG. 13 is a diagram showing the contents of a compressed code obtained according to the third embodiment. [Figure 27] FIG. 13 shows a comparison result of the number of compressed code bits (vertical axis) of an initial value X0 compressed using a modified logistic map in which the count number of bit 0 is the occurrence probability value P (dotted line) and the number of compressed code bits (vertical axis) of an initial value X0 compressed using a modified logistic map in which P is varied (solid line). [Figure 28] 10 is a flowchart showing the operation of a decompression processing device according to a third embodiment of the present invention. [Figure 29] 11 is a flowchart of a process performed by a compression processing device according to a fourth embodiment of the present invention. [Diagram 30] A graph showing a comparison of the number of initial value X0 bits (vertical axis) after compression of 8-bit (horizontal axis) data using the modified logistic map and the modified tent map. [Diagram 31] 11 is a flowchart of a process performed by a decompression processing device according to a fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0024] Hereinafter, a compression processing device, a decompression processing device, a compression processing program, and a decompression processing program according to the present invention will be described with reference to the accompanying drawings. In each drawing, the same components are given the same reference numerals and duplicated explanations are omitted. The embodiment of the present invention is a lossless compression device with a high compression ratio using a modified logistic map. Therefore, we will start by explaining the logistic map and the tent map.
[0025] <About the logistic map> The logistic map is defined by the following equation (1).
number
[0026] <About the tent map> The tent map is defined by the following equation (2):
number
number
[0027] Here, the effect of the lossless compression method by the compression processing device using the modified logistic map formula according to this embodiment will be considered. The aforementioned Patent Document 1 introduces a compression method that traces back the modified Bernoulli shift map, and describes that a compressed code equivalent to a conventional arithmetic code can be obtained. Since the slope coefficient of the modified Bernoulli shift map is linear, it is presumed that a compression rate equivalent to that obtained when compression processing is performed using a modified tent map, which also has a linear slope coefficient P, can be obtained.
[0028] On the other hand, FIG. 11 of the 23rd document of the Patent Document 2 is the same as FIG. 7 of the present application. In FIG. 7, the initial value X 0 The interval is shown, and the initial value X 0 Comparing the sections, it can be seen that the two are different. Therefore, when the logistic map is used as a compression function, the initial value X 0 Therefore, depending on the data pattern to be compressed, it is possible to obtain a high compression ratio that cannot be obtained by arithmetic coding.
[0029] A compression method using modified tent mapping is explained. When the digital data to be compressed is represented as a binary number and the frequency of “0” and “1” is represented as the occurrence probability value P, the following equation (4) is used for compression going back through the modified tent map of equation (3).
number
[0030] A flowchart showing the process of compressing data according to the formula (4) is shown in FIG. 8. As an example, the initial value X 0 The transitions for which the intervals were determined are shown in Figure 9. In this case, the number of bits 0 is three, so P is set to 3 / 4, and the lower limit X 3 =0 and upper limit X 3 = 1 and the initial value X obtained using equation (4) 0 The half-open interval for is [0.64453125,0.75). This compression code is expressed in bits as the interval [(0.10100101) 2,(0.11) 2 ) and the compression code has the fewest bits (0.1011) among them. 2 is 0.6875 in decimal, and the initial value of the compressed value is X 0 It becomes.
[0031] A flowchart showing the decompression process is shown in FIG. 10. As a specific example of the decompression process, 2 The process of expanding is shown in FIG. 11. As shown in FIG. 11, the initial value X 0 Set the mapping X i If the bit is less than 0.75, a bit "0" is output, and if it is 0.75 or more, a bit "1" is output, and the process ends after outputting four bits. In Fig. 11, the compressed code "0100" is output as the decompressed bits.
[0032] <Embodiment 1> Next, a first embodiment of the present invention will be described. Fig. 1A shows a configuration diagram of a compression processing device 100 and a decompression processing device 200 according to the first embodiment of the present invention. The compression processing device 100 includes a modified logistic map formula generation means 101, a modified logistic map compression processing means 110, and a modified logistic map compression code generation means 120. The decompression processing device 200 includes a modified logistic map formula generation means 201, a modified logistic map decompression processing means 210, and a modified logistic map decompression code generation means 220.
[0033] The compression processing device 100 and the decompression processing device 200 can be configured by different computers, or may be configured by a single computer. Reference numeral 300 denotes a storage unit for decompression code, which may be located inside the compression processing device 100 or the decompression processing device 200, or may be located in an external storage device or another computer. Reference numeral 400 denotes a storage unit for compressed code, which may be located inside the compression processing device 100 or the decompression processing device 200, or may be located in an external storage device or another computer.
[0034] The modified logistic map equation generating means 101 divides the horizontal axis interval from 0 to 1 into two intervals based on the occurrence probability value P of the code constituting the decompressed code data before compression processing, and generates an equation of the modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P, and a second quadratic function equal to or greater than the occurrence probability value P.
[0035] The modified logistic map compression processing means 110 calculates the upper and lower limit values of the compressed value for each unit data of the decompression code, and repeats the calculation to obtain the range of the compressed value. Here, the compressed value is obtained by calculating an initial value corresponding to the compressed code and this value, and is a value represented as "Xi" below. The modified logistic map compression code generation means 120 generates a compressed code from the range of compressed values finally obtained by the modified logistic map compression processing means 110. When the decompression code is composed of the numerical values 1 and 0, one unit data of the decompression code refers to these 1 and 0. Also, when the decompression code is composed of alphabets such as abcd..., one unit data of the decompression code refers to one character such as a, b, c, d,.... Furthermore, when the decompression code is composed of a character string such as "Hello World!!\ " (generally, a data string) as one code, as will be described later, it refers to this character string such as "Hello World!!\ " (generally, a data string).
[0036] The modified logistic map formula generating means 201 divides the horizontal axis section from 0 to 1 into two sections according to the occurrence probability value P of the code constituting the decompressed code data before compression processing, and generates a formula of the modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P, and a second quadratic function equal to or greater than the occurrence probability value P. Therefore, the modified logistic map formula generating means 201 may share the modified logistic map formula generating means 101.
[0037] The modified logistic map decompression processing means 210 performs an iterative calculation, which repeats a predetermined number of times, a one-unit data acquisition calculation that performs an operation using the modified logistic map equation on the compressed code to obtain a value corresponding to one unit data of the decompressed code, and a one-unit data acquisition calculation that uses the obtained value in an operation using the modified logistic map equation to obtain the next one unit data. The modified logistic map decompression code generation means 220 converts a value obtained for each one-unit data acquisition calculation by the modified logistic map decompression processing means into one unit data of the decompressed code based on the coefficients of the modified logistic map equation.
[0038] In this embodiment, the compression method is shown by assuming the logistic map of equation (1) as a modified logistic map. From the modified logistic map, X i < 0.5, get bit 0, 0.5 ≤ X i When , get bit 1 and get a bit string in succession. Then trace back the bit string to the initial value X 0 Consider finding the interval of . FIG. 12 shows an example of tracing back the logistic map when the bit string is "0010". First, the last bit is “0”, so X 3 The interval "0≦X 3 It can be seen that the value is <0.5”. Next, “X 3 =4X 2 (1-X 2 )” is a quadratic equation “4X 2 2 -4X 2 +X 3 =0” and the lower limit X 3 If we apply =0, then X 2 =1 or X 2 =-1 is obtained, but “X i >0” so “X 2 = 1". Upper limit X 3 If we apply =0.5, we get 4X 2 2 -4X2 +0.5=0” quadratic equation aX 2 The formula for the solution of +bX+c=0
number
[0039] Next, “X 2 =4X 1 (1-X 1 )” is a quadratic equation “4X 1 2 -4X 1 +X 2 = 0” and similarly, the previously calculated X 2 Upper limit X from the interval 2 =1 is “4X 1 2 -4X 1 +1=0" so the solution to the quadratic equation is X 1 =0.5 is obtained. Lower limit X 2 =0.853553390593 is “4X 1 2 -4X 1 -0.853553390593=0" From the solution of the quadratic equation, X 1 =0.308658283817 and X 1 = 0.691341716183 is obtained, and X 1 Since the observed bit is 0, it is in the range “0≦X 1 <0.5” so X1 =0.308658283817. As a result, X 1 The interval is 0.308658283817≦X 1 It can be seen that the calculation is based on <0.5". Finally, “X 1 =4X 0 (1-X 0 )” is also “4X 0 2 -4X 0 -0.5=0" and "4X 0 2 -4X 0 -0.308658283817=0” and taking into account that bit 0 is observed, we finally get the initial value X 0 The interval is “0.084265193849≦X 0 <0.146446609407”. From the narrowed down range, the initial value X 0 Repeat the iterations by giving X i This means that when is 0.5 or more, a bit “1” is obtained, and when it is less than 0.5, taking four consecutive bits of “0” always results in the bit string “0010”.
[0040] In this way, the modified logistic map compression processing means 110 calculates the upper and lower limits of the compression value for each unit data by using the quadratic equation aX 2 Formula for finding the solution to +bX+c=0
number
[0041] From the observed bit string as above, the initial value X of the logistic map is 0 The results of finding 4 bits (total of 16 patterns) in the procedure for finding the interval are shown in the table in Figure 7, “Initial value X of the logistic map 0 The example in Figure 12 is shown in Figure 7. <4> This section is as follows. Figure 7 <1> ~ <16> Each initial value X 0 Iteratively calculate the logistic map formula (1) from the interval of X i When the bit is greater than or equal to 0.5, a "1" is obtained, and when it is less than 0.5, a "0" is obtained by taking four consecutive bits, which generates a bit string that conforms to the "Gray code" string.
[0042] The modified logistic map used in this embodiment will be defined. Equation (1) is a recurrence formula for a general logistic map with a control parameter of 4, and is a quadratic polynomial. In the above example, this is treated as a modified logistic map. In this embodiment, following the modified tent map of equation (3), a modified logistic map consisting of a quadratic polynomial with a variable parameter like P in equation (4) is defined. That is, the modified logistic map equation generating means 101 generates a modified logistic map equation as follows.
[0043] As an example, FIG. 13 shows a modified logistic map similar to the modified tent map in FIG. 6 when P=0.75. In Figure 13, P=3 / 4 or less is expressed by the following second-order polynomial (first quadratic function).
number
number
number
[0044] Here, in the equation where f(x)=0 when x=0 as in equation (1), the coefficient c=0, so the equation becomes the following quadratic equation.
number
number
number
number
number
number
number
number
[0045] Next, a method for obtaining a quadratic expression of equation (6) with P=3 / 4 or more will be described. Since the quadratic curve of equation (6) above P=3 / 4 in Figure 13 does not pass through the origin (0,0) of the XY axis, we consider finding coefficients a, b, and c of the quadratic equation of the quadratic curve obtained by shifting the quadratic curve by +1 / 2 in the X-axis direction when P=1 / 4 as shown in Figure 14.
[0046] When x = 1 / 4, f(1 / 4) = 1, so the following equation (13) holds.
number
number
number
number
number
number
[0047] As described above, the modified logistic map used in this embodiment can be defined as a quadratic polynomial consisting of two types with the probability value P of occurrence of the code constituting the decompressed code data before compression in which the horizontal axis section from 0 to 1 is compressed as a variable parameter, with this P as the boundary. The modified logistic map equation generating means 101 of this embodiment generates an equation of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P, and a second quadratic function equal to or greater than the occurrence probability value P.
[0048] In this embodiment, the initial value X 0 Just as we obtained the interval, we use the formula for solving the quadratic equation to find the initial value X of the modified logistic map. 0 The modified logistic map initial value calculation means calculates the initial value of the modified logistic map. The compression processing device 100 of this embodiment realizes compression processing that traces the modified logistic map backward by executing a program corresponding to the flowchart shown in Fig. 16. In this flowchart, the initial value X 0 An example of the calculation is shown in FIG.
[0049] First, the pre-compression data is read and the data volume N is acquired (S11). In this case, the pre-compression data is "0100", so N=4. Next, the number C of 0s in the N-bit data is counted (S12). In this case, C=3. Next, the initial value parameters are set (S13). Here, the occurrence probability value P of the initial value parameters is P=C / N=3 / 4, and i=(N-1)=3.
[0050] The logistic map equation is generated by the modified logistic map equation generating means 101 as follows. That is, when P is less than 3 / 4, coefficient a=-16 / 9, coefficient b=8 / 3, and coefficient c=0 are substituted into equation (7) to generate equation (5). When P is 3 / 4 or more, coefficient a=-16, coefficient b=24, and coefficient c=8 are substituted into equation (7) to generate equation (6).
[0051] Next, the first bit of the uncompressed data is extracted (S15), and a calculation is performed to trace the modified logistic map back (S16). The initial value X 0 The result of finding the interval is a half-open interval (0.57838655, 0.75) expressed to the eighth decimal point. On the other hand, the initial value X going back through the modified tent map of P=3 / 4 shown in Figure 9 0 The half-open interval of is (0.64453125, 0.75]. In this embodiment, the modified logistic map in FIG. 17 is traced back to the initial value X 0 The interval is different from the value obtained. Note that these initial values X 0 The intervals will be explained in the second embodiment described below, but the double-headed vertical arrows in FIG. 20 indicate intervals determined by the modified tent map, and the double-headed vertical arrows in FIG. 21 indicate intervals determined by the modified logistic map.
[0052] Initial value X 0 The width of the interval is "0.75-0.64453125=0.10546875" in the modified tent map, and "0.75-0.57838655=0.17161345" in the modified logistic map of this embodiment. 0 The initial value X obtained by the modified tent mapping in the section 0 It can be seen that this is 62.7% larger than the interval.
[0053] Conventional arithmetic coding uses the initial value X calculated based on the probability of occurrence of data P. 0 The larger the deviation in the occurrence probability, the larger the initial value X 0 Since the interval can be set large, the initial value X that has the smallest number of bits within the interval is0 It is a reversible compression method that uses it as a compression code. In the compression method that traces back to the Bernoulli shift mapping described in Patent Document 1 shown above, it is shown that the arithmetic code and the compression rate are equivalent. Initial value X 0 The larger the interval of can be taken, the smaller the number of bits of the initial value X that becomes the compression code 0 can be adopted. Depending on the data pattern, it can be expected that the compression bit number can be made smaller by obtaining the interval of the initial value X 0 with the modified logistic mapping.
[0054] In the first step S16, after obtaining the initial value X 0 , it is detected that the number of times i of the backward operation becomes 0 (S17). If NO, it returns to step S15, and steps S15 to S17 are repeated. If YES in step S17, the compression code is finally calculated from the obtained semi-open interval [lower limit X i , upper limit X i ), and the compression code and the dictionary area are output (S19), and the process ends. In the dictionary area, the occurrence probability value P of the code of the data before compression and the initial value X 0 are stored.
[0055] The decompression processing device 200 of the present embodiment executes the decompression processing operation by the modified logistic mapping shown in FIG. That is, the compressed data is taken in, and the occurrence probability value P and the initial value X 0 are taken out from the dictionary area of the compressed data (S21). The formula of the modified logistic mapping is generated from the above P (S22), and the process of obtaining the compression code at that time by the operation of the modified logistic mapping is performed, and the number of processing times i is incremented by one step (S23). Next, when X i <P, 1 bit of the decompression code is output as 0, and X iWhen it is ≧P, output 1 for one bit of the extension code (S24). Next, detect whether the number of processing times i has become N (the number of bits after extension) (S25). If it is NO, return to step S23 to continue the processing. When branching to YES in step S25, it ends. That is, in the processing from step S23 to step 25, when “X i <P”, output bit “0”, and when “P≦X i ”, continuously output bit “1”, output for N bits of the data amount after extension, and end the extension process.
[0056] As a specific example, data with a data capacity to be compressed of 8 bits and an occurrence probability value P = 0.75 is processed in two ways: the modified tent mapping in FIG. 8 and the modified logistic mapping in FIG. 16, and the initial value X 0 is obtained. The range of the initial value X 0 and the compression bit code for performing the compression process on the 8-bit data “01000010” are shown in FIG. 19. When using the modified tent mapping, the initial value X 0 interval is [0.690673828125, 0.701797485351562), and the compression code is for 7 bits of “1011001”. When using the modified logistic mapping, the initial value X 0 interval is [0.683767000917306, 0.702201771440963), and the compression code is for 4 bits of “1011”. As a result, it was found that when using the modified logistic mapping, 3 bits of the compression code can be reduced.
[0057] As another example, an example of compressing 8-bit data “01010101” with an occurrence probability value P = 0.5 is shown. The initial value X 0 interval obtained by the compression process using the tent mapping is [0.3984375, 0.40234375], and the compression code is for 9 bits of “011110101”, and the initial value X 0 = 0.478515625. The initial value X 0The interval is [0.343159129800554,0.348997025340386], and the compressed code is 5 bits of “01011”, and the initial value X 0 = 0.34375. Comparing these results, the modified logistic map reduced the amount of data by 4 bits, resulting in a higher compression rate.
[0058] The formula for solving the quadratic equation uses √, but in the case of the calculation of the modified logistic map in Figure 12, the initial value X 0 The decimal representation of the range is “0.084265193849≦X 0 <0.146446609407”, but when expressed as √
number
[0059] <Embodiment 2> Next, a second embodiment of the present invention will be described. Fig. 1B shows a configuration diagram of a compression processing device 100B and a decompression processing device 200B according to the second embodiment of the present invention. The compression processing device 100B includes a modified logistic map formula generation means 101B, a modified logistic map compression processing means 110, a modified logistic map compressed code generation means 120, a modified logistic map initial value calculation means 130, and a modified logistic map optimal formula acquisition means 140.
[0060] The modified logistic map formula generating means 101B of this embodiment prepares a plurality of occurrence probability values P, and generates a plurality of modified logistic map formulas based on the plurality of occurrence probability values P. The modified logistic map initial value calculating means 130 calculates upper and lower limit values of the compressed value for the first unit data of the decompression code, using the plurality of modified logistic map formulas generated by the modified logistic map formula generating means 101B, to obtain initial values.
[0061] The modified logistic map optimal equation acquisition means 140 compares the number of bits of the initial values corresponding to the modified logistic map equation obtained by the modified logistic map initial value calculation means 130 and obtains the modified logistic map equation which results in the initial value with the smallest number of bits. The modified logistic map compression processing means 110 of this embodiment performs calculations using the modified logistic map equation obtained by the modified logistic map optimal equation acquisition means 140.
[0062] The calculation performed by the modified logistic map compression processing means 110 of this embodiment is the same as that of the first embodiment, that is, the calculation of the upper limit and the lower limit of the compression value for each unit data of the decompression code to obtain the range of the compression value, and this calculation is repeated by the modified logistic map compression processing means 110. Also, the modified logistic map compression code generation means 120 has the same configuration as that of the first embodiment.
[0063] FIG. 7 of the above-mentioned Patent Document 1 shows 16 types of 4-bit data, and the initial value X calculated by the inclination P of the horizontal axis. 0 The graph shows the intervals on the vertical axis. As in this graph, the horizontal axis is set to P, and the deformed tent map is obtained by changing P. Each map is used to calculate the initial value X 0 The intervals are calculated and assigned to the vertical axis in the graph shown in Figure 20. The horizontal axis is also set to P, and the modified logistic map is calculated by changing P. The initial value X 0 Figure 21 shows a graph in which the intervals were determined and assigned to the vertical axis.
[0064] We prepare 16 sets of 4-bit Gray code strings shown in Fig. 7, and compress them by tracing back the modified tent map to obtain the initial value X 0 A graph in which the intervals are assigned to the vertical axis and each P is assigned to the horizontal axis is shown in FIG. 20. For example, in FIG. <6> P and initial value X in the domain 0 When the decompression process is performed by the modified tent mapping according to the flowchart shown in FIG. 10, the <6> The Gray code bit sequence (0111) is always obtained.
[0065] In the flowchart of Fig. 16, the initial value X obtained as a result of performing a retroactive calculation of the modified logistic map on 16 data sets with a data volume of 4 bits is 0 A graph in which the intervals are assigned to the vertical axis and each P is assigned to the horizontal axis is shown in FIG. <6> The slope P and the initial value X in the region 0 When the operation is performed by the modified logistic map according to the flowchart shown in FIG. <6> The bit sequence (0111) is always obtained.
[0066] Next, we present the protocol that defines the dictionary area of compressed data and the file structure of compressed code. In this embodiment, the amount of data handled is 8 bits, and the first 3 bits are used as the dictionary area to store P. As shown in the modified logistic map in Figure 15, seven types of P are used: P = 1 / 8, 2 / 8, 3 / 8, 4 / 8, 5 / 8, 6 / 8, and 7 / 8.
[0067] For example, when compressing "11010010" as 8-bit data, the compression process is performed using the seven types of P shown in Figure 15 to obtain the initial value X 0 Calculate the interval. The initial value X that has the smallest number of bits among the 7 values. 0 is adopted as the compression code. In this example, P is selected as 5 / 8 (=0.625) and the bit value is “101”. The compression code is two bits, “11”, and the initial value X 0= 0.75. Therefore, by concatenating the 3-bit dictionary area P value "101" with the compression code "11", the compressed data becomes "10111". The 8 bits of data before compression described above become 5 bits after compression, so the compression ratio is 5 / 8 = 0.625.
[0068] In the compression method described in the first embodiment, the probability value P of occurrence of a bit "1" or "0" in the data is set to 0.5, so the bit string is "100" (3 bits). 0 The initial value X that has the smallest number of bits when the interval is found 0 becomes "0.671875", and the bit string becomes "101011", which is 6 bits. When concatenated, it becomes "100101011", which is 9 bits, and the compression ratio is 9 / 8=1.125, so P=0.5 results in a greater number of compressed bits than P=0.625. Changing P to 5 / 8 (=0.625) is an example of a good compression ratio, as it is 4 bits smaller in comparison.
[0069] In the second embodiment, the specifications of the compressed file for compressing 8 bits of data are such that 3 bits are reserved for storing P as a dictionary area, and the 4th bit and after are the initial value X 0 The size of the data to be compressed is handled as a power of 2, such as 8 bits → 16 bits → 32 bits → ...
[0070] The amount of information for P is 3 bits → 4 bits → 5 bits → ..., so the amount of information for P is E bits, and the uncompressed data is 2 E When expanding, the number of bits of compressed data is determined in advance, and the number of mappings is set to “2 E -1” to restore the decompressed code.
[0071] Others are defined as follows. These are for cases where the data to be compressed is all "0" and cases where it is all "1". For example, in the case of an 8-bit data capacity, since P is defined as having 7 values as above, the 3 bits of the area storing P are unused as "000". For this reason, the header part of the compression is set to "000", and the initial value X from the 4th bit onwards is 0 It is defined that when the data is all "0", a "0" is stored, and when the data is all "1", one bit of "1" is stored.
[0072] 22 shows a flowchart of the process performed by the compression processor 100B. This flowchart shows the process of changing P of the modified logistic map to obtain the initial value X 0 The operation of the compression process performed by the modified logistic map optimal equation acquisition means 140 to search for the optimal logistic map is included. As described above, the case where the pre-compression data is not all "0" or "1" is the subject. When all data is "0" or "1", P is set to "0" for E bits, and the initial value X 0 A separate routine is provided to set one bit to “0” when all bits are “0”, and to set one bit to “1” when all bits are “1”, and to compress the file.
[0073] The process performed by the compression processor 100B will be described with reference to the flow chart of Fig. 22. First, the uncompressed data is read and the data volume N is taken in, and the number k of equations of the modified logistic map to be generated is set to 1, and the initial compression code C 0 2 E+1 Then, the number of times the mapping is traced back is set to (N-1) (S31). Furthermore, the occurrence probability value P is set to k / 2 E (S32), and generate two quadratic polynomials that trace the modified logistic map back from P (S33).
[0074] Next, the first bit of the uncompressed data is extracted (S34), a calculation is performed by tracing back the modified logistic map, i is incremented by one (S35), and it is detected whether i has become 0 (S36). If the result in step S36 is NO, steps S34 to S36 are repeated. If the result in step S36 is YES, the half-open interval [lower limit X0 , upper limit X 0 ) to get the compressed code (initial value X 0 ) and the compression bit rate is C k (S37), k is incremented by one (S38), and it is detected whether the calculation has been performed until k reaches the final value E (S39). If the result is NO in step S39, the process returns to step S32 to continue the process, and if the result is YES in step S39, the process proceeds to step S40 to 0 ,C 1 ,C 2 ,···,C E The smallest number of compressed codes Ck among them is detected (S40), and a compressed code is obtained based on the pre-compression data using the modified logistic map equation which resulted in this number of compressed codes Ck. A compressed code which stores data indicating the required data used in the dictionary area together with the actual compressed code is output (S40A).
[0075] Next, the decompression processing device 200B will be described. The decompression processing device 200B includes a modified logistic map formula generating means 201, a modified logistic map decompression processing means 210, and a modified logistic map decompression code generating means 220, similar to the decompression processing device 200 of the first embodiment. The decompression processing device 200B operates according to the flowchart shown in Fig. 18, which has already been described. Here, an example of decompression processing will be described in which the compressed data obtained by compressing the pre-compression data "11010010" is "10111".
[0076] The first three bits of the compressed code are set to P. The fourth and subsequent bits are set to the initial value X. 0 Here, since the bit code of P is “101”, set P=5 / 8(=0.625). Initial value X 0 becomes 2 bits of “11”, and the initial value X 0 Select =0.75.
[0077] Since P=5 / 8, the following two equations (18) are generated as the modified logistic map, similarly to the first embodiment.
[0078]
number
[0079] In order to see the effect of compression with varying P, an experiment was carried out in which data volume to be compressed was compressed with 254 patterns excluding two patterns, "00000000" where all 8 bits are 0, and "11111111" where all 8 bits are 1, and the average data volume after compression was obtained. The graph in Fig. 24 shows the results of a comparison between compression of P as the occurrence probability of "0" in the first embodiment with compression of P as the occurrence probability of "0" in 254 patterns excluding data with all 8 bits being "0" and all being "1", and compression of P by varying in the second embodiment. In Fig. 24, the horizontal axis shows 8-bit data from 0 to 255, and the vertical axis shows the initial value X after each 8-bit data is compressed. 0 The numbers indicate the number of bits.
[0080] Data with all bit values of "0" becomes "0" shown in the "speech bubble" in Fig. 24, and as described above, the compressed code is "0" and becomes 1 bit. Data with all bit values of "1" becomes "170" shown in the "speech bubble" in Fig. 24, and as described above, the compressed code is "1" and becomes 1 bit. The solid line shows the results when P is varied, and the dotted line shows the results when P shown in embodiment 1 is set to the occurrence probability of "0". Overall, it can be seen that varying P tends to result in fewer bits after compression, resulting in a higher compression rate.
[0081] The average number of bits after compression is examined. In the method of the first embodiment shown by the dotted line in FIG. 0The average compressed code for the data was 6.65 bits, while the method of varying P, which is the method of this embodiment 2 shown by the solid line in Fig. 24, resulted in an average compressed code of 5.33 bits. In comparison, this embodiment reduces the number of bits by an average of 1.32 bits, confirming the effect of increasing the compression ratio by varying P. The method for calculating the average value was "total number of compressed bits / 256" for the number of compressed bits on the vertical axis for each data on the horizontal axis in Fig. 24. The number of bits of the average compressed code was calculated using the initial value X excluding the 3 bits of the header part P. 0 This only calculates the size of the code in the compressed file. The average bit size of the entire compressed file is 3 bits more.
[0082] In the second embodiment, the amount of data to be compressed is 8 bits. P of the dictionary area is 3 bits, and the ratio is 3 / 8, which is a large ratio, occupying 37.5%. When the amount of data to be compressed is 16 bits, it is 4 / 16=0.25, and when it is 32 bits, it is 5 / 32=0.15625. As the amount of data to be compressed increases, the ratio of the dictionary area tends to decrease, and therefore it is estimated that the compression effect will be increased.
[0083] <Embodiment 3> Next, a third embodiment of the present invention will be described. In the second embodiment, P is varied, and the initial value X 0 However, the initial value X of the logistic map with P=0.5 shown in Figure 7 is 0 When observing the section, "0 <1> " and "1 <16> The closer the section is to the initial value X 0 Since the interval is smaller than the others, the initial value X 0 It is inferred that the amount of bits in the “0 <1> " and "1 <16> The closer the area is to the initial value X 0 It was observed that the compression ratio becomes worse because the range of 0 and 255 is narrow. From the experimental results in Fig. 24, the initial value X 0 It can be seen that the bit amount of
[0084] Here, in Fig. 7 <1> ~ <4> The first two bits of the area are “00”. <13> ~ <16> Let's focus on the Gray code that results in "10" in the region. <1> ~ <4> If we invert the bits in the region <9> ~ <12> For example, <3> The bit value “0011” becomes “1100” when inverted. <9> The bit value is the same as that of <13> ~ <16> If we invert the bits in the region <5> ~ <8> Focusing on this point, we consider providing a one-bit code that determines whether or not to compress by bit inversion as the dictionary area of the compression code.
[0085] The first two bits of the data to be compressed are checked, and if they are "00" or "10", a one-bit inversion discrimination code in the dictionary area is set to "1" as described later, and the data is inverted before compression. By carrying out compression in this manner, the initial value after compression, X 0 The code is shown in Figure 7. <5> ~ <12> Since the interval is selected, the initial value X 0 It is expected that the compression ratio can be increased by expanding the space.
[0086] 1C shows a configuration diagram of a compression processing device 100C and a decompression processing device 200C according to a third embodiment of the present invention. The compression processing device 100C includes a modified logistic map formula generation means 101B, a modified logistic map compression processing means 110, a modified logistic map compressed code generation means 120, a modified logistic map initial value calculation means 130, a modified logistic map optimal formula acquisition means 140, and a code conversion means 150.
[0087] The code conversion means 150 converts the decompressed code data into a Gray code, divides the decompressed code data represented by the converted Gray code into a plurality of ranges according to size, and performs code conversion to change the decompressed code data at the ends of the ranges to decompressed code data at the center of the ranges. The modified logistic map formula generation means 101B, the modified logistic map compression processing means 110, the modified logistic map compressed code generation means 120, the modified logistic map initial value calculation means 130, and the modified logistic map optimal formula acquisition means 140 perform processing using the code changed by the code conversion means 150.
[0088] FIG. 25 shows a flowchart of a process example related to the code conversion (bit inversion) by the code conversion means 150 of the third embodiment. E The first bit of the pre-compression data (bits) is extracted (S41) and it is detected whether the first bit of the pre-compression data is "00" or "10" (S42). If the pre-compression data is binary code, it is converted into gray code. If the first bit is "00" or "10", the pre-compression data is inverted between 1 and 0. If inversion has been performed, "INV=1" is set, and if not, "INV=0" is set in the dictionary area of the compressed code (S43, S44).
[0089] As described above, the compression process after the code conversion process is performed is performed as shown in the flowchart of Fig. 22, as in the second embodiment. That is, P is varied and the same process as in the second embodiment is performed, and after compression, 1 bit of the INV value is placed at the beginning as shown in Fig. 26, and the second bit and after are dictionary areas that store E bits of P values. The "E+2" bit and after are set to the initial value X as the compressed code. 0 It stores the above and becomes a compressed file (compressed code).
[0090] An experiment was carried out in which the invention of the third embodiment, which judges whether or not to perform bit inversion on pre-compression data by referencing the first two bits, was applied to pre-compression data with a bit amount of 8 bits to obtain the average value of the compressed code. In this experiment, compression was performed by the modified logistic map while varying P for 254 patterns, excluding two patterns: an example of a pattern in which all bits are "0" such as "00000000", and an example of a pattern in which all bits are "1" such as "11111111", and the average value of the compressed code was obtained. The results of the above experiment are shown in the graph in Fig. 27. The horizontal axis of the graph shows 8-bit data from 0 to 255, and the vertical axis shows the initial value X after each 8-bit data is compressed. 0 The number of bits indicates the number of bits per second.
[0091] In FIG. 27, the result of performing the inversion process is shown by a solid line, and the result of not performing the inversion process but only varying P as in the solid line of FIG. 24 is shown by a dashed line. When the bit values are all "0", it becomes the "0" part shown in the balloon of FIG. 27, and the compressed code becomes "0" as described above as 1 bit. When the bit values are all "1", it becomes the "170" part shown in the balloon of FIG. 27, and the compressed code becomes "1" as described above as 1 bit. Also, as shown in FIG. 27, the compressed code is the same between "64 and 191" because the inversion process is not performed. It can be seen that the compressed code amount is smaller between 0 and 63 and between 192 and 255, where the inversion process is performed, than when the inversion process is not performed, and it can be seen that the compression is effective.
[0092] As a result of the above experiment, the initial value after compression X 0The average data amount was 4.76 bits when inversion processing was performed, and was 5.33 bits on average in the example where only the compression method using the logistic map that varies P as in the second embodiment was adopted. It was found that the inversion processing resulted in a reduction effect of 0.57 bits. Compared to the second embodiment, since 1 bit of information was added to the dictionary area, the average value was 5.76 bits, which was a larger result than the variation of P, but in the third embodiment, the amount of information to be compressed is small at 8 bits, so the proportion of 1 bit for the dictionary area is large. As mentioned above, since the proportion of the dictionary area decreases as the amount of information of the data to be compressed increases, it is expected that the addition of 1 bit will produce a compression effect.
[0093] Next, the decompression processing device 200C will be described. The decompression processing device 200C includes a modified logistic map formula generating means 201, a modified logistic map decompression processing means 210, and a modified logistic map decompression code generating means 220, similar to the decompression processing device 200B of the second embodiment. The decompression processing device 200C operates according to the flowchart shown in FIG.
[0094] The operation of the decompression process will be described with reference to the flow chart shown in Fig. 28. First, the compressed data is extracted, and the inversion code data INV of the first bit from the beginning of this compressed data is extracted, and the next P and the next initial value X are also extracted. 0 Next, the equation of the modified logistic map corresponding to P is generated (S52), and the initial value X 0 Here, N indicates the number of bits of the data after expansion (S53).
[0095] When the actual decompression process is completed as described above, INV is extracted from the dictionary area at the beginning of the compressed data, and it is detected whether INV is "1" (S54). If the result is YES in step S54, the bit of the decompressed data is inverted (S55). When this step S55 is completed, or if the result is NO in step S54, the decompressed data is output (S56), and the process ends. In the above decompression process, if the first bit is "0" due to the data format of the compressed file shown in FIG. 26, E bits are extracted from the second bit and the "3+E" bits and after are set to the initial value X 0 Then, the process corresponding to the flowchart shown in FIG. 18 is performed to decompress the data. When the first bit is 1, E bits of P are extracted from the second bit, and the bits after “3+E” are set to the initial value X. 0 Then, the process corresponding to the flowchart shown in FIG. 18 is carried out to decompress the data, and the decompressed data is then inverted.
[0096] <Embodiment 4> Next, a fourth embodiment of the present invention will be described. In this fourth embodiment, a compression ratio is increased by selecting a compression algorithm, either a modified logistic map or a modified tent map. FIG. 1D shows a configuration diagram of a compression processing device 100D and a decompression processing device 200D according to the fourth embodiment of the present invention. The compression processing device 100D according to the fourth embodiment includes a modified logistic side processing unit 500, a modified tent side processing unit 600, and an optimal mapping formula comparison result acquisition unit 700.
[0097] The modified logistic map processing section 500 includes a modified logistic map formula generating means 101D, a modified logistic map compression processing means 110, a modified logistic map compressed code generating means 120, a modified logistic map initial value calculating means 130, and a modified logistic map optimal formula acquiring means 140.
[0098] The modified logistic map formula generating means 101D divides the horizontal axis section from 0 to 1 into two sections according to the occurrence probability value P of the code constituting the decompressed code data before compression processing, and generates a formula of the modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P, and a second quadratic function equal to or greater than the occurrence probability value P. The modified logistic map compression processing means 110 iteratively performs a calculation to calculate the upper and lower limit values of the compressed value for each unit data of the decompressed code and obtain the range of the compressed value. The modified logistic map compressed code generating means 120 generates a compressed code from the range of the compressed value finally obtained by the modified logistic map compression processing means. The modified logistic map formula generating means 101D uses a plurality of the occurrence probability values P prepared, and generates a plurality of modified logistic map formulas based on the plurality of the occurrence probability values P.
[0099] The modified logistic map initial value calculation means 130 calculates the upper and lower limit values of the compressed value for the first unit data of the decompression code, using the multiple modified logistic map equations generated by the modified logistic map equation generation means 101D, to obtain initial values. The modified logistic map optimal equation acquisition means 140 compares the number of bits of the initial values corresponding to the modified logistic map equations obtained by the modified logistic map initial value calculation means 130, and obtains the modified logistic map equation that results in the initial value with the smallest number of bits.
[0100] The modified tent side processing unit 600 includes a modified tent mapping equation generating means 601 , a modified tent mapping compression processing means 610 , a modified tent mapping compressed code generating means 620 , a modified tent mapping initial value calculating means 630 , and a modified tent mapping optimal equation obtaining means 640 .
[0101] The modified tent mapping equation generating means 601 divides the horizontal axis section from 0 to 1 into two sections based on the occurrence probability value P of the code constituting the decompressed code data before compression processing, and generates a modified tent mapping equation consisting of a first linear function from 0 to less than the occurrence probability value P, and a second linear function equal to or greater than the occurrence probability value P. The modified tent mapping compression processing means 610 iteratively calculates the upper and lower limit values of the compressed value for each unit data of the decompressed code, and obtains the range of the compressed values. The modified tent mapping compressed code generating means 620 generates a compressed code from the range of compressed values finally determined by the modified tent mapping compression processing means.
[0102] The modified tent mapping initial value calculation means 630 calculates an upper limit and a lower limit of the compressed value for the first unit data of the decompression code to obtain an initial value, using a plurality of modified tent mapping equations generated by the modified tent mapping equation generation means 601. The modified tent mapping optimal equation acquisition means 640 compares the number of bits of the initial value corresponding to the modified tent mapping equations obtained by the modified tent mapping initial value calculation means 630, and obtains the modified tent mapping equation which results in the initial value with the smallest number of bits.
[0103] The optimal mapping formula comparison result acquisition means 700 does not belong to either the modified logistic side processing unit 500 or the modified tent side processing unit 600. However, the optimal mapping formula comparison result acquisition means 700 uses the calculation result by the modified logistic mapping initial value calculation means 130 of the modified logistic side processing unit 500 and the calculation result by the modified tent mapping initial value calculation means 630 of the modified tent side processing unit 600. That is, the optimal mapping formula comparison result acquisition means 700 compares the initial value having the smallest number of bits among the initial value number of bits corresponding to the modified logistic map formula obtained by the modified logistic map initial value calculation means 130 with the initial value having the smallest number of bits among the initial value number of bits corresponding to the modified tent map formula obtained by the modified tent map initial value calculation means 630, determines the mapping formula that resulted in the smaller initial value in the comparison result, and determines the mapping formula obtained by the modified logistic map optimal formula acquisition means 140 or the modified tent map optimal formula acquisition means 640 that corresponds to the mapping formula.
[0104] The compression processor 100D according to the fourth embodiment of the present invention performs calculations using the modified logistic map equation or the modified tent map equation found by the optimum mapping equation comparison result acquisition means 700.
[0105] 29 shows a flowchart of the process performed by the compression processing device 100D according to the fourth embodiment of the present invention, and the procedure of the process will be described. First, the uncompressed bits are extracted, P is varied to generate an equation for the modified tent map, and the compression process is performed using this equation to obtain the initial value TX 0 Then, the value TP of P at that time is obtained (S61). Next, P is varied to generate an equation for the modified logistic map, and this equation is used to perform compression processing using the uncompressed bits, and the initial value LX that results in the smallest amount of data is obtained. 0 Then, the value LP of P at that time is obtained (S62). Of course, either step S61 or step S62 may be carried out first.
[0106] When steps S61 and S62 are completed, TX 0 Data volume and LX 0The amount of data in TX is compared with that in TX (S63). 0 The amount of data is LX 0 If the amount of data is less than the amount of data in LX, the selection information of the modified tent map and the compressed code up to the initial value are output (S64), and 0 The amount of data is TX 0 If the data amount is less than the amount of data, the selection information of the modified logistic map and the compressed code up to the initial value are output (S65). The output of either the selection information of the modified tent map or the selection information of the modified logistic map is written as "0" or "1" in the first bit of the dictionary area. After step S64 or step S65, the remaining uncompressed bits are compressed using the selection information of the modified tent map and the selected mapping of the modified logistic map, and output (S66).
[0107] The bit capacity of the pre-compression bits is set to 8 bits, and 254 types of data excluding “00000000” and “11111111” are compressed according to the fourth embodiment to obtain the initial value X 0 An experiment was conducted to obtain the average number of compressed bits of the 8-bit data. The compression according to the fourth embodiment is compression by modified tent mapping or modified logistic mapping that varies P. Figure 30 shows a graph of the results. The horizontal axis represents 8-bit data from 0 to 255, and the vertical axis represents the initial value X after each 8-bit data is compressed. 0 The solid line shows the compression result using the modified logistic map with P varying, and the dotted line shows the compression result using the modified tent map with P varying.
[0108] In the compression method using the modified logistic map that varies P, which corresponds to the solid line in Figure 30, the initial value X 0 The average was 5.33 bits. In the compression method using the modified tent mapping with P varying, which corresponds to the dotted line, the initial value X 0 The average was 5.73 bits. Comparing these, the modified logistic map was 0.4 bits smaller.
[0109] It is also possible to use a program that makes the modified logistic map and modified tent map of the fourth embodiment compete with each other, compare the two compression bit numbers on the vertical axis of the graph in FIG. 30, and adopt the program that results in the smaller bit number. 0 The average bit length of the compressed code was 4.63 bits. In the inversion processing method of the third embodiment, the average was 4.76 bits. Therefore, the fourth embodiment was smaller than the third embodiment by 0.13 bits.
[0110] Next, the decompression processing device 200D will be described. The decompression processing device 200D is provided with a mapping formula determination unit 280, a modified logistic side decompression unit 800A, and a modified tent side decompression unit 800B. The mapping formula determination unit 280 refers to the first bit of the dictionary area to determine whether to use a modified logistic mapping or a modified tent mapping, and further specifies P using E bits from the second bit onwards, specifies the mapping formula used in the compression process, and sends it to the modified logistic side decompression unit 800A or the modified tent side decompression unit 800B. That is, when the first bit of the dictionary area indicates a modified logistic mapping, it is sent to the modified logistic side decompression unit 800A, and when the first bit of the dictionary area indicates a modified tent mapping, it is sent to the modified tent side decompression unit 800B.
[0111] The only difference between the modified logistic side decompression unit 800A and the modified tent side decompression unit 800B is whether the processing is performed based on the modified logistic map equation or the modified tent map equation. The modified logistic side decompression unit 800A is provided with a modified logistic map equation generation means 201, a modified logistic map decompression processing means 210, and a modified logistic map decompression code generation means 220, as with the decompression processing device 200B of the second embodiment. The modified tent side decompression unit 800B is provided with a modified tent map equation generation means 801 whose processing mapping equation is different from that of the modified tent map equation generation means 201, a modified tent map decompression processing means 810 whose processing mapping equation is different from that of the modified logistic map decompression processing means 210, and a modified tent map decompression code generation means 820 whose processing mapping equation is different from that of the modified logistic map decompression code generation means 220.
[0112] The decompression processor 200D operates according to the flowchart shown in Fig. 31. The operation will be described based on this flowchart. In the decompression process, the first bit of the compressed file is referenced to determine whether to use the modified logistic map or the modified tent map. From the second bit onwards, P is set for E bits, and from the "2+E"th bit onwards, the initial value X 0 Therefore, the compressed data is extracted, the first bit is discriminated, and P is extracted, and the initial value X 0 Then, it is detected whether the first bit is "1" (S72).
[0113] If the answer is YES in step S72, a formula for the modified logistic map corresponding to P is generated (S73), and the initial value X 0 Then, the decompression process is performed for the number of bits N of the decompressed data using the equation of the modified logistic map (S74). On the other hand, if the result in step S72 is NO, the equation of the modified tent map corresponding to P is generated (S75), and the initial value X 0 is set and the modified tent map equation is used to perform decompression processing for the number of bits N of the decompressed data (S76). Following steps S74 and S76, the decompressed pre-compression data for N bits is output (S77).
[0114] To summarize the above briefly, when the first bit of the compressed data is “0”, P is set to the modified tent map of equation (3). The initial value X 0 Then, the decompression process is performed by setting X. Also, when the first bit of the compressed data is "1", a modified logistic map formula is generated from P. The initial value X 0 The fourth embodiment solves the case where the compression ratio using the modified tent map is larger than the compression ratio using the modified logistic map formula, depending on the data pattern to be compressed.
[0115] As described above, according to each embodiment of the present invention, the modified logistic map is nonlinear and the initial value X 0Since the interval is different from that of conventional arithmetic codes, it can be said that the compression rate can be increased by the data pattern. Also, by adopting a mechanism for varying the control parameter P of the modified logistic map, a higher compression rate can be achieved. In coding compression, data with information entropy close to 1 cannot be compressed to a value lower than the original data amount, but each embodiment of the present invention uses an arithmetic formula, so a compression effect can be expected. [Explanation of symbols]
[0116] 100, 100B, 100C, 100D Compression processing device 101, 101B, 101D Modified logistic map formula generating means 110 Modified logistic map compression processing means 120 Modified logistic map compressed code generator 130 Modified logistic map initial value calculation means 140 Method for obtaining optimal equation for modified logistic map 150 Code conversion method 200, 200B, 200C, 200D Decompression processing device 201 Modified logistic map formula generating means 210 Modified logistic map extension processing means 220 Modified logistic map decompression code generation means 280 Mapping Formula Determination Unit 300 Storage area for extension cord 400 Compressed code storage 500 Transformed Logistics Side Processing Unit 600 Deformed tent side processing section 601 Deformed tent mapping generation means 610 Transformed tent mapping compression processing means 620 Transformed Tent Mapping Compressed Code Generator 630 Deformed tent mapping initial value calculation means 640 Means for obtaining optimal equation for deformation tent mapping 700 Means for obtaining optimal mapping formula comparison results 800A modified logistic side extension 800B Deformed tent side extension 801 Deformed tent mapping generator 810 Deformed tent map extension processing means 820 Modified tent mapping decompression code generator
Claims
1. a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function equal to or greater than the occurrence probability value P; a modified logistic map compression processing means for repeatedly calculating an upper limit and a lower limit of a compression value for each unit data of the decompressed code data to obtain a range of the compression value; a modified logistic map compressed code generating means for generating a compressed code from the range of compressed values finally obtained by the modified logistic map compression processing means; A compression processing device comprising:
2. The modified logistic map compression processing means calculates the upper limit and the lower limit of the compression value for each unit data by using a quadratic equation aX 2 Formula to find the solution of +bX+c=0 [0010] 2. The compression processing apparatus according to claim 1, wherein a solution is obtained using the above formula and the calculation is performed using the solution.
3. the modified logistic map formula generating means prepares a plurality of occurrence probability values P, and generates a plurality of modified logistic map formulas based on the plurality of occurrence probability values P; a modified logistic map initial value calculation means for calculating an upper limit and a lower limit of a compression value for a first unit data of the decompressed code data by using a plurality of equations of the modified logistic map generated by the modified logistic map equation generation means, thereby obtaining an initial value; a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the equations of the modified logistic map obtained by the modified logistic map initial value calculation means and obtaining the equation of the modified logistic map which has resulted in an initial value with the smallest number of bits; having 2. The compression processing device according to claim 1, wherein the modified logistic map compression processing means performs calculations using the equation of the modified logistic map obtained by the modified logistic map optimum equation acquisition means.
4. a code conversion means for converting the expanded code data into a gray code, dividing the expanded code data represented by the converted gray code into a plurality of ranges according to size, and changing the expanded code data at the ends of the ranges into expanded code data at the center of the ranges; 4. The compression processing device according to claim 3, wherein the modified logistic map formula generating means, the modified logistic map compression processing means, the modified logistic map initial value calculating means, and the modified logistic map optimal formula obtaining means perform processing using the code changed by the code converting means.
5. a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function equal to or greater than the occurrence probability value P; a modified logistic map compression processing means for repeatedly calculating an upper limit and a lower limit of a compression value for each unit data of the decompressed code data to obtain a range of the compression value; a modified logistic map compressed code generating means for generating a compressed code from the range of compressed values finally obtained by the modified logistic map compression processing means; A modified logistics side processing unit configured to include a modified tent mapping formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generating a modified tent mapping formula comprising a first linear function from 0 to less than the occurrence probability value P and a second linear function equal to or greater than the occurrence probability value P; a modified tent mapping compression processing means for repeatedly calculating an upper limit and a lower limit of a compression value for each unit data of the decompressed code data to obtain a range of the compression value; a modified tent map compressed code generating means for generating a compressed code from the range of compressed values finally determined by the modified tent map compression processing means; A deformed tent side processing section configured to include Equipped with the modified logistic map formula generating means uses a plurality of prepared occurrence probability values P to generate a plurality of modified logistic map formulas based on the plurality of occurrence probability values P; the modified tent mapping equation generating means uses a plurality of prepared occurrence probability values P to generate a plurality of modified tent mapping equations based on the plurality of occurrence probability values P; The modified logistics side processing section includes: a modified logistic map initial value calculation means for calculating an upper limit and a lower limit of a compression value for a first unit data of the decompressed code data by using a plurality of equations of the modified logistic map generated by the modified logistic map equation generation means, thereby obtaining an initial value; a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the equations of the modified logistic map obtained by the modified logistic map initial value calculation means and obtaining the equation of the modified logistic map which has resulted in an initial value with the smallest number of bits; is provided, The deformed tent side processing section includes: a modified tent mapping initial value calculation means for calculating an upper limit value and a lower limit value of a compressed value for a first unit data of the decompressed code data by using a plurality of modified tent mapping equations generated by the modified tent mapping equation generation means, thereby obtaining an initial value; a modified tent map optimal equation acquisition means for comparing the number of bits of initial values corresponding to the modified tent map equation obtained by the modified tent map initial value calculation means and determining the modified tent map equation that results in the smallest number of bits; is provided, an optimum mapping formula comparison result acquisition means for comparing an initial value having the smallest number of bits among the initial values corresponding to the modified logistic map formula obtained by said modified logistic map initial value calculation means with an initial value having the smallest number of bits among the initial values corresponding to the modified tent map formula obtained by said modified tent map initial value calculation means, to obtain a mapping formula which has brought about the smaller initial value as a result of the comparison, and for obtaining a modified logistic map optimum formula acquisition means corresponding to said mapping formula or a mapping formula acquired by the modified tent map optimum formula acquisition means, A compression processing device characterized in that a calculation is performed using the modified logistic map equation or the modified tent map equation obtained by said optimum mapping equation comparison result acquisition means.
6. a modified logistic map formula generating means for dividing the probability of occurrence P used in the compression process into two intervals and generating a formula of a modified logistic map including a first quadratic function from 0 to less than the probability of occurrence P and a second quadratic function equal to or greater than the probability of occurrence P; a modified logistic map decompression processing means for performing an iterative calculation, which repeats a one-unit data acquisition calculation, which is a calculation for obtaining a value corresponding to one unit data of decompressed code data by performing an operation using a modified logistic map formula on the compressed code, and a one-unit data acquisition calculation, which is a calculation for obtaining a next one unit data by using the obtained value in an operation using the modified logistic map formula, until a predetermined number of times; a modified logistic map decompression code generating means for converting a value obtained for each unit data acquisition calculation of the modified logistic map decompression processing means into one unit data of the decompression code data based on a coefficient of the modified logistic map equation; 13. A decompression processing device comprising:
7. 7. The decompression processing device according to claim 6, further comprising means for extracting P contained in the compressed code and sending it to said modified logistic map formula generating means.
8. A computer that realizes data compression processing is a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generating a formula of a modified logistic map constituted by a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function equal to or greater than the occurrence probability value P; a modified logistic map compression processing means for repeatedly calculating an upper limit and a lower limit of a compression value for each unit data of the decompressed code data to obtain a range of the compression value; a modified logistic map compressed code generating means for generating a compressed code from the range of compressed values finally determined by said modified logistic map compression processing means; A compression processing program characterized by causing the program to function as follows.
9. When the computer, as the modified logistic map compression processing means, calculates the upper limit and the lower limit of the compression value for each unit data, the quadratic equation aX 2 Formula to find the solution of +bX+c=0 [0025] 9. The compression processing program according to claim 8, further comprising: a function for obtaining a solution using the above formula (1) and performing calculations using the solution.
10. the computer functions as the modified logistic map formula generating means to prepare a plurality of occurrence probability values P and generate a plurality of modified logistic map formulas based on the plurality of occurrence probability values P; The computer further comprises: a modified logistic map initial value calculation means for calculating an upper limit value and a lower limit value of a compression value for a first unit data of the decompressed code data by using a plurality of modified logistic map equations generated by the modified logistic map equation generation means, thereby obtaining initial values; a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the equations of the modified logistic map obtained by the modified logistic map initial value calculation means and determining the equation of the modified logistic map which results in an initial value with the smallest number of bits; It functions as 9. The compression processing program according to claim 8, wherein the computer functions as the modified logistic map compression processing means to perform a calculation using the equation of the modified logistic map obtained by the modified logistic map optimal equation acquisition means.
11. The computer includes: functioning as a code conversion means for converting the expanded code data into a Gray code, dividing the expanded code data represented by the converted Gray code into a plurality of ranges according to size, and changing the expanded code data at the ends of the ranges into expanded code data at the center of the ranges; 11. The compression processing program according to claim 10, wherein the computer functions to perform processing using the code changed by the code conversion means as the modified logistic map equation generation means, the modified logistic map compression processing means, the modified logistic map initial value calculation means, and the modified logistic map optimal equation acquisition means.
12. A computer that realizes data compression processing is a modified logistic map formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generating a formula of a modified logistic map constituted by a first quadratic function from 0 to less than the occurrence probability value P and a second quadratic function equal to or greater than the occurrence probability value P; a modified logistic map compression processing means for repeatedly calculating an upper limit and a lower limit of a compression value for each unit data of the decompressed code data to obtain a range of the compression value; a modified logistic map compressed code generating means for generating a compressed code from the range of compressed values finally determined by said modified logistic map compression processing means; to realize a modified logistics side processing unit; Furthermore, the computer a modified tent mapping formula generating means for dividing a horizontal axis section from 0 to 1 into two sections based on an occurrence probability value P of a code constituting the decompressed code data before compression processing, and generating a modified tent mapping formula comprising a first linear function from 0 to less than said occurrence probability value P and a second linear function equal to or greater than said occurrence probability value P; a modified tent mapping compression processing means for repeatedly calculating an upper limit and a lower limit of a compression value for each unit data of the decompressed code data to obtain a range of the compression values; a modified tent mapping compressed code generating means for generating a compressed code from the range of compressed values finally determined by said modified tent mapping compression processing means; and functioning as a deformation tent side processing part, the computer causes the modified logistic map formula generation means to function to generate a plurality of modified logistic map formulas based on the plurality of occurrence probability values P by using the plurality of occurrence probability values P; the computer causes the modified tent mapping equation generating means to function to generate a plurality of modified tent mapping equations based on the plurality of occurrence probability values P by using the plurality of occurrence probability values P; The computer causes the modified logistics side processing unit to a modified logistic map initial value calculation means for calculating an upper limit value and a lower limit value of a compression value for a first unit data of the decompressed code data by using a plurality of modified logistic map equations generated by the modified logistic map equation generation means, thereby obtaining initial values; a modified logistic map optimal equation acquisition means for comparing the numbers of bits of initial values corresponding to the equations of the modified logistic map obtained by the modified logistic map initial value calculation means and determining the equation of the modified logistic map which results in an initial value with the smallest number of bits; and causing the device to function so as to include The computer causes the deformation tent side processing unit to a modified tent mapping initial value calculation means for calculating an upper limit value and a lower limit value of a compressed value for a first unit data of the decompressed code data by using a plurality of modified tent mapping equations generated by the modified tent mapping equation generation means, thereby obtaining an initial value; a modified tent map optimal equation acquisition means for comparing the number of bits of initial values corresponding to the modified tent map equation obtained by the modified tent map initial value calculation means and determining the modified tent map equation which results in the initial value with the smallest number of bits; and causing the device to function so as to include the computer functions as an optimum mapping formula comparison result acquisition means which compares an initial value having the smallest number of bits among the initial values corresponding to the modified logistic map formula obtained by the modified logistic map initial value calculation means with an initial value having the smallest number of bits among the initial values corresponding to the modified tent map formula obtained by the modified tent map initial value calculation means, to obtain a mapping formula which has brought about the smaller initial value as a result of the comparison, and which acquires a modified logistic map optimum formula acquisition means corresponding to the mapping formula or a mapping formula acquired by the modified tent map optimum formula acquisition means, The computer is a compression processing program characterized in that it causes each means of the modified logistic side processing unit or each means of the modified tent side processing unit to function to perform calculations using the modified logistic map equation or the modified tent map equation obtained by the optimal map equation comparison result acquisition means.
13. A computer that decompresses the compressed data. a modified logistic map formula generating means for dividing the probability of occurrence P used in the compression process into two intervals and generating a formula of a modified logistic map composed of a first quadratic function from 0 to less than the probability of occurrence P and a second quadratic function equal to or greater than the probability of occurrence P; a modified logistic map decompression processing means for performing an iterative calculation, which repeats a one-unit data acquisition calculation, which is a calculation for obtaining a value corresponding to one unit data of the decompressed code data by performing an operation using a modified logistic map formula on the compressed code, and a one-unit data acquisition calculation, which is a calculation for obtaining a next one unit data by using the obtained value in an operation using the modified logistic map formula, until a predetermined number of times; a modified logistic map decompression code generating means for converting a value obtained for each unit data acquisition calculation of the modified logistic map decompression processing means into one unit data of the decompression code data based on a coefficient of the modified logistic map formula; 23. A decompression processing program comprising:
14. 14. The decompression program according to claim 13, wherein the computer is caused to function as a means for extracting P included in a compressed code and sending it to the modified logistic map formula generating means.
Citation Information
Patent Citations
Compression method of natural picture data
JP2004129179A
Probability type arithmetic element and probability type arithmetic device using the same
JP2004171552A
Data encoder and data encoding communication device
JP2009033474A
Compression processing device, decompression processing device, compression processing program, and decompression processing program
JP2019047450A
Random number generator
JP2019168963A
