WO2023001659A1

WO2023001659A1 - Medium able to be recognized optically by a user, featuring digital data and the means for decoding them

Info

Publication number: WO2023001659A1
Application number: PCT/EP2022/069595
Authority: WO
Inventors: Vincent Joguin
Original assignee: Eupalia
Priority date: 2021-07-20
Filing date: 2022-07-13
Publication date: 2023-01-26
Also published as: FR3125616B1; EP4374287A1; FR3125616A1

Abstract

The invention relates to a medium (1) comprising information able to be recognized optically by a user and storing digital data and means for decoding the digital data, the optically recognizable information comprising a succession of: a) a text explaining a process to be implemented in order to decode the digital data; b) an algorithm (110) to be transcribed so as to initialize a working memory of a virtual machine and execute this virtual machine; d) a list of alphanumeric characters (118, 119, 120) to be entered coding padding data for the working memory and a program for decoding a first matrix; e) first and second matrices to be digitized (122, 123) including black and white binary symbols at a predetermined location of the medium, including digital data of a program for decoding the second matrix (123) and the stored digital data to be decoded, respectively.

Description

Title of the invention: Support optically discernible by a user, representing digital data and the means of decoding them [0001] [The invention relates to the archiving of long-term digital data, and in particular the archiving of digital data over periods when the standards of materials or coding may vary in large proportions.

[0002] For certain security devices or processes, for example in the nuclear field, the time required to monitor devices or materials may exceed one or more centuries. To ensure the safety of future responders, it is important to allow them to maintain access to digital data relating to safety.

[0003] Furthermore, for cultural applications, it is important to guarantee access to a digital cultural heritage for future generations.

[0004] Most of the media accessible by optical, electronic or magnetic reading are faced with insufficient lifetimes or too rapid obsolescence for very long-term security applications or to guarantee the preservation and consultation of cultural heritage. The invention aims to solve one or more of these drawbacks. The invention thus relates to a medium comprising information optically discernible by a user, this information storing digital data and means for decoding the digital data, as defined in appended claim 1. [0006] The invention also relates to the variants of the dependent claims. Those skilled in the art will understand that each of the characteristics of the dependent claims or of the description can be combined independently with the above characteristics, without however constituting an intermediate generalization. Other characteristics and advantages of the invention will emerge clearly from the description which is made of it below, by way of indication and in no way limiting, with reference to the appended drawings, in which:

[0008] [Fig.1] is a schematic representation of a support according to an example of implementation of the invention;

[0009] [Fig.2] is a schematic representation of an introductory page of the support of Figure 1;

[0010] [Fig.3],

[0011] [Fig.4],

[0012] [Fig.5],

[0013] [Fig.6],

[0014] [Fig.7],

[0015] [Fig.8],

[0016] [Fig.9],

[0017] [Fig.10],

[0018] [Fig.11],

[0019] [Fig.12],

[0020] [Fig.13],

[0021] [Fig.14] and

[0022] [Fig.15] are schematic representations of different pages of the support of Figure 1;

[0023] [Fig.16] is a magnified representation of an example of a stored digital data matrix.

[0024] The inventor assumed that paper supports, or any other support comprising optically discernible information by a user for their reading, could have a very long life in good storage conditions. This lifetime can be much higher to that of a medium with optical reading by laser or magnetic. Consequently, such a type of support was chosen with a view to implementing the invention.

Figure 1 is a schematic representation of a support 1 according to an example implementation of the invention. The medium 1 comprises information optically discernible by a user. This information stores digital data and means for decoding the digital data.

In the examples shown, the support 1 comprises sheets of paper 10 to 1 n, ordered in succession, for example by means of page numbers Each figure 2 to 15 corresponds to a respective page of the support 1 of this example. Although the example has been described with reference to a support 1 comprising sheets of paper, other supports comprising information optically discernible by a user can be envisaged according to the invention, for example microfilm, microfiche, cinema film, plate or disc (in glass, sapphire or ceramic for example). The sheets of paper 10 to 1n advantageously each comprise an index discernible to the eye by a user, making it possible to order them.

In Figure 2, the medium has an introductory page, including an explanation 100 of the structure of the medium and the information returned. The introductory page can include an index separated from the rest of the pages, for example an alphabetic character or a value 0, in order to be able to process this page independently of the rest of the support 1. An example of such an explanation 100 is provided in the following table as an example:

[0028] [Tables 1]

0029] Support 1 includes a preliminary part including both explanations and an algorithm for restoring a digital sequence including a virtual machine. Such a preliminary part (FIGS. 2 to 12) can also be designated by the term primer. Support 1 includes an explanatory text of a method to be implemented to decode digital data present on another part of support 1 (FIGS. 13 to 15). The description will first give an example of explanatory text of the method to be implemented to decode the digital data.

The main characteristic of the preliminary part or primer is to provide a succession of technical means making it possible to implement and progressively validate, in stages, a program for decoding the digital data stored on the other part of the support 1.

In Figure 3, the medium includes on page 1 (index corresponding to the beginning of the information and instructions to implement the decoding) part of the explanatory text. Page 1 thus advantageously includes a summary 101 of the decoding method, an instruction 102 for digitizing black and white matrices including binary symbols at a determined location on the medium, the definition 103 of a program for converting the images to be produced, an instruction 104 for generating a digital sequence from the application of the realized conversion program applied to the digitized images, and an illustration 105 of an example of digital sequence obtained from the conversion program.

An example of a summary of the decoding method 101 is reproduced below. This synthesis 101 summarizes the different steps to be implemented to decode optically discernible information. An example of such a summary 101 is provided in the following table by way of example:

[0033] [Tables 2]

Ό034] An instruction 102 for digitizing black and white matrices of binary symbols at a determined location on the medium is provided. Instruction 102 may include recommendations on digitizing and/or processing the digitized image of a matrix of binary symbols. An example of such a 102 instruction is provided in the following table as an example:

[0035] [Tables 3]

_ A definition or parameterization 103 of a program to be carried out to convert the digitized images into a digital sequence is provided. This definition may include the ordering of words or bits in the digitized image. These words can define on the one hand the geometry of the digitized image, and on the other hand a linear representation of the luminance of each point of the digitized image. These instructions, such as those contained in the table below and illustrated in a graphic illustration 105, are then intended to make it possible to generate a numerical sequence from the program produced.

[0037] [Tables 4]

Ό038] An instruction 104 for generating a digital sequence from the application of the conversion program produced, applied to the digitized matrices, is provided on the support. An example of the definition of the numerical sequence is provided with reference to the following table:

[0039] [Tables 5]

_

0040] In Figure 3, the support includes on page 2, an instruction 106 for adapting an algorithm in a programming language. Instruction 106 explains also the beginning of the course of the execution of the algorithm. An example instruction is given in the following table:

[0041] [Tables 6]

_

0042] In Figure 3, the support advantageously includes on page 2, an explanation 107 of the results returned to the display during the execution of the program corresponding to the algorithm. An example of explanation 107 is given in the following table:

[0043] [Tables 7]

0044] Advantageously, the medium includes an explanation 108 of error messages potentially returned by the program during its execution. An example of explanation 108 is given in the following table:

[0045] [Tables 8]

0046] In Figure 4, the support 1 advantageously includes on page 3 an explanatory

109 notations used in an algorithm detailed in the rest of support 1. An example of explanation 109 is given in the following table:

[0047] [Tables 9]

0048] Support 1 includes (here on page 3 and following) an algorithm to be adapted in a programming language. The algorithm is based on a reduced number of operations of a virtual machine. Page 3 has here a first part

110 of the algorithm, a second part 111 of the algorithm is illustrated on page

4. A third part 112 of the algorithm is illustrated on page 5. An example of an algorithm to be adapted is illustrated in the following table:

[0049] [Tables 10]

strong, pointer, translation, usual, vector and area

(The 15 variables above can be initialized to 0 or any other value)

► Allocate the initially empty designation sequence (its final size will remain less than 256 bytes)

► Allocate the initially empty output sequence (its final size has no predefined limit)

► Allocate a set of sixteents in the form of a memory array named content ranging from indices 0 to maximum; its size (maximum + 1) must be at least equal to (3702 784 + t) seizets, with t the size of an image matrix (maximum size if they are not all of the same dimensions), i.e. i.e. its width multiplied by its height

► content[0..10] <- 25, 73, 16521, 122, 16384, 20, 16522, 138, 137, 16409, 26

► content[11..21] <- 16524, 140, 16523, 37, 16387, 65521, 16, 16387, 49, 138, 16524

► content[22..32] <- 140, 28, 16387, 64497, 64242, 64527, 65413, 31, 27, 16387, 29

► content[33..43] <- 25, 16409, 16386, 40, 65504, 49025, 25, 73, 16409, 25, 65484

► content[44..54] <- 121, 16525, 43, 127, 16526, 18, 16524, 140, 32799, 73, 16524

► content[55..65] <- 140, 140, 140, 73, 65464, 32861, 142, 73, 16526, 142, 142

► content[66..76] <- 16526, 142, 59, 16387, 5, 16387, 133, 0, 140, 73, 65447

► content[77..87] <— 40458, 141, 73, 16525, 141, 141, 16525, 141, 76, 16387, 36

► content[88..98] <- 16387, 134, 139, 16523, 16387, 44444, 33333, 33333, 33333,

33333, 33333

► content[99..109] <- 33333, 33333, 33333, 33333, 33333, 33333, 33333, 33333, 33333, 33333, 33333

► content[110..120] <- 138, 73, 16522, 138, 139, 135, 17, 16523, 139, 136, 16387

► content[121..131] <— 4, 38, 3, 16387, 60, 16384, 52, 137, 73, 16409, 25

► content[132..136] <- 16387, 56, 65530, 65427, 65517

(Sixteens of indices 137 to maximum of the memory array can be initialized to 0 or any other value)

► index <— 0

► number <— 0

► carry <— 0

► order <— 5

► While order < 20 do (i.e.: If order < 20 then) If order > 15 then 0 order <— order - 14

0 If number > argument then

• number <— (number - argument) - carry

• carry <— 0

0 Else if number = argument then

• number <— carry * 65535 0 Else

• number <— ((number + 1) - carry) + (65535 - argument)

• carry <— 1 Else if order < 5 then

0 argument <— content[subscript]

0 If index = 65535 then

• index 0 0 Else

• index index + 1

0 If argument = 0 then

• order order + 8

0 Else if argument < 16384 then

• content argument[argument]

• order order + 14

0 Else if argument = 16384 then

_

Ό050] To synthesize, this algorithm includes:

[0051] - the validation of 137 values of the initial code of a virtual machine by the execution of a function number 5 (11 lines) calling on a subtraction function with borrowing (9 lines). This step takes place after initialization of the algorithm and in its general framework: “while” loop and referral by testing the “order” variable. A message is displayed in the event of an error at this stage. To verify its success, it is specified in the written part of the preliminary part or primer that "The first assignment of the order variable to the value 8 indicates the correct entry of the values of the memory table at the start of the algorithm";

[0052] the display of the value “1000”: following the previous step, it is indicated that “Therefore, the value 1000 must be displayed after 10 iterations of the “While” loop. » These 10 iterations call upon most of the algorithm (in particular the execution engine of a virtual machine), with the exception of the input functions and most output functions (only the value 1000 is output). The low number of iterations allows the programmer taking over the algorithm to consider a detailed study of the operation of his program (so-called “debug” mode) if it does not work as expected. It is even possible to study it “by hand”, on paper.

[0053] - the entry of a line of letters: for the next step, it is specified that: “the programmer can check that the entry of any letters results in the display of the message “Error on the line! Please enter it again. (symbol 32800), followed by the redisplay of the line number. Conversely, entering the correct line letters should directly display the next line number without an error message. This step involves the character input function, the output of an error symbol, as well as almost all of the initial code of the virtual machine. As it is a single line, the input can be done directly by the user, very quickly.

[0054] - the entry of all the letters: this step does not present any algorithmic difficulty if the previous steps work. However, the user will have to find the most efficient way (for example character recognition) to complete the input. Automatic verification of each line allows the user to quickly identify and correct any input errors.

[0055] - the display of the first success (or error) symbol: the output of symbol No. 4801 validates the principle of operation of the input of scanned image data (2 values entered at this stage);

[0056] the display of the second success (or error) symbol: the output of symbol no. 4802 reinforces the validation of the principle of the operation of the input of scanned image data (data of complete entries at this stage);

[0057] -the display of a fourth success (or error) symbol: the output of symbol No. 4804 completes the validation of the scanned image data input function and the correct formatting of an image in the numerical sequence. The output of symbol 4804 also begins to validate that a matrix has been correctly scanned (sufficient area, definition, contrast); [0058] the display of symbol No. 4806 confirms that a matrix has been correctly scanned;

[0059] the new display of symbol No. 4801 validates the correct sequence of images in the digital sequence; [0060] - the validation of the output of data from a digital sequence in a file, which will occur later, after the decoding of a stored data matrix, in the case of a small test file stored in a single matrix stored digital data to be decoded. The compaction of all the elements of the solution makes it possible to reach this stage in a reasonable time, the execution accelerating very significantly after the decoding of the first system matrix detailed below.

[0061] -the addition of an optional additional algorithm, the correct restitution of the test file in a much shorter time (acceleration factor greater than 50) allowing the correct implementation to be validated. In the event of significant difficulty (never observed during several tests carried out with technicians used as reference users), it is of course always possible to keep only the main algorithm which precedes in autonomy.

In the above algorithm, the lines relating to the allocation of values until the initialization of the index, number, carry and order parameters are intended to initialize the table of the working memory of a virtual machine. These lines have the function of launching calls to subsequent functions of the program produced according to the algorithm.

[0063] The part of the algorithm executed if the condition order=5 is verified corresponds to a function for checking the values assigned to the table of the working memory of a virtual machine, in order to be able to manage the display of a possible error message when entering these values.

The part of the algorithm executed if the condition order<5 (values 3 and 4) is verified corresponds to the interpretation engine of the virtual machine, by directing the execution of the instructions of the virtual machine. The existence of the two values makes it possible to distinguish, in other functions of the algorithm, the two modes of execution “initial entry of letters by the user” and “current execution in autonomy”. The part of the algorithm executed if the condition order=6, if order=7 or if order<10 is verified corresponds to the management of errors in the execution of the virtual machine.

The values 10 and 12 allow entry of scanned image data; the existence of the two values makes it possible to distinguish the two modes of execution "current execution in autonomy" and "current execution with an optional additional algorithm accelerator": indeed, the instruction "order <— (order - 10) * 9” gives, for the value 10, (10 - 10) * 9 = 0 (return to the fast data entry algorithm of an acceleration virtual machine), and for the value 12,

(12 - 10) * 9 = 18 (subtraction then return to the value 4: virtual machine instruction interpretation engine).

The part of the algorithm executed if the condition order=11 is verified corresponds to the management of an entry of letters or alphanumeric characters by the user.

The final part of the algorithm corresponding to the other conditions on the order values (values 13, 14 and 15) corresponds to the management of the outputs of the virtual machine. This part includes in particular a stop function, the management of an option for an acceleration function, the management of the display of an input line number, the management of a change of operating mode, as well as that the display of an explanation 107 of the results returned to the display or of an explanation 108 of error messages, if applicable; the existence of the three values makes it possible to distinguish the three modes of execution.

The values 16 to 19: subtraction operation with borrowing; the existence of the four values makes it possible to distinguish the three modes of execution as well as the function of verification of the initial code of the virtual machine (values 2 to 5 respectively), which resort to this operation.

[0070] The value 20 terminates the execution of the algorithm.

The initial code, installed with the prior filling of the working memory of the virtual machine with 137 16-bit values:

-displays a line number; - controls the entry of letters, each letter being converted by function number 11 of the main algorithm into a value from 0 to 15 (4 bits);

- concatenates four 4-bit values to form a 16-bit value which is stored at the correct address in the working memory of a virtual machine (i.e. in the "content" table of the algorithm);

- calculates a checksum per line (16 16-bit values), each value having a different weight in the sum in order to detect any permutations, and again integrates the 17th and last value provided to cancel the checksum; - displays an error message and loops to the first operation if the complete checksum is different from 0;

- goes to the next line number and loops to the first operation otherwise;

- after a few lines entered, transfers the execution to the code thus loaded in memory, namely the extension of the initial code of the virtual machine.

The support 1 advantageously includes on page 5 an explanation 113 of a notation used in an optional supplement to the algorithm 110. This explanation 113 also details the usefulness of the optional supplement to the algorithm as well as the principle of the Boolean and logical operation, supported by an example. An example of explanation 113 is given in the following table:

[0073] [Tables 11]

_

Ό074] Support 1 includes (here on page 5 and following) an optional additional algorithm to be adapted in a programming language. The optional algorithm supplement can in particular add one or more arithmetic or logical operators to accelerate the calculations carried out by a program corresponding to the algorithm. Page 5 here comprises a first part 114 of the optional algorithm supplement, a second part 115 of the optional algorithm supplement is illustrated on page 6. A third part 116 of the optional algorithm supplement is illustrated on page 7. An example of algorithm to adapt is illustrated in the following table [0075] [Tables 12]

Ό076] The 'order' variable still makes it possible to direct the execution of the optional additional algorithm to the correct function:

- values 0 to 2: reserved for the optional additional algorithm; ; value 0: fast data entry (function ensured by the software interpreted with the main algorithm in autonomy); value 1: (re)initialization of the optional algorithm supplement virtual processor; value 2: engine for interpreting the instructions of the optional algorithm supplement virtual processor.

[0077] The extension of the initial code loaded by the latter:

[0078] - load a large block of data (until the end of a set of characters to be entered from medium 1, detailed later) using the initial code of the virtual machine;

[0079] - decompresses the initial code of a second-level virtual machine (that of the optional algorithm supplement) compressed according to a method specific to this code;

-transmits the value code 2000 to the main algorithm, thus activating the execution engine of the optional additional algorithm, if this is implemented, instead of the execution engine of the virtual machine, this which has the effect of immediately abandoning the execution of the initial code of the virtual machine;

[0081] -if the optional additional algorithm is not implemented, the execution of the initial code of the virtual machine continues with the loading of the remaining letters (last part of the characters to be entered) by means of the initial code of the virtual machine ;

[0082] - decompresses the loaded values, which contain the interpreter of a second-level virtual machine (also implemented in the optional algorithm supplement) compressed according to a method specific to this code; it is a simplified interpreter containing only the functionalities strictly necessary for the execution of the code at this stage;

[0083] - transfers the execution to this simplified interpreter.

The initial code of a second-level virtual machine decompresses the continuation of a large block of data, compressed according to a third-party generic algorithm. This decompression code is used for the rest of the process, a version adapted to large data sizes (> 64 KB) being used later to decompress the user file if necessary.

The decompressed data contains the decoding program for the two system matrices 121 and 122 detailed below written in a specific language with its simplified interpreter containing only the functionalities strictly necessary for this decoding program.

As illustrated in Figure 9, the support 1 advantageously includes on page 7 instructions 117 for entering letters or alphanumeric characters. The program produced according to the previous algorithm can contain a verification of the entries by means of checksums, in order to verify that the characters entered are satisfactory. An example of instructions 117 is given in the following table:

[0087] [Tables 13]

0088] The medium 1 then includes a list of letters or alphanumeric characters to enter. The input of these characters by the user is used:

- on the one hand to fill the table of the working memory of the virtual machine, considered as a first level virtual machine. This part also includes a decompressor of the compressed code of a second level virtual machine. This part also includes a program for decoding at least a first predefined matrix of binary symbols.

The corresponding typed characters are those up to line 3560 in the following example; -on the other hand to manage the absence of implementation of the optional additional algorithm.

The support 1 includes (here in Figures 10 to 12) and a list of letters or alphanumeric characters to enter. A first part 118 of this list is provided on page 8 of support 1, a second part 119 of this list is provided on page 9 of support 1, and a third part 120 of this list is provided on page 10 of support 1.

[0090] An example of a list of characters to enter is illustrated in the following table

[0091] [Tables 14]

Ό092] Support 1 also includes a matrix 121, an example of which is shown in Figure 13 (corresponding to page 11 of support 1), a matrix 122 an example of which is shown in Figure 14 (corresponding to page 12 of support 1) and a matrix 123, an example of which is shown in Figure 15 (corresponding to page 13 of support 1). Arrays 121-123 include black and white binary symbols and can be thought of as two-dimensional barcodes. In the present case, the black and white binary symbols have different coding configurations in the matrices 121 to 123.

The matrices 121 and 122 are system matrices. Matrix 123 (and potentially other subsequent matrices) contain the stored digital data to be decoded.

The system matrices 121 and 122 contain, in a compressed way:

- a complete second-level virtual machine interpreter (implemented in the optional algorithm supplement) for the first-level virtual machine;

-an almost complete dynamic compiler for a third-level virtual machine, making it possible to multiply by more than 15 the execution speed from the decoding of the second system matrix 122;

- the complete program for decoding the 123 data matrices, including the reconstruction of the user file by having recourse, if necessary, to two types of Reed-Solomon error correction (very widespread method, used for example by QR codes): intra-array correction (15% redundancy to correct up to 7.5% errors in the same array) and inter-array correction (3 redundancy arrays per group of 17 arrays to correct up to three blocks of data remaining erroneous despite the intra-matrix correction).

Furthermore, instead of a user file, provision is made for matrix 123 (and possibly subsequent ones) to contain additional code which will make it possible to load a complete version of a specific dynamic compiler into the system, as well as than any useful operating system or software.

The system matrices 121 and 122 contain error correction information according to the Hamming(8,4) method, which is both simple and proven.

The matrix 123 of stored digital data to be decoded (illustrated in FIG. 15) here comprises: a fairly thick black continuous frame 124 encircling this matrix 123;

- 125 white space on all four sides, equal in thickness to the frame

124, delimited by a second black line 126. The thickness of the black line 126 is not an important parameter. Estate black frame 124, white frame

125, black frame 126 makes it easy to detect the presence of a frame in the image. The continuity of the framing around the matrix 123 makes it possible to discriminate it from any other elements present in the image (dust for example), and to detect the position of the four corners of the matrix 123. The same framing system can also be used for dies 121 and 122; -three entirely blackened corners 127, and a partially blackened corner 128, which makes it possible to detect the orientation of the matrix 123 in the image (including in mirror) and to discriminate it with respect to a system matrix 121 or 122 of which the blackening of the corners may be different;

-a blackening 129 of one (or more) large dots in the white space following the blackened corners 127, in order to possibly encode the version of the matrix 123;

-a white space 130 in the internal frame of one of the long sides of the matrix 123 (that of the corner 128, which indicates the starting point of the optical reading). This space 130 is used as an optical guide to run through each line of the emblem, adapting to any geometric deformations;

-a set of crenellated lines 131 carrying binary information, each line consisting of a long line 132 punctuated at regular intervals by a small perpendicular line 133, the long line 132 possibly continuing at one end or at the other of each dash 133 to denote a 0 or 1 bit value. The lines 131 are alternately shifted by half an interval so that two dashes 133 from one line 131 to the other aligned, to avoid the possible joining of the two small lines 133 which could be induced by the overflow of the ink when too many black pixels are present in an area;

Each crenellated line 131 (a region of crenellated lines 131 is illustrated in FIG. 16) optically represents a binary digital signal. The small perpendicular lines 133 constitute the clock of the signal, at the rate of one clock mark per bit. This regularity of the clock in the signal makes it possible to tolerate significant optical distortions, at different scales. A bit is detected by the narrowing of the width of the global line (transition from the small line 133 to the long line 132 on the signal). To reinforce the width contrast and the position of the long line, the end of the small line 133 opposite the long line 132 (the position of which marks the value of the bit) is lengthened by one pixel. In addition, the thickness of the other end of the small line which adjoins the long line is reduced by one pixel to ensure a more abrupt drop in the signal towards the long line, i.e. to compensate for the bulge induced by the assembly of too many black pixels printed.

It will also be possible to consider that binary symbols of the matrices 121 to 123 with a very high brightness contrast correspond to black and white symbols. The distinction between black and white is made by the image processing software process integrated into the pre-part or primer.

[0100] According to an independent aspect of the invention, the medium 1 may be devoid of the system matrices 1 and 2, the execution of the program produced directly decoding the matrix 123 of the stored digital data to be decoded.

Claims

[Claim 1] [Support (1) comprising information optically discernible by a user, this information storing digital data and means for decoding the digital data, characterized in that the optically discernible information comprises a succession of: a) a text explanation of a method to be implemented to decode the digital data, the explanatory text including: i) an instruction (102) for digitizing black and white matrices of binary symbols at a determined location on the medium; ii) an instruction (103) for setting up a program to be carried out to convert the digitized images into a digital sequence; iii) an instruction (104) for generating a digital sequence from the application of the program produced and applied to the digitized matrices; b) an algorithm (110) to be transcribed into a programming language, so as to initialize a working memory of a virtual machine and so as to execute this virtual machine; d) a list of alphanumeric characters (118, 119, 120) to be entered encoding data for filling the working memory of the virtual machine and a program for decoding a first matrix of binary symbols; e) first and second matrices to be digitized (122, 123) including binary symbols in black and white, arranged at a predetermined location on the medium, the first matrix (122) including digital data from a decoding program of the second matrix (123), the second matrix (123) containing the stored digital data to be decoded.

[Claim 2] Support (1) comprising information optically discernible by a user, according to claim 1, in which the symbols of the first matrix (122) are black or white dots.

[Claim 3] Medium (1) comprising information optically discernible by a user, in which the medium (1) is in the form of a succession of pages discernible by a user and ordered by indices discernible by the user on each page of the support (1).

[Claim 4] A medium (1) carrying information optically discernable by a user, according to claim 3, wherein the medium is selected from the group consisting of sheets of paper, microfilm, microfiche, motion picture film, and glass, sapphire or ceramic plates.

[Claim 5] Support (1) comprising information optically discernible by a user according to any one of the preceding claims, in which the explanatory text (101) includes beforehand a summary of the various steps to be implemented to decode said matrices.

[Claim 6] A medium (1) carrying user optically discernible information according to any preceding claim, wherein the first optically discernible information on the medium (1) includes an explanatory (100) of the structure of the information present on the medium and an explanation of the digital data stored.

[Claim 7] Medium (1) carrying information optically discernable by a user according to any one of the preceding claims, in which the list of alphanumeric characters (118, 119, 120) to be entered is preceded by c) an entry instruction (117) of alphanumeric characters.

[Claim 8] A medium (1) carrying information optically discernable by a user according to any preceding claim, wherein the first or the second matrix to be digitized comprises alternating a first black frame (124), d a white frame (125) and a second black frame (126).

[Claim 9] A medium (1) carrying information optically discernable by a user according to any preceding claim, wherein the first or second matrix to be digitized comprises jagged lines (131) each including a long line (132) punctuated at regular intervals with small lines perpendicular to the long line [Claim 10] A medium (1) carrying information optically discernable by a user according to any preceding claim, wherein the instruction (104) for generating a digital sequence from the application of the program performed comprises an instruction for organizing the numerical sequence into machine words.