JPH0855180A - Method for deciphering and recording digital information - Google Patents

Abstract

PURPOSE:To simplify an identification processing and reduce data throughput by setting rectangular windows which encircle a block in order and identifying marks given to boxes in the block. CONSTITUTION:Image data in an image RAM are retrieved to measure the start position and inclination of the data. After the inclination at the time of scanning is corrected, a window is generated on the basis of the size of the boxes and the number of the boxes constituting the block (s91). On the basis of the start position of the data, the window is set at a position including the block (s92). Then the border of the block included in the window is detected (s93). The detected border position of the block is stored in the RAM (s94). Further, the block whose border is detected is divided to find a position for sampling the boxes (s95). The image bit at the position is sampled to obtain the values of the boxes (s96).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital information decoding method for decoding digital information from a record carrier on which digital information is recorded by printing or the like. The present invention also relates to a digital information recording method for recording digital information on a recording surface of a record carrier by printing or the like.

[0002]

[Prior Art and Problems to be Solved by the Invention]
・ The digital information represented by the binary value of "1" is encoded,
The technique of recording on a record carrier as a two-dimensional pattern is widely used. For example, a matrix code is a means for encoding data by using two-dimensionally arrayed cells and providing optically recognizable marks on each cell, typically bright (white) and dark (black). is there.

Conventionally, when decoding digital information from such a record carrier, a two-dimensional pattern represented on a recording surface is read as an image, and the entire read image data is read.
It was dimensionally searched and compounded.

Therefore, it is necessary to access the entire read image data at one time, and when the information density and density are increased, there is a problem that the data processing amount becomes large and the processing takes time. It was Further, when the recording surface and the image sensor are misaligned when the two-dimensional pattern is read and the image data is locally distorted, it is difficult to correct the misalignment. There is known a method (Japanese Patent Laid-Open No. 12579/1990) in which clocking information is added around the pattern on the recording surface in order to prevent reading misalignment. The distortion cannot be corrected.

Therefore, an object of the present invention is a digital information decoding method for decoding the above-mentioned digital information from a record carrier on which digital information is recorded by a matrix code, and when the information has a high density and a large capacity. Even so, it is an object of the present invention to provide a digital information decoding method capable of decoding in a short time with a relatively small amount of data processing, and capable of easily correcting local distortion of data.

Further, the present applicant has previously proposed a digital information recording method capable of increasing the recording information density and allowing control information such as clocking information to be included inside the data area (Japanese Patent Application No. Hei 10 (1999) -135242). 6-70911). One of them is a pattern that a square block composed of a plurality of cells is virtually set on a plane, and an optically recognizable mark is given to each of the cells, so that the block can have a pattern. In the group, the number of consecutive marks having the same value is k in at least one of the row direction and the column direction (where k is 1 or more, and the mesh forming the side of the block in the one direction). An integer less than the number of) is adopted as the designated pattern. And
Each designated value of the block unit is made to correspond to each data value forming the digital information to be recorded, and the digital information to be recorded is expressed as a two-dimensional pattern by arranging the designated pattern of the block unit on the recording surface. To do.

An object of the present invention is to provide a digital information recording method which is an improvement of the previously proposed recording method. In particular, the present invention provides a digital information recording method capable of improving block identification accuracy during reading.

A further object of the present invention is not to represent the digital information to be recorded by the specified pattern, but to encode it by random number processing in units corresponding to the blocks so that light and dark are substantially evenly distributed. Another object of the present invention is to provide a digital information recording method capable of improving the identification accuracy of a block at the time of reading when represented by the above pattern.

[0009]

In order to achieve the above object, a digital information decoding method according to a first aspect of the present invention is such that matrix-shaped cells corresponding to bits are virtually set on a flat recording surface. A digital information decoding method for providing optically recognizable marks on each of the cells and decoding the digital information from a record carrier on which digital information is recorded as a two-dimensional pattern of the marks. , A two-dimensional pattern representing the digital information is read as image data from the recording surface of the record carrier, the image data is virtually divided into rectangular block units composed of a plurality of cells, and each block of the image data is divided. Then, the rectangular windows surrounding this block are set sequentially to identify the marks given to the cells in the block, and the information represented by the marks is decoded. It is characterized by obtaining the original digital information.

The digital information decoding method according to claim 2 is the digital information decoding method according to claim 1, wherein when the window is set in the first block of the image data, the outside of the first block is set. , A window having a size larger than the size of this block is set, and then the side of the window is moved inward in image bit units representing the above image data so that the side of the window crosses the image bit. The change in the value is detected, and the side of the window is stopped at the boundary position where the value of the image bit changes.

According to a third aspect of the present invention, there is provided a digital information decoding method in which a rectangular block composed of a plurality of cells is virtually set on a plane and each of the cells is optically recognizable. In the pattern group that can be taken by the above block, the number of consecutive marks having the same value is k in at least one of the row direction and the column direction (where k is 1 or more and It is an integer less than the number of cells forming one side of the block.) The following plural patterns are adopted as the designated patterns, and each data value forming the digital information to be recorded has the above block unit. From the record carrier in which the specified digital patterns to be recorded are expressed as a two-dimensional pattern by arranging the specified patterns in block units on the recording surface. A digital information decoding method for decoding digital information, wherein a pattern that is not adopted as the designated pattern among the pattern groups that the block can take by adding a mark to each of the cells is regarded as an invalid pattern. A preset two-dimensional pattern representing the digital information is read as image data from the recording surface of the record carrier, the image data is virtually divided into the block units, and each block of the image data is divided into blocks. By sequentially setting a rectangular window surrounding each of the blocks, the marks provided in the cells in each block are identified to determine whether the pattern is a designated pattern or an invalid pattern. When the pattern of the block is the designated pattern, the mark While the information represented by is decoded and the original digital information is obtained, the pattern of the above block is invalid. When it is over emissions, it is characterized by performing the process of correcting read error.

According to a fourth aspect of the present invention, there is provided a digital information recording method, in which a rectangular block composed of a plurality of cells is virtually set on a plane and each of the cells is optically recognizable. In the pattern group that can be taken by the above block, the number of consecutive marks having the same value is k in at least one of the row direction and the column direction (where k is 1 or more and It is an integer less than the number of cells forming one side of the block.) The following plural patterns are adopted as the designated patterns, and each data value forming the digital information to be recorded has the above block unit. A method for recording digital information in which the specified information of each block is associated and the digital information to be recorded is expressed as a two-dimensional pattern by arranging the specified patterns in block units on the recording surface. In the case of adopting a specific pattern other than the specified pattern among the pattern groups that the block can take as a boundary identification pattern for identifying the boundary of the block, and arranging the specified pattern of the block unit on the recording surface. , With the blocks already placed,
Whether or not the boundary with the block to be arranged is easy to identify is determined based on the pattern of the cells that sandwich the boundary, and when the boundary is easy to identify, the block is arranged as it is, while the boundary is When it is not easy to identify, a block having the boundary designating pattern is arranged instead of the block.

According to a fifth aspect of the present invention, in a digital information recording method, a rectangular block composed of a plurality of cells is virtually set on a plane, and digital information to be recorded is recorded in units corresponding to the block. Coded by random number processing to create a coding pattern in which optically recognizable marks are provided to each of the cells forming the block in a substantially uniform distribution, and the digital information to be recorded is recorded in the block unit. In the digital information recording method for expressing the two coding patterns as a two-dimensional pattern by arranging the coding patterns on the recording surface, a pattern in which two kinds of marks representing different values are alternately provided from the pattern group that the block can take. , Is adopted as a boundary identification pattern for identifying the boundary of a block, and the above-mentioned block-unit coding pattern is recorded on the recording surface. At the time of sliding, it is judged whether or not the boundary between the block already arranged and the block to be arranged is easy to identify based on the pattern of the meshes that sandwich the boundary, and when the boundary is easy to identify, While arranging the blocks as they are, when the boundary is not easy to identify, a block having the boundary designating pattern is arranged instead of the block.

Further, in the digital information recording method according to the sixth aspect, in the digital information recording method according to the fifth aspect, when the same coding pattern as the boundary identification pattern is generated by the random number processing, When arranging block-by-block coding patterns on the recording surface, a plurality of blocks having the same coding pattern as the boundary identification pattern are continuously arranged, and a coding pattern corresponding to the digital information to be recorded is provided. , Which is characterized by being distinguished from the original boundary identification pattern.

A digital information recording method according to a seventh aspect is the digital information recording method according to any one of the fourth to sixth aspects, in which the already arranged block and the block to be arranged are arranged. It is characterized in that whether or not a boundary is easily discriminated is determined based on how many pairs of cells having different values out of the pair of cells adjacent to each other across the boundary.

[0016]

According to the digital information decoding method of the present invention, the image data read from the record carrier is virtually divided into rectangular block units consisting of a plurality of cells, and each block of the image data is divided into blocks. Since the surrounding rectangular windows are sequentially set to identify the marks given to the cells in the block, it is sufficient to identify the bright and dark patterns in block units. In comparison, the identification process is simplified. Also,
Since you don't need to access all the image data,
The data processing amount can be reduced and the processing time can be shortened. Further, it becomes possible to correct local distortion of the image data, and the data recognition rate is improved.

In the digital information decoding method of the second aspect, when the window is set in the first block of the image data, a window having a size larger than the size of the block is set outside the first block. Then, the edge of the window is moved inward in image bit units representing the image data to detect a change in the value of the image bit that the edge of the window crosses, and the boundary at which the value of the image bit changes. Stop the above side of the above window in position. Therefore, the boundaries of the blocks are detected with high accuracy. The boundaries of the second and subsequent blocks are detected using the boundary position of the first block.

In the digital information decoding method according to a third aspect of the present invention, among the pattern groups that the block can take by adding marks to each of the cells, a pattern that is not adopted as the designated pattern is preset as an invalid pattern, For each block of the read image data, a rectangular window surrounding this block is sequentially set, and the marks provided in the cells in each block are identified to determine whether it is a designated pattern or an invalid pattern. Then, when the pattern of the block is a designated pattern, the information represented by the mark is decoded to obtain the original digital information, while when the pattern of the block is an invalid pattern, a process of correcting a read error is performed. In this case, the reading error can be corrected when the block is read, and the bright and dark patterns in the block are corrected, so that reliable reading can be performed.

According to a fourth aspect of the present invention, in the digital information recording method, a specific pattern other than the designated pattern is adopted as a boundary identification pattern for identifying the boundary of the block in the pattern group which can be taken by the block, When arranging the designated patterns on the recording surface, it is determined whether or not the boundary between the block already arranged and the block to be arranged is easily discriminated based on the pattern of the cells sandwiching the boundary. Then, when the boundary is easy to identify, the block is arranged as it is, while when the boundary is not easy to identify, a block having the boundary designating pattern is arranged in place of the block. In this case, it is possible to avoid an arrangement in which it is difficult to identify the boundary of the block at the time of reading, and the block identification accuracy is improved.

According to a fifth aspect of the present invention, there is provided a digital information recording method, wherein a pattern in which two kinds of marks representing different values are alternately provided in a pattern group that a block can take is a boundary identification for identifying a block boundary. When arranging the above-mentioned block-unit coding patterns on the recording surface, it is determined whether or not the boundary between the block already arranged and the block to be arranged is easy to distinguish. Make a decision based on the pattern. Then, when the boundary is easy to identify, the block is arranged as it is, while when the boundary is not easy to identify, a block having the boundary designating pattern is arranged in place of the block. In this case, it is possible to avoid an arrangement in which it is difficult to identify the boundary of the block at the time of reading, and the block identification accuracy is improved.

According to a sixth aspect of the present invention, in the case where the same coding pattern as the boundary identifying pattern is generated by the random number processing, when the block-by-block coding patterns are arranged on the recording surface, the boundary identifying pattern is used. A plurality of blocks having the same coding pattern as the pattern are continuously arranged. Therefore, by imposing a restriction that a plurality of original boundary identification patterns are not arranged consecutively, the encoding pattern corresponding to the digital information to be recorded and the original boundary identification pattern are distinguished from each other. .

According to the digital information recording method of the seventh aspect, it is judged whether or not the boundary between the block already arranged and the block to be arranged is easily discriminated by determining whether the boundary between the adjacent meshes is across the boundary. Among them, since it is performed based on how many pairs of cells having different values exist, it is objectively judged whether or not it is easy to identify. As a result, it is possible to reliably avoid an arrangement in which it is difficult to identify the boundaries of the blocks during reading, and the block identification accuracy is further improved.

[0023]

EXAMPLES The present invention will be described in detail below with reference to examples.

(First Embodiment) First, an optical information recording device, a record carrier and a reading device will be described with reference to FIGS. 1 to 5.

FIG. 1 shows the overall structure of the optical information recording apparatus 10. The optical information recording device 10 is a device for encoding source data to generate an image representing a matrix code and recording the image on a recording medium such as paper. In FIG. 1, the external device 14 is composed of an external recording device, a source data creation device via a communication interface, or the like. The printing device 15
It is composed of a thermal transfer printer, a laser printer, or a block making device for printing.

The CPU (central processing unit) 11 is a ROM
The operation of the entire apparatus 10 is controlled according to the program recorded in the (read only memory) 12.
Specifically, as shown in FIG. 2, the CPU 11 first stores the source data given from the external device 14 in the RAM (random access memory) 13 (S21).
Next, the source data stored in the RAM 13 is subjected to various kinds of processing, and the bits of the source data are arranged and coded so as to correspond one-to-one to the cells of the image to be recorded (S22). ). Then, an image representing a matrix code is generated from the encoded bit array (S23). Then, by controlling the appropriate printing device 15,
The image is recorded on the recording medium (S24).

FIG. 3 exemplifies an image 30 recorded on the paper (recording surface) 31 forming the optical information recording carrier by the optical information recording apparatus 10. This image 30 represents a matrix code, and is configured by arranging virtual cells two-dimensionally. Each grid represents encoded data. For example, each grid represents a bit (coded bit), a dark grid represents a bit "1", and a light grid represents a bit "0".

FIG. 4 shows the overall construction of the optical reading device. The optical reading device 40 is a device for reading and decoding the image 30 from the recording surface 31 and using it as source data. In FIG. 4, the image sensor 45 is of a line type using a CCD element, for example. The control circuit 46 can relatively move the recording surface 31 in a certain direction at a constant speed with respect to the image sensor 45 by a paper feeding motor. Control circuit 46
At the same time as controlling the paper feed motor.
Control. The image read by the image sensor 45 is binarized and sent to the CPU 41. External device 47
Is a recording device provided externally or a source data processing device via a communication interface.

The CPU 41 controls the operation of the entire device 40 according to the program recorded in the ROM 43. Specifically, as shown in FIG. 5, image data supplied from the image sensor 45 via the control circuit 46 is sequentially stored in the image RAM 44 (S51). After that, the CPU 41 reads the lightness and darkness of the cells from the image data stored in the image RAM 44 (S52), and decodes the read information (S53). Decryption is
Basically, it is realized by performing the encoding process performed by the optical information recording device 10 in the opposite direction. The result obtained by this decoding is stored in the RAM 42 as source data.
The stored source data is sent to an appropriate external device 47 and used (S54).

The image sensor 45 and the control circuit 46 may be machine vision capable of reading all the images on the record carrier at one time. Further, the grid reading process may be performed when a part of the entire image is stored in the image RAM, and in this case, the capacity of the image RAM can be reduced.

Next, with reference to FIGS. 6 to 11, a detailed description will be given of the grid reading process as an embodiment of the digital information decoding method of the present invention. This mesh reading process corresponds to step S52 in FIG. Note that the image on the record carrier does not need to be recorded in block units, unlike the second and third embodiments described later.

FIG. 6 shows image data 60 obtained by reading the image 30 shown in FIG. 3 by the optical information reading device 40 with the horizontal direction as the main scanning direction and the vertical direction as the sub-scanning direction. The image data 60 is composed of 36 rows × 24 columns = 864 cells according to the image 30 on the record carrier. The image data 60 is stored in the image RAM 44.

FIG. 7 shows a state in which the image data 60 is divided into rectangular blocks each having a grid of 6 rows × 8 columns. If a row formed by blocks arranged in the horizontal direction is called a block row and a column formed by blocks arranged in the vertical direction is called a block row, the image data 60 has 6 block rows × 3 block columns,
That is, it is divided into 18 blocks in total. In FIG.
For each block of the image data 60, alphabets a, b, ..., R are added in order in the block row direction (main scanning direction). This indicates that the reading is performed for each block in this alphabetical order and the bright and dark of the cells are identified within each block. Note that reading may be performed in block units in the block column direction (sub-scanning direction). In that case, it will be read in the alphabetical order shown in parentheses.

The bit (encoded bit) represented by one cell of the image data 60 is composed of a plurality of image bits on the image RAM 44. Image sensor 45
The line resolution in the main scanning direction is 8 dot / mm, the line resolution in the sub-scanning direction is 8 dot / mm, and it is assumed that the mesh is recorded on the record carrier in a square of 0.625 mm square, as shown in FIG. As you can see, one square is the image R
It is composed of 5 × 5 image bits on AM. Figure 11
Then, "1" or "0" is given to the image bit 431 of the image RAM 44. "1" represents a dark bit and "0" represents a bright bit.

The mesh reading process is performed according to the flow chart shown in FIG. It is assumed that the image data 44 has already been stored in the image RAM 44.

First, the image data in the image RAM 44 is searched to measure the start position and inclination of the data (S9).
1). As shown in FIG. 11, each image bit 431 on the image RAM 44 has a subscript 2 at (X, Y) of the bit position X in the main scanning direction and the bit position Y in the sub scanning direction.
Accessed as a dimensional array. A bit is searched in the positive direction of the Y-axis from each point on the X-axis where Y = 0, and the coordinates of the dark bit found first are stored. The upper side of the data area and the slope thereof are obtained by excluding the coordinate values whose Y values are far apart from each other and performing linear approximation. Similar processing is performed on the left side of the data area and its inclination. The start position of the data (here, (5,4)) is calculated by finding the intersection of these two straight lines.
To get When all the image data is stored, the data area (for example, a square) may be measured. Furthermore, after correcting the inclination during scanning (which has already been obtained), the size of the cells (determined by the size of the cells at the time of recording and the accuracy at the time of reading) and the cells constituting the block are corrected. A window is created based on the block size determined by the number (here, 6 × 8) (S91). The window is a virtual quadrangle that can be represented by the coordinate values of a square on the image RAM 44, and the initial value is larger than the block size by, for example, 1/2 cell size vertically and horizontally.

Next, a window is set at a position including the first block a based on the start position of the data (S9).
2). For example, as shown in FIG. 10, the window 110 is set near the block a of the image data 60. The cells used to identify the boundaries are marked with a circle and a circle. In FIG. 11, this window 110 is (2, 1),
The rectangles (47,1), (2,36), and (47,36) are set as vertices (indicated by broken lines in FIG. 11).

Next, the boundary of the block a included in the window 110 is detected by the following steps (i) to (iii) (S93).

(I) First, starting from each side of the window 110, for example, the upper side, the upper side is divided by the number of cells forming the block + 1, and sampling points are obtained. For example, in the window 110 shown by the dotted line in FIG. 11, since the number of cells forming a block is eight, the upper side is divided into nine so that eight sampling points 433, 4 are formed.
34, ..., 4310 are obtained.

(Ii) The image bit located at each sampling point and one unit inside (upper side here) than the upper side
The image bits located at each point of and are sampled. Compare these values to see if they are different. This process is repeated while shifting inward by one unit for each sampling point until the values differ. Then, when the values are different, it is determined that there is a boundary position.

However, the inward shifting is up to the size of one cell (5 units in FIG. 11), and when the values are not different (that is, when the bits of the same value are continuous), the boundary position regarding the sampling point is Estimate from other points. For example, in FIG. 11, even if the sampling values are compared up to the size of the first cell at the sampling point 434, the values are not different. Therefore, the boundary position regarding the sampling point 434 is estimated from the sampling points 433 and 435 on both sides. Boundary positions regarding the sampling points 436 and 438 are similarly estimated from the sampling points on both sides.

In FIG. 11, a sequence of points having (2, 4) and (47, 4) at both ends is the boundary of this block, and the upper side of the window 110 is set to this boundary. If no boundary can be detected, it can be estimated from the block size.

(Iii) The process of (i) and (ii) above is performed in the window 1
By repeating the other three sides of 10, each side of the window 110 can be made to coincide with the boundary of the block a.

In this way, the boundary of the first block a is detected. For example, in FIG. 11, the boundary 4311 of the block 102 can be determined.

In this block boundary detection method, each side of the window is divided using the number of cells forming the block, and sampling points are set for each cell. Therefore, the amount of processing can be reduced as compared with the case where all the image bits located on each side of the window are sampled. Therefore, the time required for reading can be shortened.

The block boundary detection method can be modified in various ways. For example, if all the image bits located on each side of the window are sampled, a more accurate boundary can be obtained and the local It becomes possible to accurately cope with the distortion.

Next, the detected boundary position of the block a is stored in the RAM 42 (S94). This is because it is used as an initial value when detecting the boundary of a block adjacent to the block a.

Next, the block a in which the boundary has been detected is divided into 6 rows × 8 columns to obtain the position for sampling the cells (S95).

The value of the grid is obtained by sampling the image bit at that position (S96). Then, the value of the obtained mesh is stored in the RAM 42. The value of the image bit at the adjacent point may be checked with the sampling position as the center, and the value of the grid may be obtained from the result.

Next, it is judged whether or not the sampling of the last block in the row direction has been completed (S97), and if it is not the last block in the row direction, a new window is set at a position including the block on the right. S98). Then, the processing of steps S93 to S96 is repeated to detect the boundary of the block and sample the cells. In this way, after the processing of the block a shown in FIG. 8 is completed, the blocks b and c belonging to the same block row are sampled.

Here, setting of a new window in the same block row is performed using the boundary position of the block currently being processed. For example, the boundary positions 101 and 102 of the upper side and the lower side of the block a shown in FIG. 10 can be commonly used for the same block row. The right side of the boundary block a of the block a matches the left side of the block b. Therefore, the window setting for the block b is performed by setting the boundary position 1 between the upper and lower sides of the block a
In addition to 01 and 102, the boundary position 1 on the right side of the block a
04 is also available.

When the sampling of the final block in the row direction is completed, it is judged whether or not the sampling of the final block in the column direction is completed (S99), and a new window is opened at the position including the block to be sampled next. It is set (S910). Then, steps S93 to S9.
The process of 6 is repeated, the boundary of the block is detected, and the cells are sampled.

Here, the blocks d, e, f shown in FIG.
Since the boundary of the lower side of each block of the previous block row is already known in the second and subsequent block rows to which is belongs, this boundary position is also used for setting a new window. If the block currently being processed is the last block in the row direction and is not the last block in the column direction, for example, if it is the blocks c, f, i, l, o in FIG. The power block is the first block of the new block row. Therefore, the boundary position of the first block of the previous block row, for example, the boundary position 103 shown in FIG. 10 can also be used.

In this way, the last block (see FIG. 8)
Then, the processing of steps S93 to S96 is repeated up to r blocks), the values of all cells are read, and all encoded bits are obtained (S911).

This mesh reading process can be modified and changed in various ways. For example, in the above example, each block is
The number of rows and the number of columns are eight, but the number is not limited to this. The number of vertical and horizontal grids of the image may be an integral multiple of the vertical and horizontal grids of the block. The size of this block can be freely determined without depending on the recording method. Increasing the size of the meshes that make up a block makes it easier to identify the boundaries of the block and reduce the amount of processing. However, if it becomes too large, it becomes difficult to deal with local distortion.

Further, instead of the boundary of each block, the center line passing through the grid may be obtained.

Further, each side of the window may be represented by a straight line instead of a sequence of points. In this case, the boundary position is not determined for each sampling point, but when the value at each point is different, it is set to 1 and when the value is the same, the total value of all sampling points is calculated, and this is calculated for the 1st sampling. repetition,
The block boundary is where this total value is maximum. This facilitates processing.

By the way, as shown in FIG. 14, an error may occur due to local distortion between the actual block boundary 142 and the block boundary estimated value 141.
This tendency becomes remarkable when the capacity is increased, that is, when the image area is increased or the recording density is increased.

Then, a modified example of the above-mentioned mesh reading process will be described.

FIG. 12 shows an image 32 on a recording surface 33 as a record carrier. In this image 32, a mark 121 for indicating the data start position is provided in each square of the image pattern 30 shown in FIG. 3, and clock marks 122 are provided around the image pattern 30 at a predetermined pitch.

This image 32 is stored in the image RAM 44 as shown in FIG. 13 (the whole image data is shown by reference numeral 62). In the reading process, the block boundary is set at a position passing through the center of the clock mark 122. The clock mark 122 is used as a reference point for setting the window. The sampling of each block is performed in the same manner as the above example.

In this case, by using an appropriate block size and expressing the boundary by a sequence of points, the value of the mesh in the block is sampled with high accuracy by following the local distortion. be able to. By using the estimated value of the boundary of the block by the data start position mark 121 and the clock mark 122 as a reference line for knowing the overall inclination and detecting the local distortion, the processing amount is reduced and the block is identified. The accuracy of can be improved.

(Second Embodiment) A digital information recording method and a reading method according to an embodiment of the present invention will be described with reference to FIGS. The column of the second embodiment uses only the patterns in which the number of continuous bright and dark is a certain number or less among the patterns that can be taken in the block,
It is premised on a recording method in which light and dark are almost evenly distributed on a record carrier (such a technique is disclosed in Japanese Patent Application No. 6-70).
911).

More specifically, a square block consisting of p (vertical) × p (horizontal) squares is used as a block, and the number of consecutive squares having the same value in the vertical and horizontal directions is k ( It is guaranteed that k is an integer of 1 or more and less than p) or less. In this case, the number of patterns (designated patterns) that can be used is shown in FIG.

As can be seen from FIG. 15, for example, 4 × 4
If a square consisting of squares is used as a block, and up to three consecutive vertical and horizontal directions are allowed within the block (no consecutive four), the number of designated patterns is 22,8.
There are 74 ways. Since the number of patterns required to represent 14-bit information is 16,348, block 1
14 bits of information can be represented. The original data can be recorded by dividing the data string of the recording information into 14-bit sections, mapping each to a specified pattern, and arranging the mapped specified patterns two-dimensionally on the recording surface.

The data recorded on the record carrier (recording surface) in this manner can, of course, also be read in block units as in the previous example. FIG. 16 shows a part of the image data 160. The image data 160 has blocks 161 and 1 made up of cells 4 × 4 as in the recording.
It is divided into 62, ... And reading is performed in block units.

Here, at the time of reading the lightness / darkness of the cells in the block, as shown in FIG. 17, there are cases in which a reading error occurs due to dirt or blurring at the time of recording, adhesion of dust on the medium, or distortion. In block 161, there is a possibility of misreading the light cells in the dark cells due to the dirt indicated by hatching “△”. In block 162, the dark cells are read in the light cells due to the dust indicated by “△”. You can make a mistake. At this time, the block is read in the same block unit as the block at the time of recording, and at the time of reading the block, the block is read based on a recording method that allows only a maximum of three continuous brightnesses in the vertical and horizontal directions. You can detect mistakes. This read error information is recorded so that it can be used for later error correction processing. For example, if a check code is added in units of blocks at the time of encoding, the location of the error is identified when the block is read. Therefore, so-called erasure correction (error correction when the error position is known) becomes possible. Alternatively, the reading may be performed again with the sampling position corrected.

Further, as described above, a square consisting of 4 × 4 grids is set as a block, and up to three consecutive pieces of the same lightness are permitted in the block in the vertical and horizontal directions (there are no consecutive four pieces). In this case, of the 22,874 specified patterns, the number of patterns required to represent 14-bit information is 16,384, so the remaining 6,4
90 patterns can be used for some control information. Here, the remaining patterns that are not used for the control information are set as invalid patterns, and an invalid pattern table (or 14-bit information-block pattern mapping table) is created by collecting the invalid patterns. In this case, the reading error of the block can be detected by inspecting the invalid pattern table. For example,
FIG. 18 shows an example in which the invalid pattern 182 shown in FIG. 18B is read as a result of erroneous reading of the block 163 shown in FIG. The original pattern is the effective pattern 181 shown in FIG. By inspecting the invalid pattern table, the pattern read from the block 163 coincides with the invalid pattern 182, so that it is determined that the block 163 was erroneously read. Also in this case, this information can be recorded so that it can be used for the later error correction processing. For example, if a check code is added in units of blocks at the time of encoding, the location of the error is identified when the block is read. Therefore, so-called erasure correction (error correction when the error position is known) becomes possible. Alternatively, the reading may be performed again with the sampling position corrected.

Further, as described above, a square consisting of 4 × 4 grids is set as a block, and up to three consecutive pieces of the same lightness are allowed in the block in the vertical and horizontal directions (not four consecutive rows). If you do, it is guaranteed that four or more cells with the same brightness will not continue in the block.
A maximum of 7 consecutive blocks may span adjacent blocks. As described above, depending on the source data and the arrangement at the time of encoding, it cannot always be guaranteed that the cells having the same lightness are not continuous at the block boundaries. For example, in the image data 190 shown in FIG.
And blocks 192,1 adjacent to the four sides of this block
At the boundaries with 93, 194, and 195, cells with the same brightness are arranged on both sides of the boundaries. Therefore, the boundary of the block 191 cannot be determined as it is.

Therefore, at the time of recording, the boundary between the previously arranged block and the block to be arranged adjacent thereto is checked. Then, when it is expected that the boundary is not clear, for example, a block having a boundary identification pattern 200 as shown in FIG. 20 is inserted. FIG. 22 shows a state in which the block 221 having the boundary identification pattern 200 is inserted in the image 190 of FIG. Block 1
Reference numeral 91 is arranged at a position adjacent to the right by shifting one block to the right, and the block 221 having the boundary identification pattern 200 is arranged at the position of the original block 191.

The boundary identifying pattern 200 shown in FIG.
Is one of the control patterns and is not used as a pattern representing 14-bit information. The boundary identifying pattern 200 of FIG. 20 has the characteristic that white and black appear alternately, and therefore it is easy to identify itself. In addition, since black and white appear alternately on any side, in most cases, the boundary can be easily identified.

As the boundary identifying pattern, the pattern 210 shown in FIG. 21 has the same property as the pattern 200 of FIG. 20, and is suitable as the boundary identifying pattern. FIG. 23 shows the boundary identification pattern 21 in the image 190 of FIG.
A state in which a block 231 having 0 is inserted is shown.
The block 191 is arranged at a position adjacent to the right by shifting by one block, and the block 231 having the boundary identification pattern 210 is arranged at the position of the original block 191.

For the block having the boundary identifying patterns 200 and 210, the difference between the values of the pair of cells (coded bits) adjacent to each other across the boundary at the boundary with the block adjacent to this block is It is possible to calculate with 1 being different and 0 being the same. For example, in FIG.
In the block 221 of 2 + 1 + 3 =
6, in the block 231 of FIG.
+ 3 + 1 = 6, and there is no difference in either pattern. Note that it is possible to prepare both the patterns 200 of FIG. 20 and the pattern 210 of FIG. 21 as the boundary identification patterns and use them according to the patterns of adjacent blocks. That is, the boundary identification pattern 20
At the boundary between the block having 0 or 210 and the block adjacent to the upper, left, and right sides of this block, the difference between the values of pairs of cells (coded bits) adjacent to each other across the boundary is calculated, The pattern for boundary identification with the larger value (boundary identification value) is adopted. As a result, the boundary identification accuracy can be improved.

More specifically, the encoding process is performed according to the flow shown in FIG. It is assumed that the source data has already been stored in the RAM 13 (FIG. 1).

When the blocks are arranged two-dimensionally, they are arranged in the row direction. For example, assuming that there are eight blocks arranged in each row, source data of 14 bits × 8 = 112 bits = 14 bytes can be expressed by 8 blocks in 1 row, and 16 blocks in 8 rows per line.
Bit × 8 = 128 bits = 16 bytes of data. On the RAM 13, the encoded bits can be accessed as a two-dimensional array of M rows × 8 columns of memory cells capable of storing 16 bits. The value of the number of rows M depends on the size of the source data and the like.

First, the values of the subscripts (i, j) of the two-dimensional array of blocks are initialized on the RAM 13, and i = 1, j = 1.
(S241).

Next, the data string of the source data is read out from the RAM 13 in units of 14 bits (S24).
2). Then, with reference to the block mapping table (shown in FIG. 26B) configured by 14-bit information and coded bits 4 × 4 stored in the ROM 12,
16-bit coded bits are obtained from the read 14-bit information (S243). Source data (recorded data)
The correspondence between the coded bits and the encoded bits is as shown in FIG. The 16-bit coded bits are stored as one block in the two-dimensional array (i, j) on the RAM 13 (S244).

When (i, j) = (1,1), steps S245 and S246 are omitted and 1 is added to the value of j (S247). Since the first line is not yet completed and j> 8 is not satisfied (S247), as long as there is unprocessed source data (S249), the process returns to step S242 and the data string of the source data is divided into 14-bit units. read out.

Then, as in the previous case, the read 14
Obtain 16 encoded bits from the bit information (S
243), and stores the 16 coded bits as one block in the two-dimensional array (i, j) on the RAM 13 (S244).

In step S245, the block (i, j
−1) and blocks (i−1,
j), (i, j-2), and (i, j), the boundary identification value α is calculated. That is, the difference between the values of the four pairs of cells adjacent to each other across the boundary is set to 1 if the values of the pair of cells are different, and to 0 if they are the same, and the values of each side are summed to obtain the boundary identification value α. calculate. When (i, j) = (1,2), only the block (1,1) and the block (1,2) adjacent to the right are arranged. Thus, the block (i, j-
1) When the block is not arranged on either the upper left side or the right side, the boundary identification value α is calculated for the side not arranged on the basis of the brightness of the record carrier.

Next, it is judged whether or not the boundary identification value α exceeds a prescribed threshold value (for example, 2) (S246).
If it exceeds, it is easy to identify the boundary at the time of reading, so as long as there is unprocessed source data (S24
9) Then, the process directly returns to step S242 to read the source data.

Then, the processes of steps S242 to S249 are repeated. When the value of j exceeds 8 in step S247, it means that the arrangement of one row of the block has been completed, and therefore 1 is added to the value of i and j = 1 is set (S248). Then, as long as there is unprocessed source data, the processes of steps S242 to S249 are repeated.

Now, in step S246, when the boundary identification value α is equal to or less than the specified threshold value, it becomes difficult to identify the boundary of the block (i, j-1) when reading as it is. Therefore, the processing of steps S2410 to S2415 is performed, and the boundary identification pattern 2 shown in FIGS.
Insert 00 or 210.

That is, the boundary identification patterns 200, 2
When 10 is inserted at the position of each two-dimensional array (i, j-1), the block having the inserted boundary identification pattern and the blocks (i-1, j) adjacent to the left and right of
A boundary identification value β is calculated for a boundary between (i, j-2) and (i, j-1) (S2410). Then, of the boundary identification patterns 200 and 210, a pattern is selected in which the boundary identification value β is larger than the specified threshold value (S2411) and is larger than the other boundary identification value β (S24).
12) Insert the selected pattern 200 or 210 at the position (i, j-1) (S2413, S24).
14). Then, the original block (i, j-1) is converted to (i,
j), and the original block (i, j) is moved to (i, j + 1) (S2415).

After that, the process returns to step S247 to continue the processing.

As described above, the blocks forming the upper and left sides of the entire two-dimensional array, that is, the blocks in the first row, the blocks in the first column, and the blocks in the eighth column have adjacent blocks. For the edges that do not exist, the boundary identification values α and β are calculated based on the brightness of the record carrier. Further, when the boundary identification pattern is inserted in the eighth column, the original block (i-1,8) is moved to the position (i, 1), and the next block (i, 1) is moved to (i , 2)
Move to the position. When the boundary identification pattern is inserted in the 7th column, the original block (i, 8) is moved to the position of the beginning (i + 1, 1) of the next row.

In this way, when the blocks are arranged in a two-dimensional grid, it is possible to avoid the arrangement in which the boundaries of the blocks are difficult to be identified at the time of reading, and it is possible to improve the identification accuracy of the blocks.

This encoding process can be modified and changed in various ways.

For example, when the boundary identification value β is calculated by inserting the boundary identification pattern (S2410), the boundary identification value β may be calculated not only for the upper and left sides but also for the boundaries of the upper, lower, left and right sides. The boundary identification value β is the block 22 of FIG.
1, 2 + 0 + 1 + 3 = 6 in the order of upper, lower, left and right sides, as shown in FIG.
In the block 231 of No. 3, 2 + 4 + in the order of the upper, lower, left and right sides.
3 + 1 = 10. Therefore, in this example, FIG.
By inserting the boundary identifying pattern 210, the block identifying accuracy can be improved.

Further, when there is a difference in the reading accuracy or the sampling accuracy depending on the scanning direction, the boundary identifying pattern β is calculated by inserting the boundary identifying pattern (S2).
410), each side may be weighted. For example, when the main scanning direction is more accurate than the sub scanning direction at the time of reading, the boundary identification value β may be calculated only for the boundaries between the upper and lower sides, without considering the highly accurate left and right sides. Then, based on the boundary identification value β, it may be determined whether or not to insert the boundary identification pattern or which boundary identification pattern (200 or 210) is to be used.

Next, the grid reading process and the decoding process will be described with reference to the flow chart shown in FIG.

Image data is already stored in the image RAM 44 (FIG. 4).

First, the image data in the image RAM 44 is searched to measure the start position and inclination of the data (S25).
1). Then, after correcting the calculated inclination, a window is set based on the block size determined by the number of cells (here, 4 × 4), and the first block is included based on the data start position. A window is set at the position (S251).

Next, the boundary of the block is detected, and the detected boundary position of the block is stored in the RAM 42 (S252). This is because it is used as an initial value when detecting a boundary between blocks adjacent to the block.

Next, the block in which the boundary is detected is 4 rows ×
By dividing into four columns, the position at which the cell is sampled is obtained, and the image bit at that position is sampled to obtain the value of the cell (S253). And
The value of the obtained mesh is stored in the RAM 42.

Next, it is inspected that the number of continuous cells having the same value in this block is 3 or less in both the vertical direction and the horizontal direction (not continuous 4).
4). If four are consecutive, it is determined that an error has occurred, the sampling position is corrected (S2510), and the process returns to S253 to perform sampling again. Further, the block in which an error has occurred may be recorded in the RAM and used for error correction.

Next, it is checked whether the pattern of the block is a valid pattern (S255). If it is not a valid pattern, the sampling position is corrected (S
2510), the process returns to S253 and sampling is performed again. Further, the block in which an error has occurred may be recorded in the RAM and used for error correction.

On the other hand, in step S255, if the pattern of the block is a valid pattern, it is further inspected whether it is a boundary identification pattern (S256). When the pattern of the block is the boundary identification pattern, the next new window is set (S2511), and the process returns to step S252 to detect the next block. If the block pattern is the data pattern in step S256, the mapping table stored in the ROM 43 is referenced, and the fragment of the source data is obtained from the coded bit (16 bits) (S257). It should be noted that the mapping table shown in FIG. 26 may be reverse-looked up, or another table arranged by coded bits may be referred to.

Next, it is judged whether or not it is the last block (S258), and if it is not the last block, the next new window is set (S2511), and the process returns to step S252 and the next block is selected. Detect. In this manner, by repeating the process up to the last block, it is possible to read the values of all the cells and obtain all the source data.

(Third Embodiment) Next, a digital information recording method and a reading method according to an embodiment of the present invention will be described with reference to FIGS. 27 to 38. The section of the third embodiment is based on a recording method in which, when arranging light and dark cells with a matrix code on a record carrier, random number processing is performed to balance light and dark and to distribute them substantially uniformly. To do. When the random number processing is performed, there is no specially secured pattern as the control information, so a pattern suitable for the boundary identification pattern is generated as a pattern representing the print data.

FIG. 27 shows a part of the image data 260 obtained by reading the optical information recording medium recorded by random number processing. In this example, the image data 260 is 4 × 4.
It is configured by arranging square blocks of square cells in a two-dimensional array.

In this example, the block 261 and blocks 262, 263, and 26 adjacent to the four sides of this block.
At the boundaries with 4, 265, cells with the same brightness are arranged on both sides of the boundaries. Therefore, the boundary of the block 261 cannot be determined as it is.

Therefore, at the time of recording, for example, the boundary identification patterns 270, 28 as shown in FIG. 28 or 29.
Insert a block with 0. Border identification pattern 27
Patterns 0 and 280 are composed of 4 × 4 cells, and white and black appear alternately. This facilitates identification of block boundaries. The pattern 270
In, the cells in the upper left corner are given black, and in the pattern 280, the cells in the upper left corner are given white. The bright and dark patterns of the right side and the left side or the upper side and the lower side of the patterns 270 and 280 are opposite.

When the same pattern as the boundary identification pattern is generated by the random number processing, two blocks having the pattern are made continuous. As a result, it is possible to distinguish the block representing the print data from the block representing the boundary identification pattern. At this time, in the continuous boundaries of the same pattern (border identification pattern), the meshes arranged on both sides of the boundary have different values, so that the block boundaries can be easily identified. Note that it is necessary to encode the boundary identification patterns themselves so that two patterns do not continue because they cannot be distinguished at the time of reading.

FIG. 30 shows a state in which the block 291 having the boundary identifying pattern 270 is inserted in the image 260 of FIG. The block 261 is shifted to the right by one block and arranged at the position on the right side, and the block 291 having the boundary identification pattern 270 at the position of the original block 261.
Is arranged. Further, FIG. 31 shows the image 26 of FIG.
A state in which the block 301 having the boundary identification pattern 280 is inserted at 0 is shown. The block 261 is shifted to the right by one block and is arranged at the position on the right side, and the block 301 having the boundary identification pattern 280 is arranged at the position of the original block 261. As described above, by selectively using the patterns 270 and 280 of both FIG. 28 and FIG. 29 depending on the state of the boundary with the adjacent block, it becomes easier to identify the boundary of the block.

As shown in FIG. 32, in the case where the image data 310 is formed by arranging square blocks of 5 × 5 cells in a two-dimensional array, the boundary identification pattern shown in FIG.
As shown in FIG. 3 or FIG. 34, it is conceivable to use patterns 320 and 330 which are composed of 5 × 5 cells and in which black and white appear alternately. In the pattern 320, the cells at the four corners are black, and in the pattern 330, the cells at the four corners are white.

However, when one side of the square block is composed of an odd number of cells, the boundary identification patterns 320 and 33 are formed.
The bright and dark patterns on the right side and the left side or the upper side and the lower side of 0 are the same. Therefore, when the same pattern as the boundary identification pattern 320 or 330 is generated by the random number processing, as in the case where one side of the square block is composed of an even number of squares, the block having the pattern is divided into two blocks.
If they are made continuous, it becomes difficult to identify the boundaries between the blocks.

Therefore, if the same pattern as the boundary identifying pattern 320 is generated by the random number processing, the pattern shown in FIG.
As shown in (a), the pattern 320 and the boundary identification pattern 330 are combined and made continuous. When the same pattern as the boundary identification pattern 330 is generated by the random number processing, the pattern 330 and the boundary identification pattern 320 are combined and made continuous as shown in FIG. Thereby, the boundary can be easily identified.

More specifically, the encoding process is performed according to the flows shown in FIGS. 36 and 37. It is assumed that the source data has already been stored in the RAM 13 (FIG. 1). Further, it is assumed that the block is a square and the number m of cells on one side of the square is designated.

When the blocks are arranged two-dimensionally, they are arranged in the row direction. For example, if eight blocks are arranged in each row, the coded bits can be accessed as a two-dimensional array of N rows × 8 columns of memory cells capable of storing m ² bits on the RAM 13. The number of rows N depends on the size of the source data and the like.

First, the values of the subscripts (i, j) of the two-dimensional array of blocks are initialized on the RAM 13, and i = 1, j = 1.
(S351).

Next, the data string of the source data is read out from the RAM 13 in units of m × m bits (S35).
2). Then, pseudo random number processing is performed on the read m × m bits of information to obtain m × m bits of encoded bits (S353). A two-dimensional array (i,
It is stored in j) (S354).

Next, it is checked whether the pattern of the block stored in (i, j) is the same as the boundary identification pattern (S355). If the pattern of the block is the same as the pattern for boundary identification, processing S3517-
S3519 is performed. That is, it is determined whether m is an even number or an odd number (S3517), and when m is an even number, a block having the same boundary identification pattern as the boundary identification pattern is set to (i, j + 1). (S351
8). When m is an odd number, a block having a boundary identification pattern different from the boundary identification pattern is (i, j +
It is set to 1) (S3519). When j = 8, that is, when the block is already arranged up to the eighth column, it is set at the beginning (i + 1,1) of the next row.

On the other hand, if the block pattern is different from the boundary identification pattern in step S355, the process directly proceeds to step S356.

When (i, j) = (1,1), steps S356 and S357 are omitted and 1 is added to the value of j (S358). Since the first row is not finished yet and j> 8 is not satisfied (S358), as long as there is unprocessed source data (S3510), the process returns to step S352 and the data string of the source data is set in m × m bit units. Read them in sections.

Then, as in the previous case, the read m ×
The m × m coded bits are obtained from the m bit information (S353), and the m × m coded bits are stored as one block in the two-dimensional array (i, j) on the RAM 13 (S354). .

In step S356, the block (i, j
−1) and blocks (i−1,
j), (i, j-2), and (i, j), the boundary identification value α is calculated. That is, the difference between the values of the m pairs of cells adjacent to each other across the boundary is set to 1 when the values of the pair of cells are different, and to 0 when the values of the pair of cells are the same, and the values of the respective sides are summed to obtain the boundary identification value α. calculate. When (i, j) = (1,2), only the block (1,1) and the block (1,2) adjacent to the right are arranged. Thus, the block (i, j-
1) When the block is not arranged on either the upper left side or the right side, the boundary identification value α is calculated for the side not arranged on the basis of the brightness of the record carrier.

Next, it is judged whether or not the boundary identification value α exceeds a prescribed threshold value (for example, m / 2) (S35).
7). If it exceeds, it is easy to identify the boundary at the time of reading, so as long as there is unprocessed source data (S3
510), the process directly returns to step S352 to read the source data.

Then, steps S352 to S3510.
Is repeated. If the value of j exceeds 8 in step S358, it means that the arrangement of one row of the block has been completed, so that 1 is added to the value of i and j is set to 1 (S359). Then, as long as there is unprocessed source data, the processes of steps S352 to S3510 are repeated.

Now, in step S357, when the boundary identification value α is equal to or less than the specified threshold value, it becomes difficult to identify the boundary of the block (i, j-1) at the time of reading as it is. Therefore, the processing of steps S3511 to S3516 is performed, and for example, when m = 4, the boundary identification pattern 270 or 280 shown in FIGS. 28 and 29 is inserted.

That is, the boundary identifying patterns 270, 2
When 80 is inserted in each of the positions of the two-dimensional array (i, j-1), the block having the inserted boundary identification pattern and the blocks (i-1, j) adjacent to the upper and left sides thereof,
A boundary identification value β is calculated for the boundary between (i, j-2) and (i, j-1) (S3511). Then, of the boundary identification patterns 270 and 280, a pattern is selected in which the boundary identification value β is larger than the specified threshold value (S3512) and is larger than the other boundary identification value β (S35).
13) Insert the selected pattern 270 or 280 at the position (i, j-1) (S3514, S35).
15). Then, the original block (i, j-1) is converted to (i,
j), and the original block (i, j) is moved to (i, j + 1) (S3516).

Thereafter, the process returns to step S358 to continue the processing.

As described above, the blocks forming the upper and left sides of the entire two-dimensional array, that is, the blocks in the first row, the blocks in the first column, and the blocks in the eighth column have adjacent blocks. For the edges that do not exist, the boundary identification values α and β are calculated based on the brightness of the record carrier. Further, when the boundary identification pattern is inserted in the eighth column, the original block (i-1,8) is moved to the position (i, 1), and the next block (i, 1) is moved to (i , 2)
Move to the position. When the boundary identification pattern is inserted in the 7th column, the original block (i, 8) is moved to the position of the beginning (i + 1, 1) of the next row.

When m = 5, the boundary identifying pattern 320 or 330 shown in FIGS. 33 and 34 may be used.

In this way, when the blocks are arranged in a two-dimensional grid, it is possible to avoid the arrangement in which the boundaries of the blocks are difficult to be identified at the time of reading, and the block identification accuracy can be improved.

Next, the grid reading process and the decoding process will be described in more detail with reference to the flow chart shown in FIG.

It is assumed that image data has already been stored in the image RAM 44 (FIG. 4).

First, the image data in the image RAM 44 is searched to measure the start position and inclination of the data (S36).
1). Then, after correcting the obtained inclination, a window is set based on the block size determined by the number m of the grids on one side, and the window is set at a position including the first block based on the start position of the data. Set (S3
61).

Next, the boundary of the block is detected, and the detected boundary position of the block is stored in the RAM 42 (S362). This is because it is used as an initial value when detecting a boundary between blocks adjacent to the block.

Next, the block in which the boundary is detected is m rows ×
By dividing into m columns, the position where the grid is sampled is obtained, and the value of the grid is obtained by sampling the image bit at that position (S363). And
The value of the obtained mesh is stored in the RAM 42.

Next, it is checked whether the pattern of the block is a boundary identification pattern (S364). When the pattern of the block is the data pattern, the process directly proceeds to step S365.

On the other hand, when the pattern of the block is the boundary identifying pattern, the pattern of the immediately preceding block is detected to determine whether or not two boundary identifying patterns are continuous (S365). Boundary identification pattern is 2
When not consecutive, the pattern is the original boundary identification pattern, so the next new window is set (S36).
8) Return to step S362 to detect the next block. On the other hand, when two boundary identification patterns are consecutive in step S369, it means that the pattern is a data pattern. When m is an even number, the boundary identification pattern is used as it is as a data pattern. When m is an odd number, the pattern of the immediately preceding block is used as the data pattern. Then, the process proceeds to step S365.

In step S365, pseudo-random number processing is performed to obtain m × m bit source data fragments. The obtained source data fragment is stored in the RAM 42.

Next, it is judged whether or not it is the last block (S366), and if it is not the last block, a next new window is set (S368), and the process returns to step S362 and the next block is selected. Detect. By repeating the process up to the last block in this manner, it is possible to read the values of all the cells and obtain all the source data (S367).

Various modifications and changes can be made to the third embodiment. For example, although encoding is performed in block units, if the order at the time of encoding and that at the time of decoding are the same, encoding is performed in the row direction or the column direction of the cells, and each fragment is combined at the time of decoding. Good.

In the first to third embodiments described above,
For simplicity of explanation, the grid values are expressed in black and white, but the invention is not limited to this, and other expressions can be used. For example, the cells may be painted in different colors or shades, and may be distinguished by unique optical information such as a mesh. In addition, it is possible to use multi-valued information representing one mesh by using colors, shades, mesh patterns, etc. The shape of the grid may be rectangular.

Further, depending on the characteristics of the reading device (for example, when a line sensor is used), the reading accuracy varies depending on the scanning direction, and therefore it may be advantageous to make the shape of each grid rectangular for high density.

(Fourth Embodiment) Next, a record carrier will be described with reference to FIGS.

An ink for recording invisible data by printing absorbs or reflects infrared rays or ultraviolet rays, or emits fluorescence when excited by infrared rays or ultraviolet rays (hereinafter referred to as "acting with infrared rays or ultraviolet rays"). It has been known.

For example, "10100" shown in FIG. 39 (a).
The specified pattern representing the following recording information is printed on the recording surface of the record carrier 90 shown in FIG. 2B by the ink I ₁ corresponding to “1” and the ink I ₂ corresponding to “0”. Is represented. Ink I ₁ exhibits a specific color in the visible light region,
Also, the ink is one that works with infrared rays or ultraviolet rays. For example, the ink has a property of absorbing infrared rays. on the other hand,
Ink I ₂ shows the same color as Ink I _{1 in} the visible light range,
Also, the ink has the property of not absorbing infrared rays. In this case, since the inks I ₁ and I ₂ have the same color in the visible light region, it is not possible to discriminate what information is recorded with the naked eye, and a reading device that detects infrared rays is used. Only then can the recorded information be read. Therefore, the recorded information can be made confidential.

The ink I ₁ has a specific color in the visible light region and has the property of absorbing ultraviolet rays, while the ink I ₂ has the same color as the ink I _{1 in the} visible light region. Also, an ink having a property of not absorbing ultraviolet rays may be used. Similarly, the ink I ₁ has a specific color in the visible light region and emits fluorescence when excited by infrared rays or ultraviolet rays, while the ink I ₂ has the same color as the ink I _{1 in the} visible light region. As shown, and contrary to ink I _1, it may not be excited by infrared or ultraviolet light. Even in this case, similarly, the recorded information can be made confidential.

The recording information "10100" shown in FIG. 40 (a) is printed on the recording surface of the record carrier 90 by the ink I corresponding to "1" as shown in FIG. 40 (b). ₃
And an ink-free portion corresponding to “0” may be represented. The ink I ₃ is transparent in the visible light region and acts with infrared rays or ultraviolet rays. In this case, since the ink I ₃ is transparent in the visible light region, it is not possible to discriminate what information is recorded with the naked eye, and by using a reading device that detects infrared rays or ultraviolet rays, The recorded information can be read for the first time. Therefore, the recorded information can be made confidential.

The recording information "10100" shown in FIG. 41 (a) is printed on the recording surface of the record carrier 90 by the ink I corresponding to "1" as shown in FIG. 41 (b). _Four
And an ink-free portion corresponding to “0”, and a white ink layer I ₅ that blocks visible light and transmits infrared rays or ultraviolet rays may be covered thereover. The ink I ₄ shall act with infrared rays or ultraviolet rays. In this case, it is not possible to discern what information is recorded with the naked eye because of the white ink layer I ₅ , and the recorded information is read for the first time by using a reading device that detects infrared rays or ultraviolet rays. be able to. Therefore, the recorded information can be made confidential. In addition,
The ink I ₄ may exhibit a specific color in the visible light region and may be transparent.

Further, as shown in FIG. 42, original recording information may be printed as invisible data D ₁ , and visible data D ₂ representing a rabbit picture or the like may be superimposed and printed thereon.
The invisible data D ₁ is transparent in the visible light region and acts with infrared rays or ultraviolet rays, and the visible data D ₂ is
The ink I ₇ is of a type that shows a specific color in the visible light region and transmits infrared rays or ultraviolet rays.
In this case, the recorded information D ₁ can be made secret, and the information D ₁ and D ₂ can be multiplexedly recorded on a limited recording surface, so that the recording capacity per unit area can be increased. it can. As the visible data, completely unrelated information D ₂ having a format different from that of the invisible data D ₁ may be printed as shown in FIG. 42, or the format used for the invisible data D ₃ as shown in FIG. 43. The information D ₄ may be printed using the same format as.

FIG. 44 shows a main part of the reading device 99 for reading the record information recorded by the above-mentioned multiplex printing. This reading device 99 has
A first light emitting element 91 and a second light emitting element 92 that irradiate the recording surface of the record carrier 90 with infrared light, and a first light receiving element that detects visible light and infrared light reflected by the recording surface, respectively. An element 93 and a second light receiving element 94 are provided. The light receiving element 93 is provided with a filter so as not to react with infrared light. It is assumed that the invisible data D ₁ and the visible data D ₂ are printed on the recording surface of the record carrier 90 in the state shown in FIG. Further, the ink I ₆ representing the invisible data D ₁ is of a type that is transparent in the visible light region and emits fluorescence having another infrared wavelength when excited by infrared light.

Regarding the visible data D ₂ , the light emitting element 91
Irradiates the recording surface with visible light, and the light receiving element detects the visible light reflected by the recording surface, whereby reading is performed. Regarding the invisible data D ₁ , the light emitting element 92 irradiates the recording surface with infrared light, and the light receiving element 94 detects fluorescence (infrared light) emitted by the invisible ink.
The reading is done. In this way, the visible data D ₂ and the invisible data D ₁ can be read independently of each other.

[0147]

As is apparent from the above, according to the digital information decoding method of the first aspect, the image data read from the record carrier is virtually divided into rectangular block units composed of a plurality of cells, and the image data is obtained. For each block, the rectangular windows surrounding this block are sequentially set, and the marks given to the cells in the blocks are identified, so it is sufficient to identify the light and dark patterns in block units.
Compared to accessing all image data,
The identification process can be simplified. Further, since it is not necessary to access all the image data, the data processing amount can be reduced and the processing time can be shortened. Further, it becomes possible to correct local distortion of the image data, and the data recognition rate can be improved.

In the digital information decoding method of the second aspect, when setting the window in the first block of the image data, a window having a size larger than the size of this block is set outside the first block. Then, the edge of the window is moved inward in image bit units representing the image data to detect a change in the value of the image bit that the edge of the window crosses, and the boundary at which the value of the image bit changes. Stop the above side of the above window in position. Therefore, the block boundary can be detected with high accuracy. The boundaries of the second and subsequent blocks can be detected using the boundary position of the first block.

In the digital information decoding method of the third aspect, of the pattern groups that the block can take by adding marks to each of the cells, a pattern that is not adopted as the designated pattern is preset as an invalid pattern, For each block of the read image data, a rectangular window surrounding this block is sequentially set, and the marks provided in the cells in each block are identified to determine whether it is a designated pattern or an invalid pattern. Then, when the pattern of the block is a designated pattern, the information represented by the mark is decoded to obtain the original digital information, while when the pattern of the block is an invalid pattern, a process of correcting a read error is performed. In this case, the reading error can be corrected at the time when the block is read, and the bright and dark patterns in the block are corrected, so that reliable reading can be performed.

In the digital information recording method of the fourth aspect, a specific pattern other than the designated pattern among the pattern groups that the block can take is adopted as a boundary identification pattern for identifying the boundary of the block, When arranging the designated patterns on the recording surface, it is determined whether or not the boundary between the block already arranged and the block to be arranged is easily discriminated based on the pattern of the cells sandwiching the boundary. Then, when the boundary is easy to identify, the block is arranged as it is, while when the boundary is not easy to identify, a block having the boundary designating pattern is arranged in place of the block. In this case, it is possible to avoid an arrangement in which it is difficult to identify block boundaries during reading, and it is possible to improve block identification accuracy.

In the digital information recording method of the fifth aspect, among the pattern groups that the block can take, a pattern in which two types of marks representing different values are alternately provided is used as a boundary identification for identifying the boundary of the block. When arranging the above-mentioned block-unit coding patterns on the recording surface, it is determined whether or not the boundary between the block already arranged and the block to be arranged is easy to distinguish. Make a decision based on the pattern. Then, when the boundary is easy to identify, the block is arranged as it is, while when the boundary is not easy to identify, a block having the boundary designating pattern is arranged in place of the block. In this case, it is possible to avoid an arrangement in which it is difficult to identify block boundaries during reading, and it is possible to improve block identification accuracy.

In the digital information recording method of the sixth aspect, when the same coding pattern as the boundary identification pattern is generated by the random number processing, when the block-by-block coding patterns are arranged on the recording surface, the boundary identification pattern is used. A plurality of blocks having the same coding pattern as the pattern are continuously arranged. Therefore, it is possible to distinguish the original boundary identification pattern from the coding pattern corresponding to the digital information to be recorded by imposing a restriction that a plurality of original boundary identification patterns are not arranged consecutively. You can

According to the digital information recording method of the seventh aspect, it is judged whether or not the boundary between the already arranged block and the block to be arranged is easy to discriminate between the adjacent pairs of cells with the boundary therebetween. Among them, since it is performed based on how many pairs of cells having different values exist, it is possible to objectively judge whether or not it is easy to identify. As a result, it is possible to reliably avoid an arrangement in which it is difficult to identify the boundaries of the blocks during reading, and it is possible to further improve the identification accuracy of the blocks.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration of an optical information recording apparatus used to carry out the present invention.

2 is a diagram showing an overall operation flow of the optical information recording apparatus of FIG.

FIG. 3 shows an optical information record carrier to be read.
It is a figure which illustrates the digital information recorded by the dimensional grid arrangement.

FIG. 4 is a diagram showing an overall configuration of an optical information reading device used to carry out the present invention.

5 is a diagram showing an overall operation flow of the optical information reading device of FIG.

6 is a diagram showing image data obtained by reading the image of FIG.

FIG. 7 is a diagram showing an example in which the image data of FIG. 6 is divided into a plurality of blocks.

FIG. 8 is a diagram illustrating an order of reading a plurality of blocks in FIG.

FIG. 9 is a diagram showing a flow of a grid reading process according to an embodiment of the present invention.

FIG. 10 is a diagram showing a state in which a window is set in the first block.

FIG. 11 is a diagram schematically illustrating an example in which image data is stored in an image RAM.

FIG. 12 is a diagram exemplifying image data obtained by reading an image to which a clock mark is added.

FIG. 13 is a diagram showing an example in which the image data of FIG. 12 is divided into a plurality of blocks on the basis of a clock mark.

FIG. 14 is a diagram illustrating a deviation between an estimated value of a block boundary and an actual block boundary due to a clock mark.

FIG. 15 shows the number of squares forming one side of a block, the maximum number of the same values allowed to continue in the vertical and horizontal directions within the block, and the number of designated patterns when the block is formed in a square shape. It is a figure showing the relationship of.

FIG. 16 is a diagram showing an example in which part of image data read from an optical record carrier in which blocks are formed in a square shape and digital information is recorded is divided into the same block units as in recording.

FIG. 17 is a diagram showing an example in which some of the blocks in FIG. 16 are erroneously read.

FIG. 18 is a diagram showing an example in which some of the blocks in FIG. 16 are erroneously read.

FIG. 19 is a diagram showing an example in which there is no information for determining a boundary between adjacent blocks.

FIG. 20 is a diagram showing an example of a pattern used for boundary identification.

FIG. 21 is a diagram showing another example of a pattern used for boundary identification.

22 is a diagram showing a state in which the boundary identification pattern shown in FIG. 20 is inserted in the image of FIG.

23 is a diagram showing a state in which the boundary identification pattern shown in FIG. 21 is inserted in the image of FIG.

FIG. 24 is a diagram showing a flow of encoding processing according to an embodiment of the present invention.

FIG. 25 is a diagram showing a flow of a grid reading / decoding process for reading digital information recorded by the encoding process.

FIG. 26 is a diagram showing a mapping table of recording data (14 bits) and coded bits (16 bits).

FIG. 27 shows a part of image data read from an optical record carrier on which digital information is recorded by performing random number processing.
It is a figure which shows the state divided | segmented by the square block of * 4.

FIG. 28 is a diagram showing an example of a pattern used for boundary identification.

FIG. 29 is a diagram showing another example of a pattern used for boundary identification.

30 is a diagram showing a state in which the boundary identification pattern shown in FIG. 28 is inserted into the image of FIG. 27.

FIG. 31 is a diagram showing a state in which the boundary identification pattern shown in FIG. 29 is inserted in the image of FIG. 27.

FIG. 32 shows an example in which a part of image data read from an optical record carrier on which digital information is recorded by performing random number processing is divided into 5 × 5 square blocks composed of odd-numbered cells. It is a figure.

FIG. 33 is a diagram showing an example of a pattern used for boundary identification.

FIG. 34 is a diagram showing another example of a pattern used for boundary identification.

FIG. 35 is a diagram illustrating an encoding process when the same pattern as the boundary identification pattern is generated by the random number process.

FIG. 36 is a diagram showing a flow of encoding processing according to an embodiment of the present invention.

FIG. 37 is a diagram showing the flow of encoding processing according to an embodiment of the present invention.

FIG. 38 is a diagram showing a flow of a grid reading / decoding process for reading digital information recorded by the encoding process.

FIG. 39 is a diagram showing a record carrier on which data to be recorded, the data, and an ink which acts on infrared rays or ultraviolet rays and an ink which shows the same color as the ink in the visible light region are recorded.

FIG. 40 is a diagram showing data to be recorded and a record carrier on which the data is recorded by a transparent ink which acts on infrared rays or ultraviolet rays.

FIG. 41 shows data to be recorded and a record carrier in which the data is recorded with an ink which acts on infrared rays or ultraviolet rays and is covered with a white hiding layer which transmits infrared rays or ultraviolet rays and blocks visible light. It is a figure.

FIG. 42 is a diagram showing an example in which invisible data which is invisible to the naked eye and can be recognized by irradiating infrared rays or ultraviolet rays, and visible data which has a format different from that of the invisible data and which is visible to the naked eye are overlapped and printed.

FIG. 43 is a diagram showing an example in which invisible data which is invisible to the naked eye and can be recognized by irradiating infrared rays or ultraviolet rays and visible data which has the same format as the invisible data and which is visible to the naked eye are overlapped and printed.

FIG. 44 is a diagram showing a main part of a reading device suitable for reading the record carrier of FIGS. 39 to 43.

[Explanation of symbols]

10 optical information recording device 11 CPU 12 ROM 13 RAM 14 external device 15 printing device 30 image on optical information recording carrier 40 optical information recording / reading device 41 CPU 42 RAM 43 ROM 44 image RAM 45 image sensor 46 control circuit 47 External device 110 Window 121 Image data start position mark 122 Clock mark 200, 210, 270, 280, 320, 330 Boundary identification pattern

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04N 1/409

Claims (7)

[Claims] 1. A flat recording surface is virtually formed with matrix-shaped cells corresponding to bits, and each of the cells is provided with an optically recognizable mark. A digital information decoding method for decoding the digital information from a record carrier on which digital information is recorded as a two-dimensional pattern, wherein a two-dimensional pattern representing the digital information is read as image data from a recording surface of the record carrier, The image data is virtually divided into rectangular blocks each consisting of a plurality of cells, and for each block of the above image data, a rectangular window surrounding this block is sequentially set and assigned to the cells within the blocks. A method for decoding digital information, characterized by identifying the formed mark and decoding the information represented by the mark to obtain the original digital information. 2. The digital information decoding method according to claim 1, wherein when the window is set in the first block of the image data, the size outside the first block is larger than the size of this block. Set the window,
Then, the edge of the window is moved inward in image bit units representing the image data, a change in the value of the image bit crossed by the edge of the window is detected, and the boundary position at which the value of the image bit changes is detected. A method for decoding digital information, characterized in that the above side of the window is stopped. 3. A pattern that a block can take by virtually setting a rectangular block composed of a plurality of cells on a plane and providing optically recognizable marks on each of the cells. In the group, the number of consecutive marks having the same value is k in at least one of the row direction and the column direction (where k is 1 or more, and the mesh forming the side of the block in the one direction). The following plural patterns are adopted as designated patterns, and each designated value of the above block unit is made to correspond to each data value forming the digital information to be recorded, and the above recorded Digital information decoding method for decoding the digital information from a record carrier in which the specified pattern of the block unit is arranged on the recording surface as a two-dimensional pattern In the method, a pattern that is not adopted as the designated pattern is preset as an invalid pattern in a pattern group that can be taken by the blocks by adding marks to each of the cells, and recording on the record carrier is performed. A two-dimensional pattern representing the digital information is read from the surface as image data, the image data is virtually divided into the block units, and a rectangular window surrounding the block is sequentially set for each block of the image data. Then, the mark provided in the grid in each block is identified to determine whether the pattern is a designated pattern or an invalid pattern.When the pattern of the block is the designated pattern, the information represented by the mark is decrypted to restore the original pattern. Correct the reading error when the pattern of the above block is an invalid pattern while obtaining digital information. Digital information decoding method and performing management. 4. A pattern that a block can take by virtually setting a rectangular block composed of a plurality of cells on a plane and providing optically recognizable marks to each of the cells. In the group, the number of consecutive marks having the same value is k in at least one of the row direction and the column direction (where k is 1 or more, and the mesh forming the side of the block in the one direction). The following plural patterns are adopted as designated patterns, and each designated value of the above block unit is made to correspond to each data value forming the digital information to be recorded, and the above recorded In a digital information recording method for expressing digital information as a two-dimensional pattern by arranging the specified pattern in block units on a recording surface, a pattern that the block can take Among them, a specific pattern other than the specified pattern is adopted as a boundary identification pattern for identifying the boundary of the block, and when the specified pattern of the block unit is arranged on the recording surface, it should be arranged with the already arranged block. Judging whether the boundary with the block to be identified is easy based on the pattern of the meshes that sandwich the boundary, and when the boundary is easy to identify, the block is arranged as it is,
A digital information recording method comprising arranging a block having the boundary designating pattern instead of the block when the boundary is not easily identified. 5. A rectangular block composed of a plurality of cells is virtually set on a plane, and the digital information to be recorded is encoded by a random number process in units corresponding to the block to form the block. By creating an encoding pattern in which optically recognizable marks are provided in each of the cells in a substantially uniform distribution, the digital information to be recorded is arranged on the recording surface by the encoding pattern of the block unit. In the digital information recording method of expressing as a three-dimensional pattern, a pattern in which two kinds of marks representing different values are alternately provided from the pattern group that the block can take is used for boundary identification for identifying a block boundary. When adopting as a pattern and arranging the above-mentioned block-unit coding pattern on the recording surface, Whether or not the boundary with the intended block is easy to identify is determined based on the pattern of the grids that sandwich the boundary, and when the boundary is easy to identify, the block is arranged as it is while the boundary is easy to identify. If not, a block having the boundary designating pattern is arranged in place of the block, and the digital information recording method is characterized. 6. The digital information recording method according to claim 5, wherein when the same coded pattern as the boundary identification pattern is generated by the random number processing, the coded patterns in block units are arranged on the recording surface. At this time, a plurality of blocks having the same coding pattern as the boundary identification pattern are continuously arranged so as to distinguish the coding pattern corresponding to the digital information to be recorded from the original boundary identification pattern. A method for recording digital information, characterized in that 7. The digital information recording method according to any one of claims 4 to 6, wherein it is determined whether or not a boundary between the block already arranged and a block to be arranged is easy to identify. A digital information recording method, which is performed based on the number of mesh pairs having different values among the mesh pairs adjacent to each other across the boundary.