JP6952846B2 - Two-dimensional code - Google Patents

Description

The present invention relates to a two-dimensional code, a method for analyzing a two-dimensional code, a device for analyzing a two-dimensional code, and a program for analyzing a two-dimensional code.

Compared with the one-dimensional code, the two-dimensional code can have a lot of information in a small area, and is widely used for various purposes such as article management and web guidance using a mobile phone. Since the two-dimensional code is photographed at various angles, the size or orientation of the two-dimensional code image is not constant on the screen, and image distortion and image blurring may occur. Further, when the two-dimensional code itself is dirty, a situation may occur in which some images cannot be discriminated.

The matrix type two-dimensional code generally has a substantially square shape, and after the photographed two-dimensional code image is coordinate-converted to the original shape, analysis (decoding) processing is performed based on the pixel value of each cell. It is said. The two-dimensional code has a position detection pattern having a predetermined shape in order to detect a position in the captured image. At the time of analysis, the position detection pattern is detected from the captured two-dimensional code image, and the coordinate conversion of the two-dimensional code image is performed based on the positional relationship of the position detection pattern. It is corrected and an image of the original 2D code is created. The extraction of the message contained in the two-dimensional code is performed based on the image after the coordinate conversion. As described above, the detection of the position detection pattern is a process that is the basis of the analysis and has a large influence on the recognition accuracy and the analysis time. Therefore, the detection process of the position detection pattern is an important element of the analysis process.

In addition, there is a need to recognize a plurality of two-dimensional codes at the same time. For example, multiple cardboard boxes with two-dimensional codes attached to the shelves are arranged in the same direction, and when stocking these cardboard boxes, the two-dimensional codes are collectively recognized from a distance rather than being brought close to the cardboard boxes and recognized one by one. If multiple two-dimensional codes can be recognized, work efficiency will be improved. It is known that batch recognition of a plurality of articles is performed using, for example, RFID, but since RFID requires an antenna or the like on the media side, the cost of the media becomes high. If a two-dimensional code can be used for such batch recognition, the two-dimensional code can be used simply by printing it on paper, for example, resulting in low cost.

As described above, the shape of the two-dimensional code is generally almost square, but when the two-dimensional code is printed on a narrow part of an article, or as wide as the spine of a book. When affixing to a narrow part, it is desirable to use a two-dimensional code with an elongated shape.

Japanese Unexamined Patent Publication No. 2013-30184

For example, Patent Document 1 proposes a two-dimensional code having an elongated shape.

However, the elongated two-dimensional code described in Patent Document 1 has only one position detection pattern.

In a two-dimensional code having only one position detection pattern, it is necessary to clearly separate the position detection pattern from the noise information reflected in the image in the captured image of the two-dimensional code. In particular, if the only position detection pattern cannot be identified due to dirt or image blurring, there is a risk that the two-dimensional code cannot be analyzed.

For example, in the above-mentioned inventory example in which a plurality of two-dimensional codes are photographed in one image, it is difficult to focus the photography on all the two-dimensional codes, and blurring is likely to occur. Generally, it is difficult to correct the image of a two-dimensional code that has blur / blur, so there is a limit to the improvement with software, and the blur / blur is solved by improving the hardware such as the autofocus function and increasing the shutter speed. There is a problem that it is necessary to do so, which increases the cost. Recently, there is a need to recognize a two-dimensional code with a camera installed in a mobile phone, but in the analysis of a two-dimensional code such as tilting of the camera at the time of shooting, blurring, and blurring caused by insufficient focus. Is in a problem-prone shooting environment, and it is often difficult to improve mobile phone cameras, etc. due to cost and size.

A general two-dimensional code has an error correction function, but this is a function for correcting a message error of the two-dimensional code, not a function for correcting a position detection pattern, so that the position detection pattern is dirty. When a chip occurs, it becomes difficult to recognize the two-dimensional code.

An object of the present specification is to provide a two-dimensional code that can solve the above-mentioned problems.

Another object of the present specification is to provide a method for analyzing a two-dimensional code that can solve the above-mentioned problems.

Another object of the present specification is to provide a two-dimensional code analysis device capable of solving the above-mentioned problems.

Further, it is an object of the present specification to provide a program for analyzing a two-dimensional code that can solve the above-mentioned problems.

According to the two-dimensional code disclosed herein, a plurality of cells arranged as a two-dimensional matrix-like pattern and representing data represented by the binary code, and two or more cells that can be independently detected from an image. Each of the above position detection patterns is provided with different position detection patterns, and the outer shape of each of the above position detection patterns is a quadrangle. Each of all the other position detection patterns is arranged so that at least a part of them overlap.

Further, according to other two-dimensional codes disclosed in the present specification, they are arranged as a two-dimensional matrix-like pattern, and can be detected independently from an image with a plurality of cells representing data represented by the binary code. With two or more different position detection patterns, all the other inside a pair of extension lines extending parallel to a predetermined direction from a pair of opposing points on the contour of one of the above position detection patterns. Each of the position detection patterns is arranged so as to overlap at least a part thereof, and all the cells are arranged inside the pair of the extension lines.

Further, according to the two-dimensional code analysis method disclosed in the present specification, at least three of the above-mentioned position detection patterns are based on the captured two-dimensional code image in the above-mentioned two-dimensional code analysis method. When is identified, at least one of the identified position detection patterns is selected from the identified position detection patterns, and the coordinates of a total of four points are acquired from the selected position detection patterns. Then, the coordinates of the image of the two-dimensional code are converted based on the acquired coordinates of the four points, and when the two above-mentioned position detection patterns are identified based on the captured image of the two-dimensional code. , At least one of the two identified position detection patterns is selected, and the coordinates of a total of four points are acquired from the selected position detection pattern, and the acquired four points are obtained. When the coordinates of the image of the two-dimensional code are converted based on the coordinates and one of the above position detection patterns is identified based on the captured image of the two-dimensional code, the identified one of the above The coordinates of four points are acquired from the position detection pattern, and the coordinates of the image of the two-dimensional code are converted based on the acquired coordinates of the four points.

Further, according to the two-dimensional code analysis device disclosed in the present specification, the above-mentioned two-dimensional code analysis device is the above-mentioned two-dimensional code analysis device, and at least three of the above-mentioned position detection patterns are based on the captured image of the two-dimensional code. When is identified, at least one of the identified position detection patterns is selected from the identified position detection patterns, and the coordinates of a total of four points are acquired from the selected position detection patterns. Then, the coordinates of the image of the two-dimensional code are converted based on the acquired coordinates of the four points, and when the two above-mentioned position detection patterns are identified based on the captured image of the two-dimensional code. , At least one of the two identified position detection patterns is selected, and the coordinates of a total of four points are acquired from the selected position detection pattern, and the acquired four points are obtained. When the coordinates of the image of the two-dimensional code are converted based on the coordinates and one of the above position detection patterns is identified based on the captured image of the two-dimensional code, the identified one of the above The coordinates of four points are acquired from the position detection pattern, and the coordinates of the image of the two-dimensional code are converted based on the acquired coordinates of the four points.

Further, according to the program for analyzing the two-dimensional code disclosed in the present specification, the program causes the computer to execute the above-mentioned analysis of the two-dimensional code, and is at least three based on the captured image of the two-dimensional code. When the above-mentioned position detection patterns are identified, at least one of the above-mentioned position detection patterns identified is selected from the above-mentioned position detection patterns, and the total is totaled from the selected above-mentioned position detection patterns. The coordinates of the four points are acquired, the coordinates of the image of the two-dimensional code are converted based on the acquired coordinates of the four points, and the two above-mentioned position detection patterns are generated based on the captured image of the two-dimensional code. When identified, at least one of the two identified position detection patterns is selected, and the coordinates of a total of four points are acquired from the selected position detection pattern. The coordinates of the image of the two-dimensional code are converted based on the acquired coordinates of the four points, and when one of the above position detection patterns is identified based on the captured image of the two-dimensional code, it is detected. The coordinates of four points are acquired from the one said position detection pattern, and the computer is made to execute the coordinate conversion of the image of the two-dimensional code based on the acquired coordinates of the four points.

According to the two-dimensional code disclosed in the present specification described above, the two-dimensional code can be arranged in a narrow part such as an article, and when a plurality of two-dimensional codes are photographed at the same time, or in an environment where blurring / blurring is likely to occur. Even when shooting under various shooting conditions such as, it is possible to accurately recognize the two-dimensional code in a short time.

Further, according to the other two-dimensional code disclosed in the present specification described above, the two-dimensional code can be arranged in a narrow part such as an article, and when a plurality of two-dimensional codes are photographed at the same time, or blurring / blurring is likely to occur. Accurate recognition of the two-dimensional code can be performed in a short time even when shooting under various shooting conditions such as shooting in an environment.

Further, according to the above-described two-dimensional code analysis method disclosed in the present specification, the above-mentioned two-dimensional code can be analyzed.

Further, according to the above-mentioned two-dimensional code analysis device disclosed in the present specification, the above-mentioned two-dimensional code can be analyzed.

Further, according to the above-mentioned program for analyzing the two-dimensional code disclosed in the present specification, the above-mentioned two-dimensional code can be analyzed.

It is a figure which shows the 2D code of 1st Embodiment disclosed in this specification. It is a figure explaining the data part of a two-dimensional code. (A) to (D) are diagrams for explaining an example of the relationship between a pair of extension lines and another two-dimensional code. (A) to (L) are diagrams showing various forms of position detection patterns. (A) and (B) are diagrams showing other forms of the position detection pattern. (A) to (E) are diagrams showing still another form of the position detection pattern. (A) to (F) are diagrams showing still another form of the position detection pattern. It is a figure which shows the modification 1 of the 2D code of 1st Embodiment disclosed in this specification. It is a figure which shows the modification 2 of the 2D code of 1st Embodiment disclosed in this specification. It is a figure which shows the card which printed the 2D code of 1st Embodiment disclosed in this specification. It is a figure which shows the book to which the modification 2 of the 2D code of 1st Embodiment disclosed in this specification is attached. (A) is a diagram showing a two-dimensional code of the second embodiment disclosed in the present specification, and (B) is a diagram showing a modified example of the two-dimensional code of the second embodiment. It is a flowchart explaining the method of creating a 2D code disclosed in this specification. It is a figure explaining the 2D code analysis apparatus disclosed in this specification. It is a flowchart (the 1) explaining the analysis method of the two-dimensional code disclosed in this specification. It is a flowchart (the 2) explaining the analysis method of the two-dimensional code disclosed in this specification. It is a flowchart (the 3) explaining the analysis method of the two-dimensional code disclosed in this specification. It is a figure explaining the shape check of the position detection pattern candidate. It is a figure explaining the shape check of the position detection correction pattern. It is a flowchart (the 4) explaining the analysis method of the two-dimensional code disclosed in this specification. It is a figure explaining that the coordinates of a total of 4 points are acquired from 3 detected position detection pattern candidates. FIGS. (A) to (C) are diagrams for explaining that the coordinates of a total of four points are acquired from the two detected position detection pattern candidates. FIGS. (A) to (C) are diagrams for explaining that the coordinates of a total of four points are acquired from one detected position detection pattern candidate.

Hereinafter, a preferred first embodiment of the two-dimensional code disclosed in the present specification will be described with reference to the drawings. However, the technical scope of the present invention is not limited to those embodiments, but extends to the inventions described in the claims and their equivalents.

FIG. 1 is a diagram showing a two-dimensional code of the first embodiment disclosed in the present specification.

As shown in FIG. 1, the two-dimensional code 10 of the present embodiment has a shape extending in a predetermined direction, and in the example shown in FIG. 1, it has a horizontally long shape.

The two-dimensional code 10 includes a plurality of position detection patterns 12A, 12B, 12C used for detecting the position of the two-dimensional code in the image in which the two-dimensional code is taken, and a data unit 11 having information such as a message. Be prepared.

The two-dimensional code 10 includes two or more different position detection patterns that can be detected independently from an image. In the present embodiment, the two-dimensional code 10 includes three different position detection patterns 12A, 12B, and 12C. The two-dimensional code 10 may include four or more position detection patterns.

Specifically, the two-dimensional code 10 is arranged at the first position detection pattern 12A arranged at the left end in the longitudinal direction, the second position detection pattern 12B arranged at the central portion in the longitudinal direction, and the right end in the longitudinal direction. The third position detection pattern 12C is provided.

A data unit 11 is arranged between the first position detection pattern 12A and the second position detection pattern 12B. Further, the data unit 11 is also arranged between the second position detection pattern 12B and the third position detection pattern 12C. The data unit 11 is arranged from the inside to the outside of the pair of extension lines L, but the area where the cells constituting the data unit 11 are arranged outside the pair of extension lines L is a pair of extension lines. It is preferably within 50%, particularly within 20%, and further within 10% of the width of L from the viewpoint of narrowing the width of the two-dimensional code 10.

The outer shape of each position detection pattern 12A, 12B, 12C is a quadrangle, and the pair of opposite sides of one position detection pattern 12A are all inside the pair of extension lines L that are virtually extended in the extending direction of the sides. The other two position detection patterns 12B and 12C are arranged so that at least a part of them overlap.

In the example of the two-dimensional code shown in FIG. 1, the second position detection pattern 12B is arranged so that the pair of opposite sides overlap with the pair of extension lines L, and the entire second position detection pattern 12B is arranged. , Are arranged so as to overlap the inside of the pair of extension lines L.

Further, the third position detection pattern 12C is also arranged so that the pair of opposite sides overlap with the pair of extension lines L, and the entire third position detection pattern 12C is inside the pair of extension lines L. They are arranged so that they overlap.

In the above description, as a pair of extension lines, a line in which the pair of opposite sides of the first position detection pattern 12A are virtually extended in the extending direction of the sides is used with the other position detection patterns 12B and 12C. Although the relationship has been described, all the other position detection patterns are at least one inside the pair of extension lines L in which the pair of opposite sides of one position detection pattern are virtually extended in the extending direction of the sides. It suffices if the parts are arranged so as to overlap each other. The pair of extension lines may be lines in which a pair of opposite sides of the second position detection pattern 12B or the third position detection pattern 12C are extended.

As described above, the position detection patterns 12A, 12B, and 12C are used for detecting and analyzing the position of the two-dimensional code 10 in the image in which the two-dimensional code 10 is captured. Specifically, first, the position detection patterns 12A, 12B, and 12C are identified from the captured image. Next, four coordinates are acquired based on the identified position detection patterns 12A, 12B, 12C. Next, based on the four acquired coordinates, the coefficient of the projective transformation in the two-dimensional space is obtained, and the coordinate transformation of the image of the two-dimensional code is performed by the projective transformation, and the influence of image distortion or image blurring is performed. Is corrected. Next, information such as a message is taken out based on the cell of the data unit 11 of the image of the corrected two-dimensional code. The details of this process will be described later.

As shown in FIG. 2, the data unit 11 is arranged as a two-dimensional matrix-like pattern, and has a plurality of cells 13 representing data in which information such as a message is represented by a binary code. Cell 13 represents binary information depending on the difference in brightness between light and dark.

As shown in FIG. 2, the data unit 11 is divided into a plurality of blocks 14 having the same size, and has a light or dark separation space 15 between adjacent blocks 14. One block 14 is formed by a plurality of cells 13.

In the example shown in FIG. 2, the block 14 is formed by 3 × 3 cells, and a bright separation space 15 for one cell is arranged between adjacent blocks 14. The separation spaces 15 are arranged in a grid pattern so as to separate each block 14. As a result, the accuracy of determining the brightness of the cell in each block 14 is improved. The number or shape of cells forming the block is not limited to the example shown in FIG. Further, the shape of the separation space is not limited to the example shown in FIG.

The actual data included in the data unit 11 has a plurality of segments corresponding to the number of messages, a segment containing only the termination flag, and a filling grass arranged in the remaining area of the actual data. A plurality of segments are segments in which a message and a header having a header (message type (message encoding), message size, etc.) that qualifies the message are arranged as many as the number of messages. In the data unit 11, the actual data is divided into blocks according to the data capacity per block.

Further, the data unit 11 has an error correction code for correcting an error of the data unit 11 itself at the time of decoding. The error correction code is also divided into blocks.

For example, when a Reed-Solomon code is used as the error correction code, error correction is performed in word units, so it is preferable that one word is one block. When one word spans a plurality of blocks, even if one block becomes dirty, all the words related to the block are subject to error correction, and the efficiency of correction becomes poor. Dirt that causes correction and color skipping due to spotlights are often concentrated in one place, but making blocks has the effect of consolidating the data to be corrected in one place at the same time, so efficient correction can be performed. Enable and increase the chances of your code being recognized.

Further, the data unit 11 includes position correction patterns 16A and 16B arranged at predetermined positions with respect to the position detection patterns 12A, 12B and 12C. The position correction pattern 16A is arranged at a predetermined position with respect to each of the first position detection pattern 12A and the second position detection pattern 12B. The position correction pattern 16B is arranged at a predetermined position with respect to the second position detection pattern 12B and the third position detection pattern 12C, respectively. The position correction pattern may be related to any one of the position detection patterns or to a plurality of position detection patterns.

The first role of the position correction patterns 16A and 16B is detected by identifying the position correction pattern arranged at a predetermined position with respect to the position detection pattern candidate detected from the image obtained by capturing the two-dimensional code. It is that the position detection pattern candidate is identified as the position detection pattern. For example, when only one position detection pattern candidate is detected from an image obtained by capturing a two-dimensional code, the simpler the shape of the position detection pattern, the higher the possibility of noise in the captured image. By identifying the corresponding position correction pattern, it is identified that the detected position detection pattern candidate is the position detection pattern. Therefore, four coordinates are acquired based on one identified position detection pattern. can do. The details of this process will be described later.

Further, the second role of the position correction patterns 16A and 16B is to correct the error of the coordinate of the cell of the data unit 11. An error may occur in the coordinates of the cell of the data unit 11 due to the deviation when acquiring the position of the position detection pattern from the image obtained by capturing the two-dimensional code. Therefore, the error of the coordinate of the cell of the data unit 11 is corrected based on the positional relationship between the position correction patterns 16A and 16B and the position detection pattern associated with the position correction patterns 16A and 16B. The number or shape of cells forming the position correction pattern is not limited to the example shown in FIG.

Further, the data unit 11 may have a version information block, a format information block, a picture embedding information block, and the like.

The version information block has information about the size of the two-dimensional code 10. When analyzing the two-dimensional code 10, the area of the data unit 11 can be identified based on the information contained in the version information block. For example, a vertical version information block having information on the vertical size of the two-dimensional code and a horizontal version information block having information on the horizontal size of the two-dimensional code can be used. By using the version information block, the size of the data unit 11 can be made variable to control the amount of information included in the two-dimensional code 10.

The format information block has the amount of data that one block 14 has as information. By using the format information block, for example, the amount of data contained in one block 14 can be selected as 6, 7, 9 bits / block.

The picture embedding information block is used to define an area in which a picture or the like is arranged and cells are not arranged when the two-dimensional code 10 has a picture. The picture embedding information block can have, for example, information on the size and block position of an area (area in which cells are not arranged) in which a picture or the like is embedded. In the area where the cells are not arranged, not only the pattern but also a blank or another one-dimensional code or two-dimensional code can be arranged.

The fact that the position detection pattern can be detected independently from the image means that each position detection pattern can be detected by analyzing the image based on the image obtained by taking the two-dimensional code.

Further, the difference in the position detection pattern includes, for example, a difference in the outer shape, the size, the arrangement direction on the two-dimensional code, and the ratio of the line width forming the position detection pattern. Further, the difference in the line width ratio includes, for example, that the light-dark ratio is different when scanning so as to pass through the center of the position detection pattern.

Further, the difference in the position detection pattern includes that the internal shape is different even if the outer shape is the same. Furthermore, the fact that the position detection patterns are different means that at least two position detection patterns do not match in shape even when rotated, or the outer shapes do not match when rotated, or the outer shapes match when rotated. However, the ratio of the line widths of the lines forming the position detection pattern is different.

The reason why the two-dimensional code 10 includes at least two or more position detection patterns will be described below. In the analysis of the two-dimensional code 10, it is preferable to accurately correct the influence of image distortion or image blur to obtain an image of the two-dimensional code that is less affected by distortion or blur. From this point of view, the coordinates of at least 4 points are acquired based on the position detection pattern, and the coefficients of the projective transformation in the 2D space are obtained based on the acquired coordinates of the 4 points. It is preferable to perform coordinate conversion. It is desirable that the coordinates of the four points for performing the projective transformation are close to the data unit 11 to be calculated for data acquisition. When the coordinates of 4 points are acquired from one position detection pattern, the calculation accuracy of the coordinates in the data unit 11 away from this position detection pattern deteriorates. Therefore, it is necessary to acquire 4 points that can cover the entire data unit 11 widely. Is desirable. Further, when the code captured is distorted due to the distortion of the lens of the reader or the distortion of the surface on which the two-dimensional code is printed, the distortion is not uniform. Therefore, it is more distorted to obtain the projective transformation coefficient based on a plurality of points spanning different position detection patterns than to acquire the projective transformation coefficient in two-dimensional space from the coordinates of four points acquired from the same position detection pattern. You can expect the effect of reducing the influence of.

Next, the reason why the two-dimensional code 10 has different position detection patterns will be described below. When a plurality of two-dimensional codes are captured in one image and the plurality of position detection patterns of each two-dimensional code are different, the number of combinations of position detection patterns to be examined can be significantly reduced. It is possible to reduce the amount of processing for determining whether the combination is a regular combination. It will be described below that this processing amount is reduced.

Regarding the case where a plurality of two-dimensional codes are imprinted on one captured image, the case where the two-dimensional codes include a plurality of different position detection patterns and the case where the two-dimensional codes include a plurality of the same position detection patterns. , The processing amount for determining that the combination of the position detection patterns is a regular combination is compared below.

For example, it is assumed that six two-dimensional codes are photographed in one image, and each two-dimensional code has three same position detection patterns. In this case, 18 (6 × 3) same position detection patterns are imprinted on the image. Detecting the 18 position detection pattern, when examined by either of the same two-dimensional code for three combinations of all of them, the combination becomes ways _{18 C 3} = _816. If the the single two-dimensional code has four identical position detection pattern, the number of the position detection pattern of the screen becomes 24, the combination is a street _{24 C} 4 ₌ 10624. In this way, the number of combinations of position detection patterns becomes enormous, and the processing time increases.

On the other hand, when the three position detection patterns are different as in the two-dimensional code shown in FIG. 1, the number of combinations of the three different position detection patterns is ₆ C ₁ × ₆ C ₁ × ₆ C ₁ = 216. With four different position detection patterns, there are ₆ C ₁ x ₆ C ₁ x ₆ C ₁ x ₆ C ₁ = 1296, and the number of combinations of position detection patterns is significantly reduced as compared with the above case. Therefore, the amount of processing for determining a regular combination of position detection patterns included in the same two-dimensional code can be greatly reduced.

Further, when the position detection patterns are different, the direction and position of the other position detection patterns can be easily predicted when one position detection pattern is detected, so that the amount of processing for searching the position detection pattern is reduced. be able to. When the two-dimensional code includes a position correction pattern, the position correction pattern may be used when searching for another position detection pattern.

The three position detection patterns 12A, 12B, and 12C preferably have a larger area than the block 14. As a result, the same pattern as the position detection pattern does not appear in the two-dimensional code, and the position detection pattern can be easily detected.

Further, the position detection patterns 12A, 12B, and 12C preferably have a frame shape provided with an identification space. This identification space is preferably a space that is at least twice as large as the smallest cell that constitutes the two-dimensional code.

In the two-dimensional code 10, the first position detection pattern 12A has a single square frame. The second position detection pattern 12B has a single square frame and squares arranged in the square frame. The third position detection pattern 12C has a single rectangular frame. Each position detection pattern 12A, 12B, 12C has a bright identification space which is at least twice the space of the minimum cell.

Since the position detection patterns 12A, 12B, and 12C have an identification space inside, the position detection pattern and other reflections can be easily distinguished, so that the influence of blurring / blurring is suppressed and the identification is improved. .. In this case, the identification space is preferably two or more cells. The reason is that in a two-dimensional code in which the smallest cell represents one bit, if blurring / blurring exceeding one cell occurs, even if the position detection pattern can be detected, data cannot be acquired. If there is more than one cell of blur / blur, the code cannot be recognized, and if there are two identification spaces, it is possible to deal with blur / blur of one cell in all directions. This is because when shooting multiple 2D codes on one screen, each 2D code becomes smaller, so the effect of blurring and blurring becomes larger, and the image of the 2D code becomes unclear. Is also effective.

A bright pattern for one cell is arranged around each position detection pattern 12A, 12B, 12C. As a result, even when another figure is arranged adjacent to the two-dimensional code 10, each position detection pattern 12A, 12B, 12C can be easily detected from the image taken by the two-dimensional code 10. ..

The light and darkness forming the position detection patterns 12A, 12B, 12C and the data unit 11 of the two-dimensional code 10 may be reversed. For example, when the light and darkness of the first position detection pattern 12A in FIG. 1 is reversed, a bright square frame is arranged in the background of the black square.

Next, an example of the relationship between the pair of extension lines and the other two-dimensional code will be described with reference to FIG. 3 (A) to 3 (D) described below show that the pair of opposite sides of the first position detection pattern 12A are inside the pair of extension lines L that are virtually extended in the extending direction of the sides. An example is shown in the case where at least a part of the position detection patterns 12B and 12C of is overlapped.

In the example shown in FIG. 3A, the pair of opposite sides of the first position detection pattern 12A is one of a pair of extension lines L that are virtually extended in the extending direction of the sides, and the second position detection pattern. At least part of the contour of 12B overlaps. Specifically, one side of the square forming the outer shape of the second position detection pattern 12B overlaps with the extension line L. In the second position detection pattern 12B, the portion of the second position detection pattern 12B except for one side of the square forming the outer shape does not overlap inside the pair of extension lines L. As described above, the fact that at least a part of the second position detection pattern 12B overlaps inside the pair of extension lines L includes that at least a part of the contour of the second position detection pattern 12B overlaps.

In the example shown in FIG. 3B, similarly to FIG. 1, the second position detection pattern 12B is arranged so that the pair of opposite sides overlap with the pair of extension lines L, and the second position detection The entire pattern 12B is arranged so as to overlap the inside of the pair of extension lines L. The third position detection pattern 12C (having a shape different from that in FIG. 1) is arranged so that one of the pair of opposite sides overlaps with one of the pair of extension lines L, and the third position is detected. The entire position detection pattern 12C is arranged so as to overlap the inside of the pair of extension lines L.

In the example shown in FIG. 3C, in the second position detection pattern 12B, a part (about half of the area) of the second position detection pattern 12B is arranged so as to overlap the inside of the pair of extension lines L. ..

In the example shown in FIG. 3D, the second position detection pattern 12B is arranged so that the contour of the second position detection pattern 12B is included inside the pair of extension lines L.

As shown in FIGS. 3 (A) to 3 (D) described above, a pair of opposite sides of one position detection pattern 12A are placed inside a pair of extension lines L that are virtually extended in a direction in which the sides extend. Since the other position detection patterns 12B and 12C are arranged so that at least a part of them overlap each other, the processing time for searching the other position detection patterns 12B and 12C based on one position detection pattern 12A is short. It is made to be. In particular, as shown in FIG. 3D, when the contour of the second position detection pattern 12B is arranged inside the pair of extension lines L, the search area becomes smaller than in other examples. The processing time for searching can be further shortened. A detailed explanation will be described later.

In the examples of FIGS. 1 and 3 (A) to 3 (D) described above, a pair of opposite sides of one position detection pattern 12A are virtually extended in a direction in which the sides extend, and a pair of extension lines L. , The relationship with the other position detection patterns 12B and 12C has been described. However, in the present specification, if the outer shape of each position detection pattern is a quadrangle, the pair of opposite sides of one position detection pattern can be virtually set. It suffices that all the other position detection patterns are arranged so that at least a part of them overlaps inside the pair of extension lines extending in the extending direction of the sides.

Further, in the present invention, the outer shape of each position detection pattern may have a shape other than a quadrangle. In this case, each of the other position detection patterns has at least one inside a pair of extension lines in which a pair of opposing points on the contour of one position detection pattern are virtually extended in parallel in a predetermined direction. The portions are arranged so as to overlap, and all cells are arranged inside a pair of extension lines.

Next, various forms of the position detection pattern will be described below with reference to the drawings.

4 (A) to 4 (L) are views showing various forms of a position detection pattern having a rectangular outer shape.

In FIGS. 4 (A) and 4 (B), the position detection pattern has a single square frame. The position detection pattern shown in FIG. 4 (A) is larger than the position detection pattern shown in FIG. 4 (B). In addition, the ratio of the line widths of the lines forming the position detection pattern is also different.

In FIGS. 4C and 4D, the position detection pattern has a vertically long single rectangular frame. The position detection pattern shown in FIG. 4C is wider than the position detection pattern shown in FIG. 4D. In addition, the ratio of the line widths of the lines forming the position detection pattern is also different.

In FIGS. 4 (E) to 4 (G), the position detection pattern has a horizontally long single rectangular frame. The position detection pattern shown in FIG. 4 (E) is larger than the position detection pattern shown in FIG. 4 (F). The position detection pattern shown in FIG. 4 (G) is higher in height than the position detection pattern shown in FIG. 4 (E), but is narrower in width. In addition, the ratio of the line widths of the lines forming the position detection pattern is also different.

In FIGS. 4H and 4I, the position detection pattern has a single square frame and squares arranged within the square frame. The position detection pattern shown in FIG. 4 (H) is larger than the position detection pattern shown in FIG. 4 (I). In addition, the ratio of the line widths of the lines forming the position detection pattern is also different.

In FIGS. 4 (J) to 4 (L), the position detection pattern is formed by a dark pattern. The position detection pattern shown in FIG. 4 (J) is a square. The position detection pattern shown in FIG. 4 (K) is a vertically long rectangle. The position detection pattern shown in FIG. 4 (L) is a horizontally long rectangle.

5 (A) and 5 (B) are diagrams showing other forms of the position detection pattern having a square outer shape.

The position detection pattern shown in FIG. 5A has a double square frame. The position detection pattern shown in FIG. 5B has a triple square frame. The position detection pattern shown in FIG. 5A may be composed of a double vertically long or horizontally long rectangular frame. Further, the position detection pattern shown in FIG. 5B may be composed of a triple vertically long or horizontally long rectangular frame. Further, in the position detection pattern shown in FIGS. 5A and 5B, only the square pattern inside the position detection pattern may be composed of a vertically or horizontally long rectangular frame, or a frame having four or more layers. May have.

6 (A) to 6 (E) are views showing still another form of the position detection pattern having a square outer shape.

The position detection pattern shown in FIG. 6A has a single square frame and a vertically long rectangular light pattern and dark pattern arranged in the square frame.

The position detection pattern shown in FIG. 6B has a single square frame and a light pattern and a dark pattern of triangles arranged in the square frame.

The position detection pattern shown in FIG. 6C has a single square frame and horizontally long bright patterns and dark patterns arranged in the square frame alternately in the vertical direction.

The position detection pattern shown in FIG. 6D has a single square frame and vertically long bright patterns and dark patterns arranged in the square frame alternately in the horizontal direction.

The position detection pattern shown in FIG. 6E has a single square frame, four square bright patterns arranged in the square frame, and a dark pattern that separates the bright patterns of each square. The position detection pattern shown in FIG. 6 may be a vertically long or horizontally long rectangle. Further, the ratio of the line widths forming the position detection pattern may be appropriately changed.

7 (A) to 7 (E) are views showing the form of a position detection pattern having a shape other than a quadrangle in outer shape.

The position detection pattern shown in FIG. 7A has a single circular or oval frame. The four position detection patterns are different.

The position detection pattern shown in FIG. 7B has a single circular or elliptical frame and a circular or elliptical frame arranged within the frame. The four position detection patterns are different.

The position detection pattern shown in FIG. 7C has a first rectangle and a second rectangle, and the first rectangle and the second rectangle intersect each other so as to be orthogonal to each other at the position of the center of gravity. The four position detection patterns are different.

The position detection pattern shown in FIG. 7D has a first rectangle and a second rectangle, and the second rectangle extends vertically from the end of the first rectangle. The four position detection patterns are different.

The position detection pattern shown in FIG. 7 (E) has a single right-angled triangular frame. The four position detection patterns are different. Further, for example, as the three position detection patterns constituting one two-dimensional code, those shown in FIGS. 4 (A), 5 (A), and 6 (A) are used, and the like. The position detection patterns shown in 6 and 7 may be combined.

Next, the modification 1 and modification 2 of the two-dimensional code of the first embodiment described above will be described below with reference to the drawings.

FIG. 8 is a diagram showing a modification 1 of the two-dimensional code of the first embodiment disclosed in the present specification.

The two-dimensional code 10 of the modification 1 is formed by extending a pair of opposing points P on the contour of the rectangular first position detection pattern 12A virtually in parallel in a predetermined direction inside a pair of extension lines L. All the other position detection patterns 12B and 12C are arranged so as to overlap at least a part thereof, and all the cells are arranged inside the pair of extension lines L.

As long as the pair of points P facing each other on the contour and the points on the contour of the first position detection pattern 12A of the quadrangle are points on the contour, they may be points on the sides or vertices.

Arranging the cells of the two-dimensional code 10 inside the pair of extension lines L includes arranging the contour of one cell so as to overlap one of the pair of extension lines L.

Since all the cells forming the data unit 11 are arranged inside the pair of extension lines L, the size of the two-dimensional code 10 is reduced.

The two-dimensional code 10 of the first modification has the outer shape of the quadrangle of the position detection patterns 12A, 12B, 12C, but the position detection patterns 12A, 12B, 12C may have the outer shape other than the quadrangle. .. For example, the position detection pattern may have an outer shape as shown in FIGS. 7 (A) to 7 (E).

Further, the two-dimensional code 10 of the modification 1 does not have a position correction pattern. By not arranging the position correction pattern in the data unit 11, more information can be arranged as cells in the data unit 11.

FIG. 9 is a diagram showing a modification 2 of the two-dimensional code of the first embodiment disclosed in the present specification.

The two-dimensional code 10 of the second modification has a vertically long shape. In the two-dimensional code 10 of the second modification, the pair of opposing sides extending in the vertical direction of the first position detection pattern 12A are placed inside the pair of extension lines L that are virtually extended in the extending direction of the sides. Each of the two position detection patterns 12B and 12C is arranged so that at least a part thereof overlaps with each other.

Next, an example in which the above-mentioned two-dimensional code is arranged on the article will be described below with reference to the drawings.

FIG. 10 is a diagram showing a card on which the two-dimensional code of the first embodiment disclosed in the present specification is printed.

The card 20 is, for example, a card-like medium having information that can be read by a game machine. On the card 20, a pattern 21 that characterizes the card 20 is arranged. Further, on the card 20, a two-dimensional code 10 is printed in a horizontally long area between the pattern 21 and the edge of the card 20. In the data section of the two-dimensional code 10, for example, identification information for identifying the card 20 is recorded. For the card 20, for example, the identification information of the card 20 possessed by the two-dimensional code 10 is read by the game machine, and the information of the card 20 is incorporated into the progress of the game.

FIG. 11 is a diagram showing a book to which a modification 2 of the two-dimensional code of the first embodiment disclosed in the present specification is attached.

A two-dimensional code 10 is attached to the spine 31 of the book 30. In the data section of the two-dimensional code 10, for example, a control number for managing the book and a title of the book are recorded as information. In a library or the like that manages a large number of books, the book 30 can be managed by using the two-dimensional code 10 attached to the spine 31 of the book 30.

According to the two-dimensional code of the present embodiment described above, it can be arranged in a narrow part such as an article, and when shooting a plurality of two-dimensional codes at the same time, or shooting in an environment where blurring and blurring are likely to occur, etc. Accurate recognition of the two-dimensional code can be performed in a short time even when shooting under various shooting conditions.

Next, a second embodiment of the above-mentioned two-dimensional code will be described below with reference to FIGS. 12 (A) and 12 (B). The detailed description of the first embodiment described above is appropriately applied to the points not particularly described with respect to the second embodiment. Further, the same components are designated by the same reference numerals.

FIG. 12A is a diagram showing a two-dimensional code of the second embodiment disclosed in the present specification.

The two-dimensional code 10 of the present embodiment includes a basic pattern portion 10A and a peripheral portion 10B arranged outside the basic pattern portion 10A. The basic pattern unit 10A includes three different position detection patterns 12A, 12B, 12C, and a data unit 11. The configuration of the basic pattern portion 10A is the same as the two-dimensional code of the first embodiment described above.

A data unit 11 is arranged in the peripheral unit 10B. The peripheral portion 10B is arranged outside both ends of the basic pattern portion 10A in the longitudinal direction.

The basic pattern unit 10A includes basic pattern unit movement information indicating the position of the basic pattern unit 10A in the two-dimensional code area. The basic pattern unit movement information includes basic pattern unit position information indicating the position of the basic pattern unit 10A in the two-dimensional code, or basic pattern unit movement amount information indicating the amount of movement of the basic pattern unit from a predetermined position. When the two-dimensional code 10 is analyzed, the region of the basic pattern portion 10A and the region of the peripheral portion 10B in the two-dimensional code 10 are identified based on the movement information of the basic pattern portion.

Further, the basic pattern unit 10A has version information indicating the size of the two-dimensional code 10, and the version information is used to determine the size of the two-dimensional code in a state where the distances between the plurality of position detection patterns are fixed. It is possible to design it variably.

According to the two-dimensional code of the present embodiment described above, the degree of freedom in arranging the position detection pattern is high, and when the two-dimensional code is arranged, if the position detection pattern is located in a difficult-to-read area, it is such a case. The position detection pattern can be arranged while avoiding areas that are difficult to read. As a result, the recognition accuracy of reading the two-dimensional code can be improved. Further, the size of the peripheral portion 10B can be variably designed according to the amount of data arranged in the data unit 11. Further, the same effect as that of the first embodiment described above is obtained.

FIG. 12B is a diagram showing a modified example of the two-dimensional code of the second embodiment.

In the two-dimensional code 10 of this modified example, the peripheral portion 10B is arranged around the basic pattern portion 10A on the top, bottom, left, and right. When the amount of data arranged in the data unit 11 is large, the data unit 11 may be arranged outside both ends of the basic pattern unit 10A in the longitudinal direction and outside the side edges in the longitudinal direction of the basic pattern unit 10A. good.

Next, a method of creating the two-dimensional code disclosed in the present specification (encoding method) described above will be described below with reference to FIG.

FIG. 13 is a flowchart illustrating a method of creating a two-dimensional code disclosed in the present specification.

The process shown in FIG. 13 can be realized, for example, by the computer executing a predetermined program stored in a storage device including a storage medium or the like.

First, in step S10, the main processing of encoding is started.

Next, in step S12, a message to be recorded in the two-dimensional code is input to the computer.

Next, in step S14, the computer determines the placement position of the position detection pattern in the code image. A computer has two or more different position detection patterns that can be independently detected from an image, and a position detection pattern in which the outer shape of each position detection pattern is a quadrangle, and a pair of opposite sides of one position detection pattern are used. Inside the pair of extension lines that are virtually extended in the extending direction of the sides, all the other position detection patterns are arranged in the code image so that at least a part of them overlap. Alternatively, the computer has two or more different position detection patterns that can be detected independently from the image, and a pair of extension lines that virtually extend a pair of opposite sides of one position detection pattern in a direction in which the sides extend. Inside, the code image may be arranged so that at least a part of each of the other position detection patterns overlaps. In this case, in step S24, which will be described later, all the cells are arranged inside the pair of extension lines.

Next, in step S16, the computer determines the placement position of the position correction pattern in the code image. The position correction pattern is arranged at a predetermined position with respect to the position detection pattern. If the position correction pattern is not arranged in the two-dimensional code, this step is omitted.

Next, in step S18, the computer adds a header to each message to generate segments, and creates actual data having a plurality of segments.

Next, in step S20, the computer creates an error correction code for the actual data. For example, the Reed-Solomon code may be used as the error correction code, and the error correction may be performed word by word.

Next, in step S22, the computer converts the actual data and the error correction code into a binary code to create light and dark cell information.

Next, in step S24, the computer arranges the position detection pattern, the position correction pattern, and each cell forming the data unit in the code image.

Next, in step S26, the computer outputs the code image as data to obtain a two-dimensional code.

When the two-dimensional code is printed, any medium may be used for printing the two-dimensional code, and examples thereof include paper, a resin plate, and a housing surface. The printing apparatus may be any as long as it can print a two-dimensional code on these media, and examples thereof include a simple printer, a precision printer, and a printing apparatus. The printing apparatus may be capable of color printing as well as monochrome printing. Further, the created two-dimensional code may be transmitted to the user as two-dimensional code data via a communication line without printing.

Next, an analysis device (decoding device) for analyzing the two-dimensional code disclosed in the present specification described above will be described below with reference to FIG.

The analysis device 40 includes a reading unit 41, a computer (two-dimensional code analysis processing unit) 45, a display 46, and a communication interface 47. The reading unit 41 has a lens 42, an image sensor 43, and an analog-to-digital converter (AD) 44, and outputs digital image data of a captured two-dimensional code to the computer 45. The computer 45 has a calculation unit that executes an analysis of a two-dimensional code by executing a predetermined program stored in a storage device including a storage medium or the like. The analysis device 40 shown in FIG. 14 can be configured by using, for example, a mobile terminal. The computer 45 may transmit the data obtained by analyzing the two-dimensional code to another server, a terminal, or the like (not shown) via a network (not shown) using the communication interface 47.

Next, an analysis method in which the above-mentioned analysis device 40 analyzes the two-dimensional code shown in FIG. 1 will be described below with reference to the drawings. 15 to 17 are flowcharts illustrating a method of analyzing the two-dimensional code disclosed in the present specification.

First, in step S30, the main process of analysis is started.

Next, in step S32, the computer 45 inputs an image of the two-dimensional code taken by the reading unit 41.

Next, in step S34, the computer 45 creates a binary image of the input captured image. In the binarization method, if the input captured image is a color image such as an RGB image, it is once converted to a grayscale image, the average of the maximum and minimum luminance values in the image is set as the threshold value, and the value is equal to or higher than the threshold value. If it is, it is bright, and if it is less than the threshold, it is dark. In order to convert a color image to grayscale, for example, using the RGB values of each pixel, the brightness converted according to the conversion formula of brightness (brightness) = 0.299R + 0.587G + 0.114B is treated as grayscale. Various methods have been proposed for converting a color image to a grayscale image and further converting to a binarized image, and the conversion method is not limited to the above conversion method.

Next, in step S36, the computer 45 detects the position detection pattern candidate. Specifically, the screen is scanned in the X direction starting from the upper left point of the image, and when the right end is reached, the screen is scanned in the X direction from the left end several pixels below. In other words, it will be subtracted for a few pixels. This is repeated for the entire image. The number of pixels to be thinned out is (maximum thinning amount) = (cell size that is the smallest among the side lengths of all position detection patterns) × (number of pixels per cell at the time of shooting). The minimum cell size in the side lengths of the position detection pattern is the minimum cell size in the Y direction that can occur in the rotation of the position detection pattern in consideration of the rotation of the cord. If the number of pixels per cell at the time of shooting is less than 1 pixel per cell, it becomes difficult to distinguish between the light and dark of the cell even in an ideal shot image. Considering the displacement or rotation of cells in the actual captured image, it is desirable that there are at least 3 pixels per cell.

Then, the computer 45 inspects the detected position detection pattern candidate. Specifically, when a light-dark boundary is found during scanning, the area around the dark portion is scanned. At that time, the minimum and maximum X and Y coordinates are obtained, respectively. As a result, the circumscribed rectangle of the dark mass can be obtained. The upper left coordinate of this is expressed as (x1, y1), and the lower right coordinate is expressed as (x2, y2).

FIG. 18 is a diagram for explaining the shape check of the position detection pattern candidate.

First, the computer 45 checks the shape of the position detection pattern candidate. Each position detection pattern of the two-dimensional code shown in FIG. 1 has a frame, that is, a hole shape. Therefore, first check for the presence of holes. Scanning is performed in the x-axis direction with the start point as the left midpoint (x1, (y1 + y2) / 2) of the circumscribed rectangle and the end point as (x2, (y1 + y2) / 2). Here, it is confirmed that the following conditions (1) or (2) are satisfied.

(1) Confirm that there is a line of darkness and darkness. Confirm that the central pixel from the left end (lx1) of the first darkness to the right end (rx1) of the second darkness is bright.
(2) Look for the pattern of darkness, darkness, and darkness. Confirm that the central pixel from the left end (lx2) of the first darkness to the right end (rx2) of the third darkness is dark.

In the case of (1), as shown in FIG. 18 (A), the upper and lower sides are scanned from the center coordinates ((lx1 + rx1) / 2, (y1 + y2) / 2), and the upper side is bright and dark (end point: ((lx1 + rx1)). / 2), y1-1)), and the lower side is bright and dark (end point: (((lx1 + rx1) / 2), y2 + 1)), and the y coordinate ty1 of the boundary between dark and bright in the upper scan It is confirmed that the midpoint of the y-coordinate by1 at the boundary between dark and light in the lower scan substantially coincides with the center coordinate of light.

In the case of (2), as shown in FIG. 18B, each scan is performed in the vertical direction from the center coordinates ((lx2 + rx2) / 2, (y1 + y2) / 2), and the upper side is dark and dark. (End point: (((lx2 + rx2) / 2), y1-1)), the lower side is dark and dark (end point: ((lx2 + rx2) / 2), y2 + 1)), and the second darkness and lightness of the upper scan It is confirmed that the midpoint of the y-coordinate ty2 of the boundary of the above and the y-coordinate by2 of the second dark and light boundary of the lower scan substantially coincides with the center coordinate of the darkness.

By the above processing, the second position detection pattern having a dark part in the frame (square in the square frame), the first position detection pattern having only the frame and the third position detection pattern (square frame or rectangular frame: B type) Can be distinguished into. Those that do not apply to (1) or (2) are not position detection patterns.

Next, the computer 45 searches for the vertices of each position detection pattern candidate. Scan from the upper left coordinates (x1, y1) of the circumscribed rectangle to (x2, y1) in the X-axis direction. Let the coordinates of the first contacted dark spot be vertex1: (vx1, by1). Next, scanning is performed from the upper right coordinate (x2, y1) to (x2, y2) in the Y-axis direction. Let the coordinates of the first contacted dark spot be vertex2: (vx2, by2). Scan from the lower right coordinates (x2, y2) to (x1, y2) in the opposite direction of the X axis. Let the coordinates of the first contacted dark spot be vertex3: (vx3, by3). Next, scanning is performed from the lower left coordinates (x1, y2) to (x1, y1) in the direction opposite to the Y axis. Let the coordinates of the first contacted dark spot be vertex4: (vx4, by4). vertex1-4 should be rectangular (including squares). If not, it is not a position detection pattern. The notation of FIG. 18B is omitted.

Next, the computer 45 calculates the ratio of the long side to the short side of the rectangle. ((Vx1-vx2) x (vx1-vx2) + (vy1-vy2) x (vy1-vy2)) ^1/2 and the length of one side is ((vx2-vx3) x (vx2-vx3) + ( By 2-vy3) × (vy2-vy3)) ^1/2 , the length of the other side is obtained. Then, the two obtained sides are compared, and the longer side is the long side and the shorter side is the short side.

If the ratio of the sides is (long side): (short side) = 1: 1 and the shape is (2), it is a candidate for the second position detection pattern, and the ratio of the sides is (long side) :( short side). ) = 1: 1 and (1) is the first position detection pattern candidate, and (long side): (short side) = 9: 5 and the shape is (2) is the third position detection pattern candidate. It becomes. In this way, it is possible to distinguish between the three types of position detection pattern candidates.

Next, the computer 45 obtains the area of the position detection pattern candidate. The area here is the area of the outer rectangular frame and includes the area of the inner space. In the case of the second position detection pattern, the square in the frame is also included. The area of the position detection pattern is obtained by subtracting the area of four right triangles including the four corners from the area of the outer rectangle. Area = (x2-x1) x (y2-y1)-(vx1-vx4) x (vy4-vy1) / 2- (vx2-vx1) x (vy2-vy1) / 2- (vx2-vx3) x (vy3) It can be obtained by −vy2) / 2- (vx3-vx4) × (vy3-vy4) / 2.

Next, the computer 45 obtains the rotation angle of the position detection pattern candidate. Since the first position detection pattern and the second position detection pattern are square and cannot detect an inclination of 90 degrees or more, they are obtained up to 90 degrees. The angle is clockwise.
θ = arctan ((vy2-vy1) / (vx2-vx1))
θ is the rotation angle.

The rotation angle of the rectangular third position detection pattern candidate is obtained from 0 to 180 degrees. The third position detection pattern defines a function lens (x1, y1, x2, y2) = ((x2-x1) ² + (y2-y1) ² ) ^1/2 for finding the distance between two points. From 0 to 45 degrees, Length (vx1, vy1, vx2, vy2)> Length (vx4, vy4, vx1, vy1) and (x2-x1)> (y2-y1), and θ = arctan ((vy2-vy1). ) / (Vx2-vx1)). From 45 to 90 degrees, right (vx1, vy1, vx2, vy2)> right (vx4, vy4, vx1, vy1) and (x2-x1) <(y2-y1), and θ = arctan ((vy2-by1). ) / (Vx2-vx1)). From 90 to 135 degrees, Length (vx1, vy1, vx2, vy2) <length (vx4, vy4, vx1, vy1) and (x2-x1) <(y2-y1), and θ = arctan ((vy2-by1). ) / (Vx2-vx1)) +90 degrees. From 135 to 180 degrees is Length (vx1, vy1, vx2, vy2) <length (vx4, vy4, vx1, vy1) and (x2-x1)> (y2-y1), and θ = arctan ((vy2-by1). ) / (Vx2-vx1)) +90 degrees.

Next, the computer 45 detects the center coordinates of the position detection pattern candidate. The center coordinates are the center of the circumscribed rectangle ((x1 + x2) / 2, (y1 + y2) / 2).
This completes the inspection of the position detection pattern candidates.

Next, in step S38, the computer 45 creates a combination of three types of position detection pattern candidates. First, the computer 45 selects the first position detection pattern candidate. Assuming that the area of the first position detection pattern is S, the position detection pattern candidates having the same area and the same rotation angle (0 to 90 degrees) are searched for as the second position detection pattern candidates. When the second position detection pattern candidate is found, next, the position detection pattern candidate having an area of S * 5/9 and the same rotation angle (0 to 180 degrees) is searched for as a third position detection pattern candidate. ..

When a plurality of two-dimensional codes are imprinted on one captured image, the number of combinations becomes enormous when considering all combinations of position detection pattern candidates. By making all the position detection patterns different as in the present invention, the number of combinations can be greatly reduced, and by using features such as shape, area, and rotation obtained in advance, a simple comparison operation can be performed. It is possible to narrow down the number of combinations. Further, according to the two-dimensional code disclosed in the present specification, the number of combinations can be narrowed down as described below due to the characteristics of the arrangement of the position detection patterns. This will be specifically described.

All of the pair of opposing sides of one position detection pattern or the pair of opposing points on the contour of one position detection pattern inside the pair of extension lines extending in the extending direction of the sides or in a predetermined direction. Since each of the other position detection patterns is arranged so that at least a part of them overlaps with each other, the position detection pattern existing at a position outside this condition is excluded from the candidate position detection patterns constituting one two-dimensional code. It is possible to narrow down the number of combinations. For example, in FIG. 1, all the other position detection patterns are arranged so as to overlap each other on a pair of extension lines extending in a direction extending the pair of opposite sides of the position detection pattern. Specifically, the extension lines extending the sides A4-A1 and the sides A2-A3 of the position detection pattern 12A in FIG. 21 overlap the sides B4-B1 and the sides B2-B3 of the position detection pattern 12B, and the positions The sides C4-C1 of the detection pattern 12C and the sides C2-C3 overlap. The sides A4-A1 are represented by the linear equation Y = a1X + b1- (1). The sides A2-A3 are parallel to the sides A4-A1 and are represented by the linear equation Y = a1X + b1'-(2).

The normal of the linear equation passing through the center of gravity of the position detection pattern 12A is represented by Y =-(1 / a1) X + b1''- (3), and the normal of the linear equation passing through the center of gravity of the position detection pattern 12B. The line is represented by Y =-(1 / a1) X + b2- (4), and the normal of the above linear equation passing through the center of gravity of the position detection pattern 12C is Y =-(1 / a1) X + b3- (5). Can be represented by. Let L1 be the distance between the center of gravity of the position detection pattern 12A and the coordinates of the intersection of equations (1) and (3), and the coordinates of the center of gravity of the position detection pattern 12B and the coordinates of the intersection of equations (1) and (4). Let L2 be the distance, and let L3 be the distance between the center of gravity of the position detection pattern 12C and the coordinates of the intersection of equations (1) and (5). When the ratio of L1, L2, and L3 is expressed as L1: L2: L3 = 1: R2: R3 with reference to L1, in FIG. 1, R2 = 1 and R3 = 1.

If the combination of the position detection patterns is correct, R2 = 1 = L2 / L1 and R3 = 1 = L3 / L1. Further, the distance between the center of gravity of the position detection pattern 12A and the coordinates of the intersection of the equations (2) and (3) is L4, and the coordinates of the center of gravity of the position detection pattern 12B and the coordinates of the intersections of the equations (2) and (4). Let L5 be, and let L6 be the distance between the center of gravity of the position detection pattern 12C and the coordinates of the intersection of equations (2) and (5). When the ratio of L4, L5, and L6 is expressed as L4: L5: L6 = 1: R5: R6 with reference to L4, R2 = 1 and R3 = 1 in FIG. If the combination of the position detection patterns is correct, R5 = 1 = L5 / L4 and R6 = 1 = L6 / L4. If the above relationship does not hold, the combination of position detection patterns is incorrect. By this method, it is possible to verify whether or not the combination of position detection patterns is correct. Further, even in the case of FIG. 3A, it can be determined whether or not the combination of the position detection patterns is correct by the same method only by changing the values of R2, R3, R5, and R6. When a pair of opposing points on the contour of the position detection pattern are arranged inside a pair of extension lines extending in parallel in a predetermined direction so that at least a part of each of the other position detection patterns overlaps. However, even when the position detection pattern as shown in FIG. 7 is not a quadrangle, it is possible to determine whether or not the combination is correct regardless of the shape. A pair of sides A4-A1 on a pair of extension lines extending in the extending direction of the sides and a pair of opposing points on the contour of the position detection pattern instead of the sides A2-A3 extending in parallel in a predetermined direction. By using the extension line, it can be similarly determined whether or not the combination of the position detection patterns is correct. Further, a pair of opposing sides of the one position detection pattern or a pair of opposing points on the contour of the one position detection pattern are placed inside a pair of extension lines extending in the extending direction of the sides or in a predetermined direction. Since each of the other position detection patterns is arranged so that at least a part of them overlaps with each other, the position detection pattern existing at a position outside this condition is excluded from the candidate position detection patterns constituting one two-dimensional code. This makes it possible to narrow down the number of combinations.

In step S38, the computer 45 determines whether or not a combination of three types of position detection pattern candidates that have not been examined remains, and if so, proceeds to step S40, and if not, proceeds to step S50.

In step S40, the computer 45 extracts information from the data unit, assuming that the three types of position detection pattern candidates combined are the same two-dimensional code position detection pattern. This process will be described later with reference to FIG.

In step S42, the computer 45 proceeds to step S44 if it succeeds, and if it fails, the computer 45 sets the combination of the failed position detection patterns as the position detection pattern candidate. After performing the process of excluding from the combination target, the process returns to step S38.

In step S44, the computer 45 records the three position detection pattern candidates used for the two-dimensional pattern that succeeded in extracting the data as a combination of the three identified position detection patterns of the same two-dimensional code. do.

Next, in step S46, the computer 45 excludes the three position detection pattern candidates used for the two-dimensional pattern that succeeded in extracting the data from the combination target of the position detection pattern candidates, and returns to step S38. If there are unused position detection pattern candidates within a predetermined range of the two-dimensional pattern for which data has been successfully retrieved, the position detection pattern candidates may be excluded from the combination target of the position detection pattern candidates. good.

By repeating steps S38 to S46 described above, it is confirmed whether or not the combination of the three types of position detection pattern candidates is included in the same two-dimensional code. Here, if there is no combination of three types of position detection pattern candidates that have not been examined, the process proceeds to step S50.

In step S50, the computer 45 determines whether or not a combination of two types of position detection pattern candidates that have not been examined remains, and if so, proceeds to step S52, and if not, proceeds to step S60.

In step S52, information is extracted from the data unit, assuming that the two types of combined position detection pattern candidates are the same two-dimensional code position detection patterns. This process will be described later with reference to FIG.

In step S54, the computer 45 proceeds to step S56 if it succeeds, and if it fails, the computer 45 sets the combination of the failed position detection patterns as the position detection pattern candidate. After performing the process of excluding from the combination target, the process returns to step S50.

In step S56, the computer 45 records the two position detection pattern candidates used for the two-dimensional pattern that succeeded in extracting the data as a combination of the two identified position detection patterns of the same two-dimensional code. do.

Next, in step S58, the computer 45 excludes the two position detection pattern candidates used for the two-dimensional pattern that succeeded in extracting the data from the combination target of the position detection pattern candidates, and returns to step S50. If there is an unused position detection pattern candidate in a predetermined range of the two-dimensional pattern for which data has been successfully retrieved, the position detection pattern candidate is excluded from the combination target of the position detection pattern candidate.

By repeating steps S50 to S58 described above, it is confirmed whether or not the combination of the two types of position detection pattern candidates is included in the same two-dimensional code. Here, if there is no combination of two types of position detection pattern candidates that have not been examined, the process proceeds to step S60.

In step S60, the computer 45 determines whether or not one type of position detection pattern candidate that has not been examined remains, and if it remains, the process proceeds to step S62, and if not, the process proceeds to step S72. In step S72, the analysis main process is terminated.

In step S62, when the detected position detection pattern candidate is a position detection pattern, the computer 45 identifies the position correction pattern arranged at a predetermined position with respect to the position detection pattern. Is determined. If this position correction pattern is identified, the process proceeds to step 64. Since the computer 45 can identify the position correction pattern arranged at a predetermined position with respect to the position detection pattern candidate, the computer 45 identifies that the position detection pattern candidate is the position detection pattern. On the other hand, when the position correction pattern is not identified, the process of excluding one detected position detection pattern candidate from the target of one unexamined position detection pattern candidate is performed, and then the process returns to step S50. ..

Here, the process of detecting the position correction pattern will be described below with reference to FIG. If a position correction pattern exists in the two-dimensional code, the position correction pattern is detected. When detecting the position correction pattern, as in the position detection pattern shown in FIG. 18, the screen is scanned in the X direction starting from the upper left point of the image, and when the right end is reached, the position is scanned in the X direction from the left end several pixels below. do. In other words, it will be subtracted for a few pixels. This is repeated for the entire image. The number of pixels to be thinned out is (maximum thinning amount) = (minimum cell size in the side length of the position correction pattern) × (number of pixels per cell at the time of shooting). The minimum cell size among the side lengths of all the position correction patterns is the minimum cell size in the Y direction that can occur in the rotation of the position correction pattern in consideration of the rotation of the cord. If the number of pixels per cell at the time of shooting is less than 1 pixel per cell, it becomes difficult to distinguish between the light and dark of the cell even in an ideal shot image. Considering the displacement or rotation of cells in the actual captured image, it is desirable that the number of pixels is at least 3 pixels per cell. Since this process is almost the same as the scanning for detecting the position detection pattern except for the pixel size to be thinned out, it may be performed at the same time. In that case, adjust the pixel size to be thinned out to the smaller one. Then, the computer 45 inspects the detected position correction pattern.

Specifically, when a light-dark boundary is found during scanning, the area around the dark portion is scanned. At that time, the minimum and maximum X and Y coordinates are obtained, respectively. As a result, the circumscribed rectangle of the dark mass can be obtained. The upper left coordinates are represented as (x1', y1'), and the lower right coordinates are represented as (x2', y2'). FIG. 19 is a diagram illustrating a shape check of the position correction pattern. First, the computer 45 checks the shape of the position correction pattern.

The values of (x2'-x1') and (y2'-y1') are almost the same. The position correction patterns 16A and 16B of the two-dimensional code shown in FIG. 1 are 3 × 3 dark cells. Therefore, first make sure that there are no bright holes. Scanning in the x-axis direction is performed with the start point as the left midpoint (x1', (y1' + y2') / 2) of the circumscribed rectangle and the end point as (x2', (y1' + y2') / 2). Here, it is confirmed that everything from the left end of darkness (lx1') to the right end of darkness (rx1') is dark. Further, as shown in FIG. 19, scanning up and down from the center coordinates ((lx1'+ rx1') / 2, (y1'+ y2') / 2), also in the y-axis direction, the upper end of darkness and the lower end of darkness are scanned. Make sure everything is dark.

Next, the computer 45 searches for the vertices of each position correction pattern. Scan from the upper left coordinates (x1', y1') of the circumscribed rectangle to (x2', y1') in the X-axis direction. Let the coordinates of the first contacted dark spot be vertex1: (vx1', vy1'). Next, scanning is performed from the upper right coordinates (x2', y1') to (x2', y2') in the Y-axis direction. Let the coordinates of the first contacted dark spot be vertex2 :( vx2', vy2'). Scan from the lower right coordinates (x2', y2') to (x1', y2') in the opposite direction of the X-axis. Let the coordinates of the first contacted dark spot be vertex3 :( vx3', vy3'). Next, scanning is performed from the lower left coordinates (x1', y2') to (x1', y1') in the direction opposite to the Y axis. Let the coordinates of the first contacted dark spot be vertex4 :( vx4', vy4'). vertex1-4 should be square. If not, it is determined that the pattern is not a position correction pattern.

Next, the computer 45 obtains the area of the position correction pattern. The area of the position correction pattern is obtained by subtracting the area of four right triangles including the four corners from the area of the circumscribed square. Area = (x2'-x1') x (y2'-y1')-(vx1'-vx4') x (vy4'-vy1') / 2- (vx2'-vx1') x (vy2'-vy1' ) / 2- (Vx2'-vx3') x (vy3'-vy2') / 2- (vx3'-vx4') x (vy3'-vy4') / 2.

Next, the computer 45 obtains the rotation angle of the position correction pattern candidate. Since the position correction pattern is a square and cannot detect an inclination of 90 degrees or more, an angle up to 90 degrees is obtained. The angle is clockwise. θ = arctan ((vy2'-vy1') / (vx2'-vx1')) θ is the rotation angle. These tests are simple, and there are many similar shapes (black squares) that will pick up a lot of noise. The position correction pattern is used to correct the coordinates of a cell and create a position detection pattern candidate after the position of the two-dimensional code in the image is roughly determined by the position detection pattern, and the detected position correction is performed. It is not necessary to determine whether or not the pattern is correct by itself, and it is sufficient if it is possible to determine whether or not the pattern is the correct position correction pattern based on the identified position detection pattern or the positional relationship with the position detection pattern candidate. On the other hand, even if the position correction pattern is not found, it is possible to acquire the position detection pattern candidate, and the position correction pattern is not essential unlike the position detection pattern.

In step S64, the computer 45 extracts information from the data unit, assuming that the detected position detection pattern candidate is the position detection pattern of the two-dimensional code. This process will be described later with reference to FIG.

In step S66, the computer 45 proceeds to step S68 if it succeeds, and if it fails, the computer 45 targets the failed position detection pattern candidate as the target of the position detection pattern candidate. After performing the process of excluding from, the process returns to step S60.

In step S68, the computer 45 records one position detection pattern candidate used for the two-dimensional pattern that succeeded in extracting the data as the identified position detection pattern of the two-dimensional code.

Next, in step S70, the computer 45 excludes one position detection pattern candidate used for the two-dimensional pattern that succeeded in extracting data from the target of the position detection pattern candidate, and returns to step S60. If there is an unused position detection pattern candidate in a predetermined range of the two-dimensional pattern for which data has been successfully retrieved, the position detection pattern candidate is excluded from the target of the position detection pattern candidate.

By repeating steps S60 to S70 described above, it is confirmed whether or not one type of position detection pattern candidate is the position detection pattern of the two-dimensional code. Here, if one type of position detection pattern candidate that has not been excluded does not remain, the process proceeds to step S72. In step S72, the analysis main process is terminated. In order to shorten the processing time, if there is no combination of two types of position detection pattern candidates that have not been examined in step S50, the analysis main processing may be terminated in step S72.

Next, the information retrieval process in steps S40, S52, and S64 described above will be described below with reference to FIGS. 20 to 23.

First, in step S80, the computer 45 starts the information retrieval process.

Next, in step S82, the computer 45 acquires the coordinates of a total of four points from one or a plurality of position detection patterns. First, the process in the case of step S40 will be described below.

In the case of step S40, the computer 45 selects two position detection pattern candidates from the three detected position detection pattern candidates, and at least one point from the selected one position detection pattern candidate. The coordinates of a total of four points are acquired from at least two selected position detection pattern candidates so as to acquire the coordinates. At this time, it is desirable that two outer positions are selected for the position detection pattern, two points at the four corners are selected from each position detection pattern, and the outer shape connecting the four selected points is rectangular. If the information is correctly obtained from the extracted message in step S96 described later, the detected position detection pattern candidate is the identified position detection pattern. The computer 45 may select at least one notation position detection pattern from the at least three identified position detection patterns and acquire the coordinates of a total of four points from the selected position detection pattern. ..

FIG. 21 is a diagram illustrating that coordinates of a total of four points are acquired from the three detected position detection pattern candidates.

As shown in FIG. 21, it is assumed that three position detection pattern candidates are detected based on the captured image of the two-dimensional code. The computer 45 preferably acquires the coordinates of the four points from the points forming the contours of the plurality of position detection patterns. In the example shown in FIG. 21, the outer shape of the position detection patterns 12A, 12B, 12C is a quadrangle, and the computer 45 acquires the coordinates of the four points from the vertices of the quadrangle forming the contour of the position detection patterns 12A, 12B, 12C. do. Since the coordinates of the four points are acquired and used as parameters for performing the projective transformation of the two-dimensional code, it is preferable to acquire the four points that are not arranged on the same straight line. This idea also applies to the processes of S52 and S64 described later.

The first position detection pattern 12A has four vertices A1, vertices A2, vertices A3, and vertices A4 in a counterclockwise direction from the upper left position. The second position detection pattern 12B has four vertices B1, vertices B2, vertices B3, and vertices B4 in a counterclockwise direction from the upper left position. The third position detection pattern 12C has four vertices C1, vertices C2, vertices C3, and vertices C4 in a counterclockwise direction from the upper left position.

The computer 45 selects the first position detection pattern 12A and the third position detection pattern 12C, which are the farthest from the three different position detection pattern candidates. Then, the computer 45 acquires the coordinates of the two vertices A1 and A2 from the first position detection pattern 12A, and acquires the coordinates of the two vertices C3 and C4 from the third position detection pattern 12C. In this way, the reason why the two position detection patterns that are farthest apart are selected and the four vertices that are farthest apart are acquired is that they are close to the data unit 11 that is the calculation target of data acquisition. Is desired. When the coordinates of four points are acquired from one position detection pattern, the calculation accuracy of the coordinates in the distant data unit 11 deteriorates. Therefore, it is desirable to acquire four points that can cover the entire data unit 11 widely. Further, when the code captured is distorted due to the distortion of the lens of the reader or the distortion of the surface on which the two-dimensional code is printed, the distortion is not uniform. Therefore, it is better to calculate the coefficient based on multiple points that span different position detection patterns than to obtain the coefficient of projective transformation in two-dimensional space from the coordinates of four points acquired from the same position detection pattern. The effect that can be reduced can be expected.

Then, it is desirable that the computer 45 satisfies the following conditions (a) and (b) for the combination of the acquired four vertices. If these conditions are not satisfied, the process may proceed to step S100. In step S100, the retrieval of information fails, and the process ends.

(A) The angles of the four vertices of the quadrangle formed by connecting the four vertices are at right angles, and the lengths of the opposite sides are 1: 1.
(B) The quadrangle formed by connecting the four vertices is separated from the data unit by a predetermined distance or more.

The above is the description of the process in the case of step S40.

Next, the process in the case of step S52 will be described below.

In the case of step S52, the computer 45 is detected so as to acquire the coordinates of at least one point from one position detection pattern candidate based on the two detected position detection pattern candidates. The coordinates of a total of 4 points are acquired from the position detection pattern candidates. If the information is correctly obtained from the extracted message in step S96 described later, the detected position detection pattern candidate is the identified position detection pattern. The computer 45 may select at least one position detection pattern from the two identified position detection patterns and acquire the coordinates of a total of four points from the selected position detection patterns.

22 (A) to 22 (C) are diagrams for explaining that the coordinates of a total of four points are acquired from the two detected position detection pattern candidates.

As shown in FIGS. 22 (A) to 22 (C), it is assumed that two of the three position detection pattern candidates are detected based on the captured two-dimensional code image. One of the three position detection patterns has not been detected due to dirt on the two-dimensional code or blurring / blurring during shooting.

In the example shown in FIG. 22 (A), the first position detection pattern 12A and the third position detection pattern 12C are detected, but the second position detection pattern 12B is not detected. The computer 45 acquires the coordinates of the two vertices A1 and A2 from the first position detection pattern 12A, and acquires the coordinates of the two vertices C3 and C4 from the third position detection pattern 12C. The reason why the four vertices most distant from each other are acquired is that it is desirable that the coordinates of the four points for performing the projective transformation are close to the data unit 11 to be calculated for data acquisition. When the coordinates of four points are acquired from one position detection pattern, the calculation accuracy of the coordinates in the distant data unit 11 deteriorates. Therefore, it is desirable to acquire four points that can cover the data unit 11 widely. Further, when the code captured is distorted due to the distortion of the lens of the reader or the distortion of the surface on which the two-dimensional code is printed, the distortion is not uniform. Therefore, it is better to calculate the coefficient based on multiple points that span different position detection patterns than to obtain the coefficient of projective transformation in two-dimensional space from the coordinates of four points acquired from the same position detection pattern. The effect that can be reduced can be expected. This idea also applies to the processes of FIGS. 22 (B) and 22 (C).

In the example shown in FIG. 22B, the second position detection pattern 12B and the third position detection pattern 12C are detected, but the first position detection pattern 12A is not detected. The computer 45 acquires the coordinates of the two vertices B1 and B2 from the second position detection pattern 12B, and acquires the coordinates of the two vertices C3 and C4 from the third position detection pattern 12C.

In the example shown in FIG. 22C, the first position detection pattern 12A and the second position detection pattern 12B are detected, but the third position detection pattern 12C is not detected. The computer 45 acquires the coordinates of the two vertices A1 and A2 from the first position detection pattern 12A, and acquires the coordinates of the two vertices B3 and B4 from the second position detection pattern 12B.

Then, it is desirable that the computer 45 satisfies the following conditions (a) and (b) for the combination of the acquired four vertices. If these conditions are not satisfied, the process may proceed to step S100. In step S100, the retrieval of information fails, and the process ends.

(A) The angles of the four vertices of the quadrangle formed by connecting the four vertices are at right angles, and the lengths of the opposite sides are 1: 1.
(B) The quadrangle formed by connecting the four vertices is separated from the data unit by a predetermined distance or more.

The above is the description of the process in the case of step S52.

Next, the process in the case of step S64 will be described below.

In the case of step S64, the computer 45 acquires the coordinates of four points from the identified position detection pattern.

23 (A) to 23 (C) are diagrams for explaining that the coordinates of a total of four points are acquired from one detected position detection pattern candidate.

As shown in FIGS. 23 (A) to 23 (C), it is assumed that one of the three position detection patterns is identified based on the captured two-dimensional code image. Two of the three position detection patterns have not been detected due to dirt on the two-dimensional code or blurring / blurring during shooting. If the position detection pattern candidate is correct, the position correction pattern should be found at a predetermined position. The midpoint of the apex A1 and the apex A2 of the position detection pattern 12A is M1, the midpoint of the apex A2 and the apex A3 is M2, the midpoint of the apex A3 and the apex A4 is M3, and the midpoint of the apex A4 and the apex A1 is M4. .. From the coordinates of the center of gravity of the position detection pattern, a straight line passing through the four points M1, M2, M3, and M4 is drawn. The position correction pattern should exist on the straight line and at a predetermined distance (for example, 14.5 cells) from the coordinates of the center of gravity. Further, the area of the position detection pattern and the area of the position correction pattern are compared. For example, if the area of the position detection pattern: the area of the position correction pattern = 64: 9 (8 × 8 cells: 3 × 3 cells), the position detection pattern Exclude from candidates. Furthermore, the rotation angles of the position detection pattern and the position correction pattern are compared, and if they are not the same, they are excluded from the candidates for the position detection pattern. From the above determination, the correct position detection pattern can be obtained from the position detection pattern candidate and the position correction pattern.

In the example shown in FIG. 23 (A), the first position detection pattern 12A is identified by identifying the position correction pattern 16A, and the second position detection pattern 12B and the third position detection pattern 12C are not detected. .. The computer 45 acquires the coordinates of the four vertices A1, the vertices A2, the vertices A3, and the vertices A4 from the first position detection pattern 12A. The reason why the four vertices of the quadrangular position detection pattern are acquired is to avoid acquiring locally soiled points on the position detection pattern. In the example shown in FIG. 23 (A), since the four vertices are acquired from one position detection pattern, a data unit located at the end of the code far away from the four vertices is generated, so that the first position Compared with the case where four points are acquired from the detection pattern 12A and the third position detection pattern 12C, the accuracy is lower, but the data can be detected. This idea also applies to the processes of FIGS. 23 (B) and 23 (C).

In the example shown in FIG. 23B, the second position detection pattern 12B is identified by identifying the position correction pattern 16A or 16B, and the first position detection pattern 12A and the third position detection pattern 12C are detected. Not. The computer 45 acquires the coordinates of the four vertices B1, the vertices B2, the vertices B3, and the vertices B4 from the second position detection pattern 12B.

In the example shown in FIG. 23C, the third position detection pattern 12C is identified by identifying the position correction pattern 16B, and the first position detection pattern 12A and the second position detection pattern 12B are not detected. .. The computer 45 acquires the coordinates of the four vertices C1, the vertices C2, the vertices C3, and the vertices AC2 from the third position detection pattern 12C.

Then, it is desirable that the computer 45 satisfies the following conditions (a) and (b) for the combination of the acquired four vertices. If these conditions are not satisfied, the process may proceed to step S100. In step S100, the retrieval of information fails, and the process ends.

(A) The angles of the four vertices of the quadrangle formed by connecting the four vertices are at right angles, and the lengths of the opposite sides are 1: 1.
(B) The quadrangle formed by connecting the four vertices is separated from the data unit by a predetermined distance or more.

The above is the description of the process in the case of step S64.

When the computer 45 finishes the process of step S82, the computer 45 proceeds to step S84.

Next, in step S84, the computer 45 obtains the coefficient of the projective conversion in the two-dimensional space based on the acquired coordinates of the four points, performs the coordinate conversion of the image of the two-dimensional code by the projective conversion, and distorts or causes it. Generates an image of the original shape of the blurred 2D code.

Next, in step S86, the computer 45 obtains the coordinates and pixel values of each cell in the image of the two-dimensional code whose coordinates have been converted to the original shape.

Next, in step S88, the computer 45 extracts real data in binary format including a message, an error correction code, and the like based on the pixel values of each cell.

Next, in step S90, the computer 45 uses the error correction code to detect the number and position of errors with respect to the data unit and the error correction code. If the number of errors is zero, the process proceeds to step S96. If the number of errors is not zero, the process proceeds to step S92.

In step S92, the computer 45 determines whether or not the error can be corrected, and if the error can be corrected, the process proceeds to step S94, and if the error cannot be corrected, the process proceeds to step S100. In step S100, the retrieval of information fails, and the process ends.

In step S94, the computer 45 corrects the error of the actual data by using the error correction code.

Next, in step S96, the computer 45 restores and stores the message based on the actual data.

Next, in step S98, since the computer 45 succeeded in extracting the information, the extraction of the information is finished, and the process proceeds to the next process of steps S40, S52, or S64.

As shown in the examples of FIGS. 22 and 23, the number of position detection patterns is preferably large in consideration of the fact that some of the position detection patterns cannot be detected when the two-dimensional code is dirty. However, if the number is large, the ratio of the area of the area in the two-dimensional code occupied by the position detection pattern increases, so that the data efficiency decreases and the processing amount for detecting the position detection pattern also increases. Therefore, it is preferable that the number of position detection patterns is appropriately determined according to the usage pattern of the two-dimensional code and the like.

In the present invention, the above-mentioned two-dimensional code, two-dimensional code analysis method, two-dimensional code analysis device, and program for analyzing the two-dimensional code can be appropriately changed as long as the gist of the present invention is not deviated. Further, the constituent requirements of one embodiment can be appropriately applied to other embodiments.

10 2D code 10A Basic pattern part 10B Peripheral part 11 Data part 12A 1st position detection pattern 12B 2nd position detection pattern 12C 3rd position detection pattern 13 cells 14 blocks 15 Separation space 16A, 16B Position correction pattern L Extension line 20 cards 21 Patterns 30 31 Back cover 40 Analyzer 41 Reading unit 42 Lens 43 Image sensor 44 Analog-to-digital converter (AD)
45 Computer 46 Display 47 Communication interface

Claims (10)

Multiple cells arranged as a two-dimensional matrix pattern and representing data represented by a binary code,
Alone a position detection pattern of three or more shapes detectable different from the image, the two-dimensional code image based on the positional relationship of the three or more position detecting patterns having different shapes when parsing the two-dimensional code A position detection pattern for performing coordinate conversion and
With
The outer shape of each of the position detection patterns is a quadrangle, and each of the other position detection patterns has a pair of opposite sides of the position detection pattern inside a pair of extension lines extending in the extending direction of the sides. However, a matrix-type two-dimensional code that is arranged so that at least a part of it overlaps. The matrix-type two-dimensional code according to claim 1, wherein at least a part of the contour of the other position detection pattern overlaps with one of the pair of extension lines. The matrix-type two-dimensional code according to claim 1, wherein the contour of the other position detection pattern is arranged inside the pair of extension lines. The matrix-type two-dimensional code according to claim 1, wherein the pair of opposite sides of the other position detection pattern are arranged on the pair of extension lines. The matrix-type two-dimensional code according to any one of claims 1 to 4, wherein the at least one position detection pattern has a single square frame and squares arranged in the square frame. The matrix-type two-dimensional code according to any one of claims 1 to 5, wherein the at least one position detection pattern has a single square frame or a rectangular frame. A position correction pattern arranged at a predetermined position with respect to the position detection pattern, and a pattern arranged at the predetermined position with respect to the position correction pattern during analysis of a two-dimensional code is identified as the position detection pattern. In any one of claims 1 to 6, the position correction pattern is provided in which the error of the coordinate of the cell is corrected based on the positional relationship with the position detection pattern at the time of analyzing the two-dimensional code. The matrix type two-dimensional code described. The position correction pattern, the matrix type two-dimensional code of claim 7 which is arranged in the center between the two different that prior Symbol position detection pattern shape adjacent. The first position detection pattern, the second position detection pattern, the third position detection pattern, the first position correction pattern, and the second position correction pattern are provided.
The first position correction pattern is arranged at the center between the adjacent first position detection pattern and the second position detection pattern.
The matrix-type two-dimensional code according to claim 7 or 8, wherein the second position correction pattern is arranged at the center between the adjacent second position detection pattern and the third position detection pattern. The matrix type according to any one of claims 1 to 8, wherein the position detection pattern and all the other position detection patterns are arranged at equal intervals in the extending direction of the pair of extension lines. Two-dimensional code.

Images