JPH0431435B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical field to which the invention is directed] This invention relates to a method for reading the mesh code of each field type code in a code matrix in which a plurality of field shape codes each representing several bits of information are arranged in one code. .

[Prior art and its problems]

Conventionally, codes are printed on the packaging of products and are used for various management purposes. This code generally uses a so-called barcode that expresses the code by the thickness and arrangement of black and white lines, but this barcode has the following problems. In other words, since barcodes arrange information in one dimension, they can only express a small amount of information, and barcode printing requires high precision, making it impossible for humans to directly draw the code. Moreover, the barcode cannot be easily read by humans.

In order to solve this problem, it is constructed by combining four rectangular meshes in a square shape.
In addition, a so-called Tagata code was devised, which displays a code by selectively coloring each mesh in black or white. This results in code 2
Dimensional arrangement is possible, allowing more information to be expressed; therefore, it is possible to draw only the frame of the field and write codes by selectively filling in each mesh like a mark sheet. It became like this,
Moreover, since the code is basically 4 bits, it can be easily read by humans.

However, there is almost no innovation in how to read these codes, how to arrange the codes, or with auxiliary graphics.

[Purpose of the invention]

This invention takes into consideration these conventional drawbacks,
The purpose is to accurately know the position of the Tagata code and to read the data accurately.

[Summary of the invention]

In order to achieve the object, the present invention forms a code matrix using a plurality of Tagata code, and provides a rectangular reference line around the code matrix at a predetermined distance from the code matrix. The central coordinates of the Tagata code are calculated from the relative positional relationship with the code matrix, and the code of each mesh is read from the center. The position of each mesh can be read accurately even if the images are shifted in the rotational direction.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

In other words, in Fig. 2, reference numeral 1 indicates the basic configuration of the well-known Tagata cord. For example, four square meshes each having a side of 2 mm are combined in a square shape to form a Tagata cord having a side of 4 mm as a whole. be done. This field code can represent information of a predetermined number of bits by filling out each mesh.
For example, if four meshes are used, there will be 4 bits of information, and the number will be 16. As shown in Figure 3, if all the meshes are filled in, the value becomes 0, and if none of the meshes are filled in, it becomes 15, and if the upper left mesh is left and the other meshes are filled in, then it becomes 15. 1. If you leave the bottom left and fill in the other meshes, you can get 2. If you leave the top right and fill in the other meshes, you can get 4. If you fill out the upper half, you can get 8.

In Figure 1, a code matrix 2 is formed by arranging a plurality of Tagata codes 1 at predetermined intervals in a matrix of M rows and N columns. In this embodiment, 20 Tagata codes are arranged vertically and horizontally. It is formed by arranging them in a matrix of 4 rows and 5 columns with an interval of 2 mm between each. A rectangular reference line 3 having a predetermined dimension is arranged around the code matrix 2, maintaining a predetermined distance from the code matrix. This reference line is the first
As shown in the figure, they are arranged 2 mm apart from each field code 1 located on the outer periphery of the code matrix 2, and the width, that is, the thickness of the line, is set to 2 mm. Therefore, the dimension of the reference line 3 in the column direction is 6(M+1), and the dimension in the row direction is 6(N+1).

Next, a method for reading tag format codes according to the present invention will be briefly explained with reference to FIGS. 4 to 8. First, the equation of each straight line is determined by scanning the image vertically and horizontally (X direction, Y direction) to obtain coordinates on each straight line forming a rectangle, that is, on the reference line 3. Next, from the equation of each straight line, the intersection of each straight line is 4
The coordinates of the four corners (intersection points) C ₁ , C ₂ , C ₃ , and C ₄ are calculated, and the coordinates of each corner code 1 are calculated based on the coordinates of the four corners C ₁ , C ₂ , C ₃ , and C ₄ . After calculating the center coordinates P _ij of each field code 1 from the relative positional relationship, the center coordinates Q of the four meshes (rectangular codes) constituting each field code 1 are calculated from the center coordinates P _ij of each field code 1. ₁ , Q ₂ , Q ₃ , and Q ₄ are calculated. Finally, pixels near the center coordinates Q ₁ , Q ₂ , Q ₃ , and Q ₄ of the mesh are read, and the information of each field shape code 1 is read by determining the shade of color of the pixels. In this way, once the equation of each straight line is determined, the corners C ₁ , C ₂ ,
The reason why we are finding the coordinates of C ₃ and C ₄ is because the corners C ₁ , C ₂ ,
This is because it is difficult to obtain a stable image by scanning C ₃ and C ₄ and the error becomes large.

Figure 4 shows 4 determined by the rectangular reference line 3.
Three corners C ₁ , C ₂ , C ₃ and C ₄ are shown. In the figure, x and y are shown in directions inclined with respect to the reference line 3. This generally corresponds to the x and y directions of the captured image.
This is because the direction is not necessarily parallel to the reference line.

Next, the method for reading the Tagata code of the present invention will be explained in detail.

Assume now that the image magnification of the camera section of the reading device is constant. First, consider obtaining one edge. It is assumed that the rectangular frame in the image has an inclination within a range of ±45 degrees. FIG. 5 shows a rectangular frame that fits within the image. While scanning in the x direction in the figure, search for a line where the number of black pixels can be detected. For a certain number, the thickness of the frame is 80% of the corresponding number of pixels in the image. The coordinates of the first black pixel on this first detection line and the coordinates of the first black pixel when scanning this line in reverse are memorized (see Figure 5).
_A1 and _B1 ). Next, scan lines separated by y ₀ in the y direction and store the coordinates of A ₂ and B ₂ . Finally, scan the scan line y ₀ apart in the y direction once again.
Memorize the coordinates of A ₃ and B ₃ . This y ₀ is the maximum interval between three scanning lines that can detect the shorter side when the rectangular frame is tilted by 45 degrees.

Of the obtained points A ₁ to _{A 3} and B ₁ to _{B 3} , the point sequence with the least change in the x direction is adopted. For example, if the long side of the rectangular frame has an inclination of 0 degrees to 45 degrees, A ₁ to
A ₃ is adopted (Figure 7a). Tilt is 0 degrees to -45
In the case of degree, B ₁ to _{B 3} are adopted (Figure 7 b).

When the obtained point sequence is A ₁ to A ₃ , the seventh
Scanning is performed as shown in Figure A. Store A ₁ and B ₁ obtained from the first scan. Next, start scanning by rotating the scanning starting point and scanning direction 90 degrees clockwise. For the first time in this scan, a scanning line in which black pixel columns continue for a predetermined number of times is determined. Then, find the coordinates of the black pixel A ₄ that appears for the first time when scanning from the left on this line, and the coordinates of the first black pixel B ₄ when scanning in the opposite direction. In the same way, find the coordinates of A ₅ , B ₅ , A ₆ , and B ₆ in Figure 7A. The eight points found (A ₁ , B ₁ , A ₄ , B ₄ , A ₅ , B ₅ , A ₆ , B ₆ )
Then, find the equations of the straight lines of the four sides, and find the coordinates of the four corners as their intersection points. The obtained point sequence is
When B ₁ to _{B 3} , scan as shown in Figure 7 B,
The coordinates of the four corners are determined in the same manner as when the obtained point sequence is _A1 to _A3 .

That is, C ₁ = (x ₁ , y ₁ ), C ₂ = (x ₂ , y ₂ ), C ₃ = (x ₃ , y ₃ ), C ₄ = (x ₄ , y ₄ )...(1) .

Next, the coordinates of the center of each field code are determined from the coordinates of the four corners. Here, the straight line C ₁ C ₂ can be considered as a straight line that passes through the center of the row when one more row is added to the upper end of the code matrix 2. Similarly, the straight lines C ₂ C ₃ , C ₃ C ₄ and
C ₄ C ₁ also has Tagata code 1 outside the code matrix,
When considering the row or column of 1, it can be thought of as a straight line passing through the center of the Tagata code of this extra row or column. Therefore, by dividing the length of the line segment C ₁ C ₂ by N+1, the pitch xp of the field codes 1 and 1 in the row direction can be found. Similarly, by dividing the length of the line segment C ₁ C ₄ by M+1, the pitch yp in the column direction of the tag type codes 1 and 1 can be found.
In actual measurements, to reduce errors, xp uses line segments C ₁ C ₂ and C ₃ C ₄ , and yp uses line segments C ₂ C ₃ and C ₁ C ₄ .

The rotation angle θ of the entire code with respect to the screen can also be determined from C ₁ and C ₂ .

When C ₁ is the origin, C ₁ C ₂ is the x-axis, and C ₁ C ₄ is the Y-axis, find the center coordinates of the Tagata code closest to C _1.
Let's say P ₁₁ . The interval between the Tagata cords is equal to half of one side of the Tagata cords in both the x and y directions. Therefore, the center coordinate Pij of the Tagata code in the i-th row and the j-th column is expressed by the following equation.

Px=xpi, py=ypj...(4) Figure 5 shows the relationship between the center of the Tagata code and the center of each mesh. Center coordinates of each mesh Q ₁ , Q ₂ ,
Q ₃ and Q ₄ are expressed as follows.

The coordinates in the actual image form an angle θ with the coordinate system with the axis of the sides of the rectangle as the following equation.

1 centered on the center of each mesh thus determined
Pixels within a square area with side length a are read, and the value within that area is determined to be 0 or 1. After determining the values for the four meshes, the value of the field code 1 is decoded. Do this for all Tagata codes. For example, if the mesh values of Q ₁ , Q ₂ , Q ₃ , and Q ₄ are 1, 0, 1, and 0, respectively (black is 0 and white is 1), the value of this taga code is It becomes 5.

When reading the Tagata code using the above algorithm, it can be read even when the Tagata code is tilted, and even codes written by humans can be read as long as the center of the mesh is properly marked.

A block diagram of a reading device for realizing the above algorithm is shown in FIG. 4 is a sheet with the matrix of Tagata code 1, 1 written, 5 is a light source, 6
is a lens, and 7 is a two-dimensional image sensor. The field codes 1, 1 written on the sheet 4 are illuminated by a light source 5. Illuminated Tagata code 1,1
is formed onto a two-dimensional image sensor 7 by a lens 6. The two-dimensional image sensor is driven by a drive circuit 1 from a clock generated by a clock generator 11.
Analog data of each pixel is output by a drive signal generated through 0. The output analog data is amplified through a preamplifier 8 and sent to a binarization circuit 9. The binarization circuit compares the analog signal sent to it with a predetermined reference voltage,
Convert to binary data. At this time, the DMA controller 12 receives the clock generated from the clock generator and generates and provides an address to the frame memory 13. The binarized data sent at this time is stored in the frame memory.
The control circuit 14 receives the signal from the trigger circuit 17 and sends the image data input signal to the DMA controller 1.
Give to 2. Therefore, the image data when the trigger is applied is binarized and stored in the frame memory 13.
is memorized. After inputting the image, the control circuit 14 executes the above-described taga code reading algorithm, displays the result on the display 16, and outputs a signal to an external device through the output circuit 15.

FIG. 9 shows an example of the appearance and usage of the reading device. Reference numeral 19 denotes a reading section of the reading device, which illuminates the sheet 4 with light from the light source 5 and captures an image. Image input is performed by pressing the trigger switch 20. When the decoding of the Tagata codes 1, 1 is completed, the results are displayed on the display 16. The results are also sent to external equipment through the output terminal 18.

〔Effect of the invention〕

As described above, according to the present invention, the formula of each straight line is obtained by scanning the image vertically and horizontally to obtain the coordinates on each straight line forming a rectangle, and from the formula of each straight line, the four intersection points where each straight line intersects are determined. Calculate the coordinates of the four intersection points, calculate the center coordinates of each field code from the relative positional relationship of each field code, and then calculate each field code from the center coordinates of each field code. Calculate the center coordinates of the four rectangular codes that make up the code, read the pixels near the center coordinates of the four rectangular codes, and read the information of each field code by determining the shade of color of the pixel. Therefore, even if the image is tilted, the position of each tag code can be accurately recognized as long as the tilt is less than 90°. This makes it possible to read the information on each tag code accurately, and also to determine the position of the reader. This has the effect of eliminating the need to strictly perform alignment.

[Brief explanation of drawings]

Fig. 1 shows an example of the method for reading the Tagata code according to the present invention, and shows the arrangement of its patterns, Fig. 2 shows the basic configuration of the Tagata code, and Fig. 3 shows an example of the display of the Tagata code. Configuration diagram shown, No. 4
5, 6, 7 and 8 are explanatory diagrams of the reading method, FIG. 9 is an external perspective view of the reading device, and FIG. 10 is a block diagram of the reading device. 1...Tagata code, 2...Code matrix,
3...Reference line, 4...Sheet, 5...Light source, 6...
... Lens, 7 ... Two-dimensional image sensor, 8 ... Preamplifier, 10 ... Drive circuit, 11 ... Clock generator, 12 ... DMA controller, 13 ...
...Frame memory, 14...Control circuit, 15...
Output circuit, 16... Display, 17... Trigger circuit, 18... Output end, 19... Reading section.

Claims (1)

[Claims] 1. When reading information from a plurality of Tagata codes arranged in rows and columns at predetermined intervals within a rectangular frame used as a reference line, the image is scanned vertically and horizontally on each straight line that makes up the rectangle. By obtaining the coordinates of
After calculating the center coordinates of each field code from the relative positional relationship of each field code using the coordinates of the four intersection points as a reference, the four directions constituting each field code are calculated from the center coordinates of each field code. A method for reading tag shape codes that calculates the center coordinates of the shape code, reads pixels in the vicinity of the center coordinates of the four rectangular codes, and reads information on each tag shape code by determining the shade of color of the pixel. .