JP2798013B2 - Two-dimensional code reading method and apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for reading a two-dimensional code which displays various information such as a product name on various kinds of products or other articles or their packaging.

[0002]

2. Description of the Related Art Conventionally, as this kind of information display code,
One-dimensional barcodes are widely practiced. The one-dimensional barcode is configured to recognize data by a combination of a thick bar, a thin bar, and a space.

However, a one-dimensional bar code for recognizing such data becomes sensitive to printing accuracy when reading the code, and has no error restoring ability. If they are attached, they often become unreadable.

[0004] Therefore, with a one-dimensional bar code, there is no space for marking electronic components, high-density mounting boards, small components, and the like, and a sufficient management system cannot be constructed.

[0005] In addition, a one-dimensional barcode requires a larger marking space when the amount of input data is increased, and a label size is also increased.

Further, when managing codes of electronic parts, medical instruments, ceramic parts, metal parts, and the like, since the labeling method in which the one-dimensional bar code is printed is not practical, the codes are directly marked with a laser marker or the like. However, the one-dimensional barcode has a drawback that the reading rate is not good.

[0007] Since the one-dimensional barcode is only an ID code for searching the database, the one-dimensional barcode system requires the construction of an access network and the updating of the database, and the introduction and management costs increase. I do.

Therefore, in recent years, a dynamically variable two-dimensional matrix code has been proposed instead of the one-dimensional bar code having many problems described above.
That is, the two-dimensional code is displayed as a matrix of black and white cells, which is a computer language consisting of binary numbers 0 and 1.

Therefore, this two-dimensional code can dynamically expand and contract its matrix in accordance with the actually coded information. Also,
The two-dimensional code can be printed on the object by a computer printer, a label making machine, a laser beam, chemical etching, lithography, flexography, photoengraving, or the like. These printed two-dimensional codes can be easily read by a CCD camera.

In addition, the two-dimensional code can adopt a large and small size widely, and can not only encode one character to about 2,000 characters, but also read the two-dimensional code. There is an advantage that the error occurrence rate is extremely low.

[0011]

SUMMARY OF THE INVENTION However, the prior art
Various devices are known as means for reading a dimensional code. For example, a gun type laser scanner, a raster scan type gun type laser scanner, an area sensor type scanner or the like is used.

However, in the former gun type laser scanner, the operator must perform a reading operation while visually confirming that the direction and range of the laser beam match the direction and range of the two-dimensional code. However, there is a problem that the reading operation of the operator is troublesome and the reading efficiency is reduced.

The latter area sensor type scanner is
Although the configuration itself is capable of automatic reading, it has the advantage of eliminating the trouble of reading work by the operator, but the image of the two-dimensional code is often inclined and formed with respect to the element array direction of the area sensor. However, in such a case, there is a problem that the resolution is reduced substantially and it becomes difficult to read a code recorded at high density.

An object of the present invention is to provide a method and apparatus for reading a two-dimensional code which can efficiently read and decode a two-dimensional code efficiently with a good resolution by a relatively simple processing operation. It is in.

[0015]

In order to achieve the above object, a method for reading a two-dimensional code according to the present invention comprises extracting a candidate for the approximate center position of a two-dimensional code from image data containing the two-dimensional code, A first processing step of determining a required center position candidate group of the two-dimensional code, and a guideline of the two-dimensional code using the rough center position candidate group of the two-dimensional code extracted in the first processing step A second processing step of searching for and determining four corner coordinates of the two-dimensional code;
The corresponding coordinates of the matrix of the two-dimensional code are calculated from the four corner coordinates obtained in the second processing step, 0 or 1 binary data is determined from the pixel value, and the two-dimensional code of the two-dimensional code is determined from the matrix data. And a third processing step of performing a decoding process of converting the data into character data according to a standard.

In this case, in the first processing step, a concavo-convex function is created from the image data to measure the concavity and convexity of the pixel values in the X and Y directions of the image data, and then the filter length is calculated from the concavo-convexity function. The moving average is applied to perform smoothing processing of the image data by the smoothing function, and further, the coordinates of the conditional maximum value are extracted from the smoothing function to determine the center position candidate group of the two-dimensional code. be able to.

In the second processing step, a candidate group of line segments considered to be a guideline is obtained from one candidate coordinate of the approximate center position candidate group using a guide line and a quiet zone of the two-dimensional code. Then, four candidate coordinate groups are calculated from the guideline candidate group using the orthogonality of the guideline of the two-dimensional code, and some consistency is checked from the four corner candidate groups to remove candidate points, It can be configured to determine four corner coordinates.

On the other hand, a two-dimensional code reading apparatus according to the present invention for carrying out the above-mentioned method comprises means for taking in image data including a two-dimensional code by a CCD camera, and two-dimensional code conversion from the obtained image data. Means for extracting approximate center position candidates, means for determining four corner coordinates of a two-dimensional code using the extracted group of approximate center positions, and a matrix of two-dimensional code from the determined four corner coordinates Means for calculating coordinates corresponding to the pixel data, discriminating 0 or 1 binary data from the pixel values, and further performing decoding processing for converting the matrix data into character data in accordance with the two-dimensional code standard. And

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, a method for reading a two-dimensional code according to the present invention will be described in relation to an apparatus for implementing the method.
This will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic configuration diagram of an apparatus for implementing a two-dimensional code reading method according to the present invention. That is,
The device configuration shown in FIG. 1 basically includes a camera device unit 10,
It comprises an image processing unit 20 and a computer processing unit 30.

However, the camera unit 10 mounts a CCD camera 14 on a camera fixing jig 12 and
It has a configuration in which a lens 16 and a lighting device 18 are provided on the image input side of the D camera 14. The CCD camera 1
The operation power supply 4 and the power supply of the lighting fixture 18 are provided in the image processing device unit 20, respectively.

The image processing unit 20 includes an image processing unit 22 connected to input an image output from the CCD camera 14 via a signal line 19, and a CCD.
Camera power supply 24 and illumination power supply 26, CRT
And a monitor device 28 comprising a display.

The computer processing unit 30
It comprises a personal computer 34 connected to the image processing unit 20 via a download cable 32, and has a CRT display 36 attached.

Next, a decoding process of the two-dimensional code reader having the above-described configuration by the reader will be described.

In the camera unit 10, the image data of 256 grayscale levels captured by the CCD camera 14 is
In order to decode the two-dimensional code via the image processing unit 20 and the computer processing unit 30, the following three processes are executed.

First processing (processing for determining a center position candidate group)
As preprocessing, a candidate for the approximate center position of the two-dimensional code is extracted. The purpose of this process is to determine a group of center position coordinate candidates for the two-dimensional code. That is, the image data captured from the CCD camera 14 does not always include only the two-dimensional code, and includes various noises. Therefore, the center position candidate group is determined from such an image using the features of the two-dimensional code.

Second Processing (Cut-out Processing) The purpose of this processing is to determine four corner coordinates of a two-dimensional code using a group of approximate center position candidates. This process also employs a method of searching for a guideline of a two-dimensional code using characteristics of the two-dimensional code and calculating four corner coordinates therefrom.

Third process (decoding process)
The corresponding coordinates of the matrix of the two-dimensional code are calculated from the four corner coordinates obtained by the clipping process, and 0 is calculated from the pixel value.
Or, determine 1 binary data. The matrix data is converted into character data according to the two-dimensional code standard. At this time, if the decoding of the two-dimensional code is completely successful, the decoded character data is output. However, when the decoding is incomplete, the processing is started from the next approximate position coordinates.

FIG. 2 is a flowchart of the above three processes.
Shown in

Next, each of the above three processes will be described in detail.

1. In the process of this <br/> processing for determining the center position candidates, for example, as shown in FIG. 3, the image data taken-in CCD camera 14, some candidates schematic center position coordinates of the two-dimensional code It is to list. FIG.
, Four approximate center position candidate groups 1 to 4 are listed. In determining these approximate center position candidate groups, the following three processes are performed.

That is, as shown in FIG. 4, (1) unevenness degree measurement processing, (2) smoothing processing, and (3) local maximum value extraction processing are sequentially performed.

(1) Asperity degree measurement processing This asperity degree measurement processing is based on the image data Pj, k and the asperity degree functions Wx (j) and Wy (j) shown in the following equations (1-1) and (1-2). Is the process of creating In this processing, as shown in FIG. 5, focusing on the fact that the two-dimensional code has a pixel value change at a certain interval, the degree of unevenness of the pixel values in the X and Y directions of the image data is measured. It is. At this time, it is important that the number of pixels of one cell of the two-dimensional code (n-
1) Sampling of image data at intervals.

[0034]

(Equation 1)

Here, Wx (j) is an unevenness degree function in the X-axis direction,
Wy (j) is the unevenness function in the Y-axis direction, ABS (h) is a function that returns the absolute value of h, and Pj, k is the pixel value of the image data at the coordinates (j, k).

(2) Smoothing processing This smoothing processing is shown in FIG.
As shown in the above, the above-mentioned unevenness degree functions Wx (j) and Wy (j)
From above, moving average is performed with the filter length m, and the smoothing function is
This is a process calculated by (1-3) and (1-4). The filter length m represents the number of cells in a two-dimensional code matrix as shown in FIG.

[0037]

(Equation 2)

(3) Maximum value extraction processing This maximum value extraction processing is processing for extracting the coordinate (Xi, Yj) of the conditional maximum value from the smoothing function Fx (j) and Fy (j), respectively. It is. The condition here is that, as shown in FIG. 7, a local maximum value less than 1/2 of the maximum value is not extracted.
Therefore, as shown in FIG. 8, the approximate center position candidate is Fx
There are as many as the product of the number Nx of the maximum values extracted in (j) and the number Ny extracted in Fy (j).

[0039] 2. The clipping processing for clipping process is a process based on the general center position obtained as described above, to determine the four angular coordinates of the two-dimensional code. A flowchart of this process is as shown in FIG. 9. Therefore, in this process, the following three processes are performed.

That is, as shown in FIG. 9, (1) a guideline search, (2) orthogonality check, and (3) determination of four corner coordinates are sequentially performed.

(1) Guideline Search In this process, a line segment that is likely to be a guideline is extracted. Accordingly, in this case, a process of obtaining a candidate group of line segments that are considered to be guide lines from one candidate coordinate (Xc, Yc) of the approximate center position candidate group is performed using the guide line and the quiet zone of the two-dimensional code. . That is, FIG.
As shown in FIG. 0, pixel values are extracted along two lines (scan 1 and scan 2) that are centered on the coordinates (Xc, Yc) and are in contact with the basic circle having a radius of L / 2 in the X-axis direction. go. At this time, coordinates matching the edge condition are stored in a register.

As shown in FIG. 11, a threshold value (threshold value) is determined as to whether or not a line segment connecting the edge coordinates extracted in scan 1 and scan 2 is a line (thick black line). Then, the pixel value on the line segment is checked and checked. As a result, if the line is confirmed, it is added to the guideline candidate group.

This processing is performed from directions 1 to 8 shown in FIG. The line segments 1 to N of the guideline candidate group at this time are:
It is as follows.

Line 1 (X10, Y10), (X11, Y11)
Line segment 2 (X20, Y20), (X21, Y21) Line segment 3 (X
30, Y30), (X31, Y31) line segment n (Xn0, Yn0),
(Xn1, Yn1) line segment N (XN0, YN0), (XN1, YN
1)

FIG. 13 is a flowchart of this process.

(2) Checking the orthogonality In this process, the orthogonality of the line segment expressed as an equation is checked. Therefore, in this case, a process of calculating four candidate (coordinate) groups from the guideline candidate group described above using the orthogonality of the guideline of the two-dimensional code is performed. In this case, FIG.
, The angle θ between the extraction lines n and m is given by the following equation (2).
-1), (2-1.1), (2-1.2).

[0047]

(Equation 3)

According to the above equation (2-1), if θ is approximately 90 degrees, the intersection (Xcj, Ycj) is calculated by the following equation (2-c).
2) Calculate using (2-2.1), (2-2.2), (2-2.3), and (2-2.4).

[0049]

(Equation 4)

The result obtained by the above equation (2-2) is added to four corner candidate groups based on the following 1 to N points (coordinates).

Point 1 (Xc1, Yc1) Point 2 (Xc2, Yc2) Point 3 (Xc3, Yc3) Point n (Xcn, Ycn) Point N (XcN, YcN)

FIG. 15 is a flowchart of this process.

(3) Determination of Four Corner Coordinates In this process, the consistency is confirmed and the coordinates are predicted. That is, in this case, some consistency is checked from the four corner candidate groups described above, candidate points are eliminated, and a process of determining four corner coordinates is performed. Here, for the consistency, an angle check and a distance check are applied.

First, as shown in FIG. 16, the angle check is a process of eliminating candidate points by utilizing the fact that the angle from the center position does not become 45 degrees or less if there are four angles. In this case, for example, points 2 and 4 are dropped as shown in FIG.

Next, the distance check is a process of sifting out candidate points using the fact that the distances of the four corners are constant values.

The candidate points that survive the above consistency check are the following four corner coordinates.

Point 1 (Xc1, Yc1) Point 2 (Xc2, Yc2) Point 3 (Xc3, Yc3) Point 4 (Xc4, Yc4)

If less than four candidate points remain, the remaining points are obtained by calculation.

If the flowchart of these processes is shown,
This is as shown in FIG.

[0060] 3. Regarding the decoding process, the corresponding coordinates of the matrix of the two-dimensional code are calculated from the four corner coordinates obtained in the above-described clipping process, 0 or 1 binary data is determined from the pixel value, and 2 By performing the decoding process of converting the data into character data in accordance with the standard of the dimensional code, appropriate decoding of the two-dimensional code can be easily and quickly achieved.

Although the preferred embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various design changes can be made without departing from the spirit of the present invention. it can.

[0062]

As described above, the method and apparatus for reading a two-dimensional code according to the present invention extracts the approximate center position candidate of the two-dimensional code from the image data including the two-dimensional code, and obtains the required two-dimensional code. A first processing step of determining a group of candidate center positions of the code; and a two-dimensional code using the group of approximate center positions of the two-dimensional code extracted in the first processing step.
Search the guideline of the two-dimensional code and
A second processing step of determining one corner coordinate, and calculating a corresponding coordinate of a matrix of a two-dimensional code from the four corner coordinates obtained in the second processing step, and calculating 0 or 1 binary from the pixel value. A third processing step of discriminating data and converting the matrix data into character data in accordance with the standard of a two-dimensional code; The processing can be realized efficiently and accurately with a good resolution by a relatively simple processing operation.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of an apparatus for implementing a two-dimensional code reading method according to the present invention.

FIG. 2 is a flowchart of a program for executing a two-dimensional code reading method according to the present invention.

FIG. 3 is an explanatory diagram showing an example of extracting a group of approximate center position candidates in the two-dimensional code reading method according to the present invention.

FIG. 4 is a flowchart of a program for extracting a general center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 5 is an explanatory diagram showing a relationship between cells of a two-dimensional code and the number of pixels of image data in the two-dimensional code reading method according to the present invention.

FIG. 6 is a waveform diagram showing a smoothing process of image data extracted from a group of approximate center position candidates in the two-dimensional code reading method according to the present invention.

FIG. 7 is a waveform diagram showing a local maximum value extraction process after smoothing the extracted image data of the approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 8 is an explanatory diagram showing position coordinates of an extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 9 is a flowchart of a program for extracting a position coordinate of an extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 10 is an explanatory diagram showing a method of extracting an edge of a guideline in a process of extracting position coordinates of an extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 11 is an explanatory diagram showing a method of performing a line check between extracted edges in a process of extracting position coordinates of an extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 12 is an explanatory diagram showing a method of performing a guideline search in a process of extracting position coordinates of an extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 13 is a flowchart of a program for performing a guideline search at the time of extracting position coordinates of the extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 14 is an explanatory diagram showing angles of two guideline candidate groups in a guideline search in a process of extracting position coordinates of the extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 15 is a flowchart of a program for checking the orthogonality of the guideline in the process of extracting the position coordinates of the extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 16 is an explanatory diagram showing angles of four corner candidate points as position coordinates of an extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 17 is an explanatory diagram showing the sieving by the angles of four corner candidate points as the position coordinates of the extracted approximate center position candidate group in the two-dimensional code reading method according to the present invention.

FIG. 18 is a flowchart of a program for determining four corner coordinates as position coordinates of an extracted group of approximate center position candidates in the two-dimensional code reading method according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Camera apparatus part 12 Camera fixing jig 14 CCD camera 16 Lens 18 Lighting equipment 20 Image processing apparatus part 22 Image processing apparatus 24 Camera power supply part 26 Illumination power supply part 28 Monitoring device 30 Computer processing part 32 Download cable 34 Personal computer 36 CRT display

Claims (4)

(57) [Claims] A first processing step of extracting an approximate center position candidate of the two-dimensional code from image data including the two-dimensional code, and determining a required group of center position candidates of the two-dimensional code; A second processing step of searching for a guideline of the two-dimensional code and determining four corner coordinates of the two-dimensional code by using the group of approximate center position candidates of the two-dimensional code extracted in the processing step; The corresponding coordinates of the matrix of the two-dimensional code are calculated from the four corner coordinates obtained in the second processing step, 0 or 1 binary data is determined from the pixel value, and the standard of the two-dimensional code is determined from the matrix data. And a third processing step of performing a decoding process of converting the data into character data according to the following. 2. A first processing step comprises: forming a concavo-convex function from image data, measuring the concavity and convexity of pixel values in the X and Y directions of the image data, and then performing a moving average with a filter length from the concavity and convexity function Then, a smoothing function is performed on the image data by the smoothing function, and the coordinates of the conditional maximum value are extracted from the smoothing function to determine a center position candidate group of the two-dimensional code. The method for reading a two-dimensional code according to claim 1. A second processing step of obtaining, from one candidate coordinate of the approximate center position candidate group, a candidate group of line segments considered as a guideline using the guideline and the quiet zone of the two-dimensional code; Next, four candidate coordinate groups are calculated from the guideline candidate group using the orthogonality of the guideline of the two-dimensional code, and some consistency is checked from the four corner candidate groups, and candidate points are eliminated. 2. The method for reading a two-dimensional code according to claim 1, wherein the method is configured to determine two corner coordinates. 4. A means for capturing image data containing a two-dimensional code by a CCD camera,
Means for extracting approximate center position candidates of the dimensional code, means for determining four corner coordinates of the two-dimensional code using the extracted group of approximate center position candidates, and two-dimensional coordinates from the determined four corner coordinates Means for calculating coordinates corresponding to the code matrix, determining 0 or 1 binary data from the pixel values, and further performing decoding processing for converting the matrix data into character data in accordance with the two-dimensional code standard. An apparatus for reading a two-dimensional code, characterized in that: