JP2002150213A - Bar code reader - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a bar code reader having high reliability for accurately detecting and decoding a bar code located at an optional position in a document image having various background patterns at a high speed. SOLUTION: An object is read by an image pickup means 1, and an image binarized by a binarization means 2 is stored in an image storage means 3. A bar code candidate region on the binarized image is detected by a region extracting means 4, the binarized image is scanned by a scanning means 5 along a plurality of scanning lines parallel with the reading axis, with the regression straight line of the pattern included in the candidate region used as the reading axis, and the bar code is decoded by a decoding means 6 from a data train of bars and spaces obtained by scanning.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for specifying the position and direction of a bar code, reading and decoding the bar code in a document image including characters, figures, patterns, and bar codes such as postal matter. The present invention relates to a bar code reader.

[0002]

2. Description of the Related Art Barcodes are used in a wide range of fields as a means for displaying management information of articles in product sales, distribution, production processes and the like. As a method of reading such a barcode, a method in which an operator directly instructs a barcode surface to read a barcode, such as a hand scanner or a pen input scanner, or a fixed scanner, in which a beam is scanned in a plurality of directions, at least, It is known that one beam reads a barcode by scanning the barcode surface correctly. There is also a method of inputting an object as two-dimensional image data, extracting a barcode area included in the image by image processing, and reading and decoding the barcode area. This is Tominaga, et al., "A Method for Reading Barcodes in Document Images", IEICE Spring Convention, 1991,
As described in D-558, a region where the length, direction, and adjacent distance of an outline vector in an image are uniform is extracted and read as a barcode region, and is limited to the position and direction of the barcode. Also, even if a plurality of barcodes exist in the object, all barcodes can be decoded.

[0003]

However, in the above-mentioned conventional method, it is necessary to scan the entire screen of the binary image in advance, track all the contours and convert them to vector data, and the background area becomes complicated. Accordingly, there has been a problem that the processing time increases drastically, and it is difficult to select a binarization threshold value for an object having uneven density, and it becomes difficult to maintain reading accuracy.

The present invention solves the above-mentioned problems of the prior art, and provides a bar code reading apparatus capable of processing at high speed even if various character patterns exist in a background area of a bar code existing at an arbitrary position. The purpose is to: It is another object of the present invention to provide a highly reliable barcode reading device capable of correctly decoding even if the barcode has uneven density.

[0005]

In order to solve the above-mentioned problems, a bar code reading apparatus according to the present invention comprises: an image pickup means for scanning and decomposing a document including a bar code to convert the document into a gray image; A binarizing means for converting into a value image;
Image storing means for storing the binary image, area extracting means for detecting circumscribed rectangle information of a barcode candidate area, and calculating a reading axis of a barcode included in the circumscribed rectangle; Scanning means for scanning a binary image along a plurality of scanning lines; and decoding means for decoding a bar code from a data string of bars and spaces obtained by the scanning, and a bar code from the entire image. By narrowing down the candidate areas based on the rectangular features of the code and scanning and decoding the data for each candidate area, even if various character patterns exist in the background area of the barcode at an arbitrary position, it can be processed at high speed In addition, even if the bar code has uneven density, the bar code can be decoded correctly.

In the bar code reading apparatus of the present invention, the binarizing means scans the grayscale image with an m × m window,
An average value of pixels in a window including a pixel of interest is calculated, and compared with a value obtained by adding a predetermined offset value to the average value, thereby performing binarization. Even in this case, since the threshold value changes in accordance with this, the bar code can be accurately binarized without crushing or blurring.

Further, in the bar code reading apparatus according to the present invention, the area extracting means converts the binary image into one in the x and y directions.
/ K, a labeling means for converting each image on the binary image into a unique label number, and a plurality of labels adjacent in the x and y directions. Label integration means for generating a group, and area determination means for detecting a barcode area from the characteristic information of the group, characterized in that even if the background of the barcode becomes complicated, By performing the labeling and the determination processing, the bar code area can be detected at high speed.

In the bar code reading apparatus of the present invention, the blocking means divides the binary image into mesh regions each having k × k pixels, and calculates the sum of the binary data in the mesh regions. A pixel value at a corresponding position on the reduced image is compared with a value obtained by adding a predetermined offset value to an average value of non-zero pixels in the reduced image, and is binarized. Even if the luminance changes, an accurate reduced image can be obtained, and the barcode candidate region can be reliably detected.

In the bar code reading apparatus according to the present invention, the area determining means selects a bar code candidate area from a size of a circumscribed rectangle of the label group and a total area of constituent labels. This has the effect that the candidate area can be detected without omission by simple determination processing.

In the bar code reading apparatus of the present invention, the scanning means labels the pattern in a circumscribed rectangle of the bar code candidate area, and integrates labels among labels adjacent within a predetermined distance. A label in which sides of a circumscribed rectangle overlap in a direction orthogonal to the direction is collected, and a barcode reading axis is set for the collected label, and even if noise exists near the barcode, Has the effect that decoding can be performed accurately.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a block connection diagram of a bar code reader according to an embodiment of the present invention. The barcode reading device according to the present embodiment includes:
An imaging unit 1 for imaging a document including a barcode and converting the image into a gray image; and converting the obtained gray image into a binary image 2
Value converting means 2, image storing means 3 for storing the obtained binary image, area extracting means 4 for detecting rectangular information of the bar code candidate area, and a bar code reading axis is calculated and parallel to the reading axis. Scanning means 5 for scanning a binary image along a plurality of scanning lines, and decoding means 6 for decoding a bar code from a data string of bars and spaces obtained by scanning.

The operation of the bar code reading apparatus configured as described above will be specifically described below. The document including the barcode is read by the imaging unit 1,
It is converted to a quantized multi-tone image. In this embodiment, an image scanner using a CCD line sensor is used, the number of quantization bits is 8, and the scan density is 8 lines / mm. For example, when reading a postal matter on which the customer barcode shown in FIG. 2 is printed, the standard barcode is a bar length W = 640 pixels, a long bar height h1 = 29 pixels, a semi-long bar height h2 = 19 pixels, The image is captured with the timing bar height h3 = 10 pixels, the bar width w1 = the space w2 = 5 pixels, and the bar pitch p = 10 pixels.

The grayscale image obtained by the imaging means 1 is 2
The image is converted into a binary image by the binarizing unit 2. FIG. 3 shows a scanning window for performing threshold processing in the binarizing means 2, which is composed of m × m pixels centered on the pixel of interest. The grayscale image input through this scanning window is raster-scanned, and (Equation 1)
The target pixel is binarized under the following condition. Assuming that the value of the target pixel indicated by a circle in FIG. 3 is Tin, the pixel value in the window is dij (i, j = 1 to m), and the offset value is g, the binary output Tout of the target pixel becomes Become like

(Equation 1)

The above processing is performed by hardware in synchronization with the line synchronization of the scanner and the pixel clock, so that
A binary image is generated in synchronization with the capture of the grayscale image, and stored in the image storage unit 3. Thus, the target pixel and mx
The reason why the binarization is performed by comparing the average value in the vicinity of m is that, when there is density unevenness in the printing of the barcode, barring or space collapse occurs at a fixed threshold value. Even if the amplitude of the luminance of the space fluctuates, a threshold value suitable for the surrounding luminance distribution is set by calculating the average value, and the relationship between the bar and the space required for decoding is maintained, and the binary value is obtained. Be transformed into In this embodiment, the scanning window size is a width that covers one bar and one space in the window, and m = 11. The offset value g is a value that is empirically determined according to the luminance distribution of the imaging target, and is set to a value that is about 1/10 of the luminance of the ground of the target object as a guide.

Next, the area extracting means 4 will be described with reference to FIGS. FIG. 4 shows a processing block configuration of the area extracting means 4, in which a binary image is converted by one in the x and y directions.
/ K, a blocking means 41 for converting the image into a reduced image, a labeling means 42 for assigning an individual label number to each pattern on the binary image, and a plurality of labels adjacent in the x and y directions. It comprises a label integrating means 43 for generating a group and an area determining means 44 for detecting a barcode area from the characteristic information of the label group.

The blocking means 41 first includes the binarizing means 2
Is divided into mesh regions in units of k × k pixels, and a gradation image in which the sum of the binary data in the mesh region is a representative value of the mesh is generated as an intermediate image. That is, an image (gradation number k × k) reduced to 1 / k in the x and y directions is formed. Next, an average value Cv of non-zero pixels in the intermediate image is calculated, and a value obtained by subtracting an appropriate offset value f from Cv is binarized as a threshold value. That is, the multivalued reduced image is converted into the binary reduced image under the condition of (Equation 2).

(Equation 2) In the present embodiment, k = 8. As shown in FIG. 5, the output image 5
1 is converted to a reduced image 52. 2 multi-tone reduced images
The reason why the threshold value is set based on the average luminance value of the entire reduced image when the value is converted is to prevent the barcode or the character symbol on the object from disappearing or being connected with the fixed threshold value.

Next, in the labeling means 42, individual labels are attached to individual patterns on the binarized reduced image. The labeling processing procedure will be described below with reference to FIGS. FIG. 6 is a flowchart showing a labeling processing procedure, FIG. 7 is a structure table for registering data of a label area, and FIG. 8 is a view for explaining a connection relationship of run data. In FIG. 6, a line counter y = 1 in the sub-scanning direction is set in step 601.
At 02, run data of the y-th line is obtained from the image memory. In the binarized image, it is assumed that the pattern portion has the value “1” and the background portion has the value “0”, and the x coordinate Rs of the start point and the x coordinate Re of the end point where the value “1” continues are taken in.
The following judgments are made. In the determination of step 603,
A label table having the components shown in FIG. 7A is searched, and among the already registered labels, the valid flag flag is searched.
Is 1, and the y coordinate LastY of the latest run belonging to the label
Picks up the label area that is the line immediately before the current line y, and proceeds to step 605. Step 60
5 is a step of determining whether or not the run of interest is connected to the label area. The run of the latest line belonging to the label area (Xs [0], Xe [0]), (Xs [1 ], Xe
[1])... Are checked by the comparison processing of (Equation 3).

FIG. 8A shows a target run 802 and a front line 80.
It shows the relationship that one run is connected,

(Equation 3) When satisfies, it is determined that the target run and its label area are connected, and the process proceeds to step 606 to update the data. In step 606, the circumscribed rectangular coordinates are updated by adding the run length of the target run to the area s of the label area. FIG. 7B shows the coordinates of the circumscribed rectangle, and the maximum value of the x coordinate of the label area.
Xmax and the minimum value Xmin of the x-coordinate are compared with the start point Rs and the end point Re of the target run, and the coordinate values are updated. Also, the maximum value Ymax of the y coordinate and the minimum value Ymin of the y coordinate of the label area
Then, the coordinate y of the target run is compared in magnitude and the coordinate value is updated.
Then, the center coordinates Cx and Cy of the label area are updated using the updated circumscribed rectangular coordinates. Cx is the average of Xmax and Xmin, and Cy is the average of Ymax and Ymin.

As shown in FIG. 8B, the noticeable run 804 is
One label area 803 and another label area 805
Are linked to both the label area 805 and the label area 805, the validity flag of the label area 805 is
The flag is set to 0 and the label is invalid. Step 607 is a step of judging whether or not the connection check has been performed for all the registered label areas of the target run. If so, the process proceeds to step 608. If not, the process proceeds to the next label area. A similar connection determination process is performed. Step 608 is a step for determining whether or not the target run is connected to any of the existing label areas. If the target run is connected, the process proceeds to step 610. If not, the process proceeds to step 609. The run is registered as a new label area, and the flag is made valid. In step 610, it is determined whether or not the connection determination has been completed for all runs of the current line. If not completed, the process is repeated from step 602. If completed, the current line y
Update the y-coordinate LastY of the latest run of all the label areas connected to the upper run to y, and run RunCn of each label area.
The t and the coordinates Xs and Xe of the run are updated, and the process proceeds to step 611. In step 611, it is determined whether or not the processing has been completed for all the lines of the image. If not, the line counter is updated in step 612 and the processing is repeated from step 602. . In this way, in the label table, the area where the valid flag flag is 1 is detected as an independent pattern included in the binary image 52.

Next, the label integrating means 43 will be described with reference to FIGS. FIG. 9 shows the label integrating means 43.
And FIG. 10 is a diagram showing a structure table for registering a label group. The label integrating means 43 applies an x direction to each label on the reduced image,
And a group adjacent to each other in the y direction are integrated to generate a label group. First, a procedure for integrating labels in the x direction while referring to label information will be described. In FIG. 9, one piece of label information is acquired from a label table in step 901 and it is determined to which group the label belongs by the following processing. In step 902, the valid flag flag of the label is checked.
The process proceeds to step 901 if invalid, and returns to step 901 to acquire label information of the next ID. In step 903, labels with a small area less than Smin are excluded from the integration processing as noise. If the area condition is satisfied, the process proceeds to step 904. If not, the process returns to step 901 to acquire the next ID label information. I do. In the present embodiment, Smin = 2 pixels. Step 904 is a step of determining whether or not the label of interest is adjacent to the existing group in the x direction. Referring to mem label information of the group structure in FIG. 10, LID [0] to LID [mem- Look at the positional relationship with all the labels indicated by [1]. Set the circumscribed coordinates of the label of interest to the starting point (Tsx,
Tsy), end point (Tex, Tey) If the circumscribed coordinates of the labels that are the partners of the existing group are the start point (sx, sy) and the end point (ex, ey), first, the labels to be integrated by (Equation 4) It is determined that the lengths of the sides are substantially the same. If the tolerance on the y side of the label is DIFF,

(Equation 4) When satisfying the following (Equation 5) under the above-described (Equation 4), it is determined that the labels are arranged in a line in the x direction. If the distance to be integrated is DIST,

(Equation 5) Becomes In this embodiment, the tolerance DIFF of the side of the label is within 6 pixels, and the distance DIST of label integration is within 6 pixels.
When any of the labels from LID [0] to LID [mem-1] meets the above integration condition, the label of interest is included in the group.

Next, in step 905, all groups to which the label of interest belongs are integrated to generate one group. At this time, the area SG of the group of the group structure in FIG. 10, the circumscribed coordinates, the number of members mem, and the label ID of the member are updated and added. If it is determined in step 906 that the processing has not been completed for all existing groups, the next group data is obtained, and the processing in steps 904 and 905 is performed. If it is determined in step 908 that the label of interest is not integrated with any group, the process proceeds to step 909, where the label of interest is assigned to a new GID as a new label group, registered in the group structure, and the group type is set to 0,
Set the group direction x. In step 910, it is determined whether or not the processing has been completed for all valid labels on the label table.
The process is repeated from 1 and when completed, the label integration process in the x direction is completed. The label integration process in the x direction is also performed for the y direction. In this case, (Equation 4) and (Equation 5)
Exchange the suffixes x and y of As a result, label groups in the x and y directions are obtained.
In the case of the object of FIG.
The area indicated by the broken line in (a) is obtained as a group in the x direction, and the area indicated by a broken line in FIG.

Next, the area determining means 44 will be described with reference to FIGS. With reference to the group structure table of FIG. 10, a customer barcode area is selected for all the groups obtained by the label integrating means under the condition of (Formula 6) or (Formula 7). If the margin is Δw for the width W of the barcode area and Δh is the margin for the height h1 of the barcode area,

(Equation 6) Or

(Equation 7) Becomes In this embodiment, W = 640 pixels, h1 = 29 pixels, Δw = 10 pixels, and Δh = 5 pixels. As a result, of the label groups indicated by broken lines in FIG. It is extracted as a code candidate area.

Next, the scanning means 5 and the decoding means 6 will be described with reference to FIGS. FIG.
FIG. 13 is a flowchart showing the processing procedure of the scanning unit and the decoding unit, and FIG. 13 is a diagram showing the integration relationship of labels. FIG.
In step 121, the starting point (GXmin, GYmi) of the rectangular area determined as a barcode candidate by the area determining means 44 in step 121
n) and the end points (GXmax, GYmax),
As shown in FIG. 31, the start point 132 (x1, y1) and the end point 133 (x2, y2) of the barcode candidate area on the original image are defined by (Equation 8).
Ask by

(Equation 8)

Next, in step 122, the pattern in the area 131 is labeled, and the barycentric coordinates and the circumscribed rectangular coordinates of each label are obtained. The labeling process is the same as that of the labeling means 42, and the description is omitted. Next, in step 123, it is determined whether the direction attribute of the region of interest is the X direction or the Y direction. If the direction is the X direction, the process proceeds to 124, and the labels are integrated in the X direction. Integrate labels into. For example, in FIG. 13, when the labels are integrated in the X direction, the starting point (s
x, sy) If the end point is (ex, ey), as shown in (Equation 9), the labels are within a predetermined pitch p pixels, and

(Equation 9) In addition, as shown in (Equation 10), the condition is that the start point and the end point of each label in the Y direction overlap each other.

(Equation 10) In the case of FIG. 13, after the labels 11 and 12 are integrated, the labels 13 and 14 satisfy (Equation 9) and rise as candidates.
According to (Equation 10), the label 13 in which the start point and the end point in the Y direction overlap the label 12 is selected as an integration target. Thereafter, similarly, labels 15 and 17 belong to the same group, and labels 14 → 16 → 18 are integrated as another group. As a result, the barcode area 134 in the barcode candidate area 131 is separated and extracted from the other area 135. Next, in step 126, using the barycentric coordinates (x, y) of the n labels in each barcode area 134, the coefficients a and b of the reading axis Y = aX + b are obtained according to the least square method of (Equation 11).

[Equation 11]

If it is determined in step 123 that the direction of label integration is the Y direction, the read axis X = aY + b
Are obtained according to (Equation 12).

(Equation 12) A plurality of scanning lines are drawn along the read axis thus determined, and binary data on the scanning lines are sequentially picked up to generate a bar and space data string. Step 127
Converts a bar and space data string into characters according to a predetermined conversion rule, and decodes a barcode. At step 128, it is checked whether or not decoding attempts have been completed for all barcode candidate areas, and if there are still candidate areas, the processing from step 121 is repeated. With the above processing, even if a plurality of barcodes exist at an arbitrary position in the document image, all the barcodes can be correctly decoded.

As described above, according to the present embodiment, first, the grayscale image is binarized and stored, and the candidate area is narrowed down from the entire binary image based on the rectangular feature of the barcode. By detecting a group of bars included in the data, determining a reading axis, and decoding a bar code from a data string obtained by scanning in parallel along the axis, a plurality of bar codes existing at arbitrary positions can be decoded. The effect is obtained.

Second, in order to extract a candidate area, the binary image is reduced to 1 / k in the x and y directions, labeling processing is performed on each pattern on the reduced binary image, and x and y By generating labels by integrating labels adjacent to each other in the direction and detecting the barcode area from the feature information of this group, the barcode area can be detected at high speed even if the background of the barcode is complicated. can get.

Third, the input gray-scale image is scanned in an m × m window, and a binary value is obtained by adding a predetermined offset value to the average value of the pixels in the scanning window as a threshold, thereby performing localization. Even if there is a large density fluctuation, the threshold value changes in accordance with the fluctuation, so that an effect is obtained that the barcode can be accurately binarized without crushing or blurring.

Fourth, the binary image is divided into meshes in units of k × k regions, and the sum of the binary data in the mesh regions is used as the representative pixel value of the mesh. By binarizing the entire reduced image with a value obtained by adding a predetermined offset value to the average value of the pixels, the bar code candidate area can be reliably detected even if the luminance of the object changes.

Fifth, by selecting a barcode candidate area from the size of the circumscribed rectangle of the label group on the reduced image and the total area of the constituent labels, an effect is obtained that the candidate area can be detected without fail.

Sixth, labeling is performed on the pattern in the bar code candidate area, and among labels adjacent to each other within a certain distance, a label whose sides of the circumscribed rectangle overlap in a direction orthogonal to the direction in which the labels are integrated is determined. By collecting the labels and determining the read axis for the collected labels, an effect is obtained in that even if noise exists near the barcode, decoding can be performed accurately.

[0032]

As described above, according to the bar code reading apparatus of the present invention, the input image is binarized by the threshold value according to the local density distribution, and the binary image is reduced to 1 / k vertically and horizontally. Labeling is performed on the reduced image, barcode candidate areas are narrowed down from the characteristics of each label group, and for this barcode candidate area, only the adjacent bar labels are re-integrated by referring to the original size binary image. By obtaining the reading axis, tracing the binary image along the reading axis, and decoding the binary image, even if the background of the barcode area at an arbitrary position in the document image is complicated, A highly reliable barcode reading device capable of detecting a barcode region at high speed and accurately decoding without being affected by density unevenness or a nearby pattern can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram of a barcode reading device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a barcode to be read by the barcode reader according to the embodiment;

FIG. 3 is a diagram showing a scanning window of a binarizing unit in the barcode reading apparatus according to the embodiment;

FIG. 4 is a block configuration diagram of an area extracting unit in the barcode reading apparatus according to the embodiment;

FIG. 5 is an exemplary view showing a processing result of a blocking unit in the barcode reader according to the embodiment;

FIG. 6 is a flowchart showing a processing procedure of a labeling unit in the barcode reader according to the embodiment;

FIG. 7 is a view showing a structure table for registering label information of a labeling unit in the barcode reading apparatus according to the embodiment;

FIG. 8 is an exemplary view for explaining connection determination of run data by a labeling unit in the barcode reading apparatus according to the embodiment;

FIG. 9 is a flowchart showing a processing procedure of a label integrating means in the barcode reading apparatus of the embodiment.

FIG. 10 is a view showing a structure table for registering group information of a label integrating means in the barcode reading apparatus according to the embodiment;

FIG. 11 is a view showing a processing result of the label integrating means in the barcode reading apparatus of the embodiment.

FIG. 12 is a flowchart showing a processing procedure of a scanning unit and a decoding unit in the barcode reading apparatus of the embodiment.

FIG. 13 is a view for explaining a label integration relationship of scanning means in the barcode reading device of the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Imaging means 2 Binarization means 3 Image storage means 4 Area extraction means 5 Scanning means 6 Decoding means 41 Blocking means 42 Labeling means 43 Label integration means 44 Area judgment means 51 Binarized image by binarization means 52 Reduced image by blocking means 111 Barcode candidate area 112 Barcode candidate area

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 3F079 AA01 CA06 CB28 CB35 5B072 AA01 AA02 BB00 CC02 CC24 DD02 DD15 DD23 FF02 FF03 FF04 LL19

Claims (6)

[Claims] 1. An image pickup means for scanning and decomposing a document including a barcode into a grayscale image, a binarizing means for converting the grayscale image into a binary image, and an image storing means for storing the binary image. Area extracting means for detecting circumscribed rectangle information of a barcode candidate area; calculating a reading axis of a barcode included in the circumscribed rectangle; and forming a binary image along a plurality of scanning lines parallel to the reading axis. A bar code reading device comprising: scanning means for scanning a bar code; and decoding means for decoding a bar code from a data string of bars and spaces obtained by the scanning. 2. The binarizing means scans a grayscale image with an m × m window, and calculates an average value of pixels in a window including a pixel of interest.
2. The barcode reading device according to claim 1, wherein the binarization is performed by comparing a value obtained by adding a predetermined offset value to the average value. 3. An area extracting means for converting a binary image into a reduced image of 1 / k in x and y directions, and assigning a unique label number to each pattern on the binary image. Labeling means, label integrating means for integrating a plurality of labels adjacent in the x direction and y direction to generate a label group, and area determining means for detecting a barcode area from characteristic information of the group The bar code reader according to claim 1 or 2, wherein: 4. The blocking means converts the binary image to k ×
k is divided into mesh regions in units of k pixels, the sum of the binary data in the mesh region is defined as a pixel value at a corresponding position on the reduced image, and a predetermined offset value is added to an average value of non-zero pixels in the reduced image. Is compared with the value obtained by adding
4. The bar code reader according to claim 3, wherein the bar code is converted into a value. 5. The barcode reading device according to claim 3, wherein the area determination means selects a barcode candidate area from a size of a circumscribed rectangle of the label group and a total area of constituent labels. 6. The scanning means performs labeling on a pattern in a circumscribed rectangle of a barcode candidate area, and forms a circumscribed rectangle in a direction orthogonal to a direction in which labels are integrated among labels adjacent within a predetermined distance. 3. The barcode reading apparatus according to claim 1, wherein labels whose sides overlap each other are collected, and a barcode reading axis is set for the collected labels.