JP4100273B2 - Two-dimensional code reading method and two-dimensional code reading apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
  The present invention relates to a technique for imaging a two-dimensional code in which predetermined information is encoded, performing image processing on the two-dimensional code in the generated image, and reading the content of information displayed by the two-dimensional code. . In this specification, visually recognizable information represented by a two-dimensional code (generally appearing as an array pattern of white and black cells) is referred to as “optical information”.
[0002]
[Prior art]
  The two-dimensional code is a code in which white or black small areas called “cells” are two-dimensionally arranged. Each cell represents the minimum unit of optical information. In addition to the cells that make up specific information, the cells that make up the finder pattern for identifying the position and orientation of the two-dimensional code, and the cells in the code The cell which comprises the timing pattern for carving and identifying is included.
[0003]
  FIGS. 8A and 8B show a data matrix and a QR code, which are representative two-dimensional codes, respectively. In the data matrix, an L-shaped finder pattern 40 indicating the code direction is formed on two orthogonal sides, and an L-shaped timing pattern 41 is formed on the other two orthogonal sides. The finder pattern 40 is configured by arranging black cells, and the timing pattern 41 has a configuration in which white cells and black cells are alternately arranged (hereinafter, white cells are referred to as “white cells”, and black cells are referred to as “white cells”. Black cell ").
[0004]
  On the other hand, the QR code is provided with a finder pattern 42 composed of three rectangular positioning symbols 42a, 42b and 42c. Each positioning symbol 42a, 42b, 42c is formed by arranging a white cell frame around a rectangular pattern of a plurality of black cells, and further arranging a black cell frame outside thereof. These positioning symbols 42a, 42b, and 42c are respectively arranged at positions corresponding to the vertices of the code, and white cells and black cells are alternately arranged so as to connect corners inside the respective symbols 42a, 42b, and 42c. The L-shaped timing pattern 43 thus formed is formed.
  In both the data matrix and the QR code, the relationship between the white cell and the black cell shown in the figure may be reversed depending on the background state.
[0005]
  When reading the above two-dimensional code, first, the finder patterns 40 and 42 on the image are extracted by image processing such as edge extraction processing or pattern matching processing to recognize the position and orientation of the code. Next, timing patterns 41 and 43 are extracted based on the extracted finder patterns 40 and 42.
[0006]
  The area surrounded by the timing patterns 41 and 43 is an area where specific contents of optical information are encoded (hereinafter, this area is referred to as “information display area”). When the timing patterns 41 and 43 are extracted, the position of each cell in the information display area is specified based on the arrangement of white cells and black cells in the pattern. Then, it is determined whether each cell is black and white by binarization processing or the like, and optical information represented by each cell is encoded according to the determination result. Furthermore, the optical information is decoded by arranging the code of each cell for each group of cells (symbol character) representing information for one unit such as characters. In this decoding, a process (error correction process) for correcting an error that has occurred when determining the monochrome state of a cell is also executed. Hereinafter, the process from the extraction of the timing pattern to the completion of the decoding of the optical information is referred to as “decoding process”.
[0007]
  In the above-described reading process, only one finder pattern should be originally extracted. However, if there is noise (such as a pattern or shadow) similar to the shape of the finder pattern in the background of the two-dimensional code. Noise may be extracted as a finder pattern candidate. In order to cope with such noise, the conventional two-dimensional code reader performs a decoding process for each extracted finder pattern candidate, and when one of the finder patterns is successfully decoded, the decoding is performed. Read data indicating the contents of the information is output.
[0008]
[Problems to be solved by the invention]
  In the reading process using the conventional two-dimensional code reading apparatus, the reading process is started after adjusting the installation position of the apparatus, the illumination state, the work feeding position with respect to the field of view of the camera, and the like. However, if these adjustments are poor, optical information reading may fail (hereinafter, this failure is referred to as “reading failure”). Even if the adjustment is appropriate, reading of the optical information may fail because the marking state of the two-dimensional code on the workpiece is bad or the noise is in the background of the two-dimensional code.
[0009]
  However, in the conventional reading process, even if a reading failure occurs, there is no information indicating the cause of the failure, so the user has to investigate the cause by trial and error. For this reason, there is a problem that the setting at the time of introduction takes time and the start of full-scale reading processing is delayed.
[0010]
  Patent Document 1 below describes that in the optical information reading process, when the decoding process fails, the cause is analyzed, and the cause of the reading error and the coping method are reported based on the analysis result. Yes.
[0011]
[Patent Document 1]
JP 2002-236870 A
              (See paragraphs [0008] and [0009].)
[0012]
  In the above-mentioned Patent Document 1, a barcode reading process is given as a specific example of the reading process, and the cause of the reading error is analyzed based on the overall voltage level (waveform) of the image, the bar width, the left and right margins of the code, and the like. (Refer to paragraphs [0025] to [0031].) However, for the two-dimensional code, paragraph [0039] only describes that the same method as the barcode can be applied, and there is nothing about a specific processing method for reading failure of the two-dimensional code. Not shown.
[0013]
  An object of the present invention is to make it possible to determine in detail the cause of the reading failure when the reading process for the two-dimensional code fails and to notify the user of the determination result.
[0014]
[Means for Solving the Problems]
  The two-dimensional code reading method according to the present invention includes:Consists of two-dimensional array of two types of cells with different display modesImage with 2D code as imaging targetTo form a plurality of cells respectively located outside the information display area in the two-dimensional code.Extract specific patternsAndSpecific patterns extractedBy specifying the information display area based on the image and performing a decoding process on the area,Reads optical information represented by a two-dimensional codeIs.
[0015]
  In the above methodInA pattern called a finder pattern in the existing two-dimensional code can be a specific pattern. However, the specific pattern is not limited to the finder pattern. For example, a combination of the finder pattern and the timing pattern may be the specific pattern.
[0016]
  As an example of image processing for extracting a specific pattern, pattern matching processing using a model of the specific pattern can be given. Further, if the specific pattern is a linear pattern or a pattern including a portion in which black cells or white cells are arranged in a straight line, after extracting the edge points included in the image, the density is determined for each extracted edge point. It is also possible to adopt a method in which a line segment having a predetermined length is set in a direction orthogonal to the gradient, and an image region in which the overlapping degree of the line segments is a predetermined value or more is extracted.
[0017]
  The decoding processThen, a process of extracting a timing pattern, a process of identifying and binarizing each cell in the information display area, a process of decoding the information encoded by binarization while collecting it for each symbol character, etc. Can do. Further, in the decoding process, it is possible to perform a process of correcting an error that has occurred in the identification for each cell for each symbol character.
[0018]
  On the backgroundNoise such as patterns and shadowsIf the number of extractions for a particular pattern increases due toSince the processing time required for the extraction becomes long, there is a possibility that the time required for the decoding process is insufficient and decoding becomes impossible.
[0019]
  Also, for example, if the density difference between the black cell and the white cell is small or the size of the cell is not appropriate, the timing pattern cannot be extracted or the cell is in a monochrome state. The number of errors that occurred during the determination exceeds the correctable number, and decoding is impossible. In such a caseCan be considered that the number of extractions of a specific pattern is greater than zero and less than the number of extractions when decoding is impossible due to the noise.
[0020]
  In view of the above points, in the method of reading a two-dimensional code according to the present invention, a processing time of a predetermined length is assigned to reading one two-dimensional code, and the decoding process is repeated until the processing time elapses. When a reading failure in which optical information cannot be read occurs as a result of the decoding process being executed in accordance with the extraction of the specific pattern, the number of extractions of the specific pattern is set to read the one two-dimensional code. When the number of extractions exceeds the threshold value based on the processing time assigned to the image, the number of noises erroneously extracted as the specific pattern is the cause of the reading failure. If the number of extractions of the specific pattern falls below the threshold value, it is possible that the two-dimensional code in the image cannot be decoded. So that it is determined that it is.
[0021]
  The above-mentioned patent document 1 describes that the cause of the reading failure is determined based on the brightness level of the image, the code width, and the left and right margins. Nothing is shown about using the number of extractions for a particular pattern.
[0022]
  Above methodAccording to the decoding processWhether the failure is due to the amount of noise mis-extracted as a specific pattern or because the 2D code cannot be decodedIt can be discriminated and presented clearly to the user.
[0023]
  Take the above methodOne aspectThen, the specific patternFailed to be extracted and reading failedWhen the brightness of the image is compared with a predetermined numerical range. If the brightness is within the numerical range, it is determined that the position of the two-dimensional code at the time of imaging is the cause of the reading failure. If the brightness is outside the numerical range, illumination of the two-dimensional code is determined. It is determined that the state is the cause of the reading failure.
[0024]
  In the above aspect, the brightness of the image can be obtained by averaging the gradation for each pixel constituting the image. The numerical range is preferably set based on the brightness when the image is halated and the minimum brightness when the image is not illuminated (brightness only due to the influence of noise). If the brightness is within a numerical range, it is possible to identify the shade of the image. Therefore, if there is a specific pattern in the image, it can be extracted. Nevertheless, it can be considered that the specific pattern is not included in the field of view of the imaging device when the number of extraction of the specific pattern is zero.
[0025]
  On the other hand, if the brightness exceeds the numerical range, it can be considered that the illumination is too bright. Moreover, when the said brightness is less than a numerical range, it can be considered that illumination is too dark.
  Therefore, according to the above aspect,If a specific pattern is not extracted and reading fails, the failureIt is possible to accurately determine whether or not this is due to an inadequate position of the two-dimensional code at the time of imaging, or to indicate the cause.
[0026]
  Of the two-dimensional code reading methodSecondIn the aspect ofAccording to the determination of the cause of reading failureDisplay information indicating the cause. This display information is preferably comment information that allows the user to easily understand the cause.
[0027]
  Next, the two-dimensional code reader according to the present invention is as follows.Consists of two-dimensional array of two types of cells with different display modesImage acquisition means for acquiring an image with a two-dimensional code as an imaging target, and the acquired imageTo form a plurality of cells respectively located outside the information display area in the two-dimensional code.A specific pattern extracting means for extracting a specific pattern, and a specific pattern extracted by the specific pattern extracting means;The information display area is identified based on theThe image in the regionmake use ofDecoding processing means for executing decoding processing, read data output means for outputting read data indicating the content of information decoded by the decode processing means, and reading failure in which the read data output means cannot output read data occurred. And determining means for determining the cause of the reading failure based on the number of extractions of the specific pattern, and output means for outputting a determination result by the determining means.
[0028]
  The decoding processing means is set to be able to repeat the decoding processing until a processing time allocated for reading one two-dimensional code elapses. In addition, the threshold value determined based on the processing time assigned to the reading of the one two-dimensional code is set in the determination means, and the decoding process is executed in accordance with the extraction of the specific pattern. As a result, when the reading failure occurs, the number of extractions of the specific pattern is compared with the threshold value, and when the number of extractions exceeds the threshold value, a large amount of noise erroneously extracted as the specific pattern. Is the cause of the reading failure, and when the number of extractions of the specific pattern is less than the threshold value, the reason for the reading failure is that the two-dimensional code in the image cannot be decoded. Determine that there is.
[0029]
  The image acquisition means includes at least an input interface for inputting an image generated by an imaging unit such as a camera. When an analog image signal is input, it is desirable that the image acquisition unit includes an A / D conversion circuit.
[0030]
  The specific pattern extraction unit, the decoding processing unit, and the determination unit execute the specific pattern extraction process, the decoding process, and the process of determining the cause of the reading failure described in the two-dimensional code reading method, respectively. These means can be constituted by a computer incorporating a program, and each means can be constituted by one computer. However, the function of performing image processing on an image of a two-dimensional code can be configured by a separate computer or a dedicated hardware circuit for image processing.
[0031]
  Each of the read data output unit and the determination result output unit can be configured as an interface circuit for an external device. Further, these output means can be configured as a display device for displaying the discrimination result or an audio circuit for outputting the discrimination result as audio. Moreover, even if it uses any said form, each output means can be comprised integrally.
[0032]
  First aspect of the two-dimensional code readerThen, the discrimination means isThe specific pattern isReading failed without extractionThe brightness of the image is compared with a predetermined numerical range, and if the brightness is within the numerical range, it is determined that the position of the two-dimensional code at the time of imaging is the cause of the reading failure, and the brightness Is outside the numerical range, it is determined that the illumination state for the two-dimensional code is the cause of the reading failure.
[0033]
  further,The above two-dimensional code readerSecondIn the aspect, the output means is determined by the determination means.Reading failureIt is configured to include display means for displaying information indicating the cause. In this way, when a reading failure occurs, information indicating the cause of the failure can be displayed on the two-dimensional code reader itself, so that the user at the site of the reading process can quickly and easily determine the cause of the failure. , To make it easy to understandit can.
[0034]
  AboveThe two-dimensional code reading device according to each configuration can be connected to an image pickup unit (camera, scanner, etc.) for picking up a two-dimensional code, and a monitor for displaying a reading result of the two-dimensional code. Or these image pick-up parts and a monitor can also be integrated in the inside and outside of a single housing | casing with the structure concerning the above-mentioned various means.
[0035]
  According to the above two-dimensional code reading method and apparatus, when a reading failure occurs, the cause of the failure can be presented to the user. Therefore, for example, at the time of adjustment before starting full-scale reading processing, the cause of reading failure can be quickly removed and appropriate adjustment can be performed, and the setting work can be made more efficient.
[0036]
DETAILED DESCRIPTION OF THE INVENTION
  FIG. 1 shows the configuration of a two-dimensional code reader according to the present invention. In the figure, a lens 101 and an imaging device (CCD) 102 are for imaging a two-dimensional code 7, and an illumination lamp 6 is for illuminating the two-dimensional code 7 to be processed.
[0037]
  Further, the two-dimensional code reader 1 includes an imaging control unit 104, an illumination control unit 105, a trigger input unit 106, an A / D conversion unit 107, an image storage unit 108, an image processing unit 109, a decoding processing unit 110, a determination A processing unit 111, an output processing unit 112, and the like are included. Reference numeral 11 in the figure denotes a host system that controls the operation of the two-dimensional code reader 1, and is composed of a personal computer, a programmable controller (PLC), and the like. The monitor 4 can also be set as a component of the two-dimensional code reader 1 as will be described later.
[0038]
  The trigger input unit 106 receives a trigger signal that instructs the start of imaging from the host system 11. This trigger signal is given to the imaging control unit 104 and the illumination control unit 105. Thereby, the CCD 102 operates under illumination by the illumination lamp 6 and the two-dimensional code 7 is imaged.
[0039]
  The image signal from the CCD 102 is digitally converted by the A / D conversion unit 107 and then stored in the image storage unit 108. The image processing unit 109 extracts the finder pattern of the two-dimensional code 7 from the image data stored in the image storage unit 108 (hereinafter referred to as “processing target image”), and then based on the extraction result, The position and inclination of the two-dimensional code are discriminated. The image processing unit 109 extracts a timing pattern and recognizes the position and size of each cell in the code based on the arrangement. Further, a binarization process or the like is performed to determine whether each cell is a black cell or a white cell, or to determine a cell size.
[0040]
  The decode processing unit 110 encodes the optical information for each cell based on the processing result of the image processing unit 109. Further, the decoding processing unit 110 collects these codes for each symbol character, determines whether or not there is an error in the value (code word) indicated by the symbol character, and corrects the error if there is an error. To do. Thus, when the optical information is decoded, the decoding result is output to the output processing unit 113.
[0041]
  When the optical information reading fails in the series of processing by the image processing unit 109 or the decoding processing unit 110, the determination processing unit 111 executes processing for determining the cause of the failure. This determination result is also given to the output processing unit 112. The output processing unit 112 outputs the processing results from the decoding processing unit 110 and the determination processing unit 111 to the host system 11 and the monitor 4.
[0042]
  2 and 3 show a specific configuration example of the two-dimensional code reader 1. FIG. First, FIG. 2 shows the appearance of the two-dimensional code reading device 1 and its use state. The two-dimensional code reader 1 of this embodiment is incorporated into an information management system in a printed wiring board assembly and production line. A controller 2 incorporating a computer and a CCD camera 3 (with a cable connected to the controller 2) Hereinafter, the basic configuration is simply referred to as “camera 3”. In addition to the two connectors 8a for camera connection provided on the front surface of the controller 2, there are provided connectors 8b and 8c for connecting the monitor 4 and the console 5 (both shown in FIG. 3). In addition, the above-described illumination lamp 6 is accommodated in the camera 3.
[0043]
  The two-dimensional code reader 1 according to this embodiment is installed for the purpose of reading the two-dimensional code 7 attached to the surface of the component 9 mounted on the printed wiring board. The marking of the two-dimensional code 7 is performed upstream of the conveyor 10 by a marking device (not shown). This marking is performed by a method of directly marking the two-dimensional code 7 on the component 9 (direct marking), such as a method of forming a hole with a pin in a portion corresponding to a black cell or a method of printing a black cell with a laser beam. Done.
[0044]
  The camera 3 is installed above a conveyor 10 for conveying parts (hereinafter simply referred to as “conveyor 10”). The controller 2 is installed in the vicinity of the conveyor 10 and is connected to the host system 11 described above. The controller 2 receives the command from the host system 11 and operates the camera 3 to image the two-dimensional code 7, and then decodes optical information included in the two-dimensional code in the image.
[0045]
  The decoding result finally obtained by the controller 2 is transmitted to the host system 11. The two-dimensional code 7 stores various information necessary for assembly processing, such as the type of part, part number, part transport destination, and board number on which the part is assembled. The host system 11 performs control such as switching the route of the conveyor 10 or sending a command to an assembly robot (not shown) using such information.
[0046]
  FIG. 3 is a hardware block diagram of the two-dimensional code reader 1 with the controller 2 as the center. The controller 2 of this embodiment includes a CPU 21, a main memory 22, an input / output control unit 23, an image processing unit 24, a model memory 25, an image memory 26, and the like as main components. The camera 3, the monitor 4, the console 5, and the illumination lamp 6 in the camera 3 are connected to the input / output control unit 23 via interface circuits 203, 204, 205, and 206, respectively. The camera interface circuit 203 includes the A / D converter 107 described above.
[0047]
  The CPU 21 also exchanges data and commands with the host system 11 via the communication interface circuit 201. The trigger input unit 106 shown in FIG. 1 has a function for receiving a trigger signal in the communication interface circuit 201 and the CPU 21.
[0048]
  The input / output control unit 23 and the image processing unit 24 are connected to the CPU 21 via the CPU bus 27 and configured to exchange image data with each other via the image bus 28. The image output from the camera 3 is provided from the input / output control unit 23 to the image processing unit 24 and stored in the image memory 26. The image memory 26 corresponds to the image storage unit 108 in FIG. The image processing unit 24 reads out the image stored in the image memory 26 in accordance with an instruction from the CPU 21 and provides it to the input / output control unit 23.
[0049]
  The input / output control unit 23 can switch and display the image provided from the image processing unit 24 and the real-time image input from the camera 3 on the monitor 4. In addition to the display of the image, the result of the reading process for the image can be displayed. Further, when the reading of the two-dimensional code fails, the input / output control unit 23 receives comment information described later from the CPU 21 and displays it on the monitor 4. Further, the CPU 21 outputs an error code indicating the same content as the comment information to the upper system 11.
[0050]
  The main memory 22 stores programs necessary for the operation of the CPU 21, temporarily stores data generated in the course of processing, a comment information setting table (shown in FIG. 4), a threshold NM, L_max, L_minIt is used for the purpose of memorizing.
[0051]
  The model memory 25 is for registering a model used for extracting a finder pattern of a two-dimensional code, and is connected to the image processing unit 24 together with the image memory 26. The image processing unit 24 includes a processor dedicated to image processing and a digital arithmetic circuit, and corresponds to the image processing unit 109 of FIG.
[0052]
  The main memory 22 stores programs corresponding to the processing units of the imaging control unit 104, the illumination control unit 105, the decoding processing unit 110, the determination processing unit 111, and the output processing unit 112 in FIG. The CPU 21 can function as each processing unit by executing processing based on these programs.
[0053]
  The controller 2 of this embodiment performs a process of operating the camera 3 to capture the image of the two-dimensional code 7 and a process of reading the two-dimensional code of the image for each component 9 conveyed on the conveyor 10. , Is set to run repeatedly at high speed. The “reading process” means a series of processes from taking a two-dimensional code image into the controller 2 to ending the decoding, and is roughly divided into a finder pattern extraction process and a decoding process. The
[0054]
  In the finder pattern extraction process of this embodiment, a pattern matching process is executed in which a correlation calculation is performed while scanning the model in the model memory 25 on the processing target image (hereinafter, the finder pattern is extracted by this pattern matching). The search process is called “correlation search”.) However, instead of the correlation search, a finder pattern can be extracted by processing using an edge point on the image and its density gradient direction.
[0055]
  Here, the finder pattern extraction process will be briefly described. In this process, first, differential processing for edge extraction is performed on the processing target image. Next, at each edge point extracted, a line segment having a predetermined length is set along a direction orthogonal to the density gradient direction (direction from black to white).
[0056]
  This processing is suitable for processing a code in which black cells are represented as a dot pattern, such as a two-dimensional code by engraving marking. For example, in the finder pattern of the data matrix, dots indicating black cells are arranged in an L shape with a predetermined interval, and therefore, for each edge point on the outline of these dots, a line segment that approximates the tangent of the dot Can be set. Here, when attention is paid to the overlapping degree of these line segments, it is considered that the overlapping degree along the L-shaped arrangement direction becomes the largest. Therefore, from the line segments set in the processing target image, an image area whose degree of overlap is a predetermined value or more is extracted, and an image area that approximates the shape of the finder pattern is extracted from the image area. The position and orientation can be specified.
[0057]
  In the controller 2 of this embodiment, when the reading of the two-dimensional code fails in the above reading processing, processing for determining the cause of the failure is executed. The result of the discrimination process is converted into predetermined comment information and displayed on the monitor 4.
[0058]
  FIG. 4 shows an example of the comment information setting table. This setting table is stored in the main memory 22 and has a structure in which individual identification numbers are associated with five pieces of comment information. In the following, each comment information is indicated as “comment 1”, “comment 2”,...
[0059]
  In FIG. 4, comment 1 indicates that the illumination for the two-dimensional code 7 is too bright, and comment 2 indicates that the illumination is too dark. Comment 3 indicates that the two-dimensional code 7 is not correctly included in the field of view of the camera 3, and is applied when the entire two-dimensional code 7 or at least a part including the finder pattern is missing.
[0060]
  Comment 4 indicates that the two-dimensional code 7 on the image cannot be decoded. For example, when the timing pattern could not be extracted due to insufficient contrast, or in the identification process of white cells and black cells, there were a number of errors that could not be corrected due to the poor marking state of the two-dimensional code. Sometimes this comment 4 is applied.
[0061]
  Comment 5 indicates that the background state of the two-dimensional code 7 is bad. Here, a bad background state is a state in which there are many noises that are erroneously extracted as a finder pattern due to unevenness and patterns. As in this embodiment, when the two-dimensional code 7 is sequentially sent into the field of view of the camera 3 to perform the reading process, it is necessary to set a predetermined allocation time for each reading process. The comment 5 means that finder pattern candidates are extracted so that the decoding process cannot be completed within the allocated time.
[0062]
  Although not shown in FIG. 4, an error code corresponding to each comment information is also stored in this setting table. The error code is output to the host system 11 and also displayed on the monitor 4 together with the comment information.
[0063]
  FIG. 5 shows a reading process procedure (ST1 to ST17) for image data of one two-dimensional code in the controller 2.
  First, when a reading start signal is input from the host system 11, the camera 3 is driven to image a two-dimensional code to be processed. As described above, the image from the camera 3 is transferred from the input / output control unit 23 to the image processing unit 24 and then stored in the image memory 26 (ST1 to ST3).
[0064]
  In the next ST4, the image processing unit 24 performs a correlation search using the model in the model memory 25 on the image stored in the image memory 26, and extracts a finder pattern. The correlation search result is transmitted from the image processing unit 24 to the CPU 21. The CPU 21 temporarily stores in the main memory 22 the number of extracted finder pattern candidates (hereinafter referred to as “candidate number N”) and the similarity of each candidate to the model.
[0065]
  If the candidate number N is greater than 0, ST5 becomes “YES” and the process proceeds to ST6 decoding. Here, a specific procedure of the decoding process will be described with reference to FIG.
[0066]
  This decoding process is sequentially performed on each finder pattern candidate extracted in ST4. Note that the processing order starts from the one with the highest similarity to the model.
[0067]
  In the first ST61, a timing pattern is extracted based on the finder pattern candidate. In the next ST62, the position of each cell in the code is extracted based on the cell arrangement state of the timing pattern. Next, in ST63, for the information display area surrounded by the timing pattern, it is determined whether the cell is white or black by binarizing the density value for each cell in the area, and the like. Based on the determination result, the optical information represented by each cell is encoded. Further, in ST64, these codes are decoded while performing error correction processing (ST64).
[0068]
  Although not shown in FIG. 6, if any of the steps ST61 to 64 is in a state where the process cannot be executed, the following steps are canceled and the process proceeds to ST65. In ST65, it is determined whether or not the processing in each step has been successful. If the process has failed in any step, ST65 is “NO” and the process proceeds to ST66, where it is checked whether or not the processes for all the extracted finder pattern candidates have been completed. If there is an unprocessed candidate here, the process returns from ST66 to ST61, and the processes of ST61 to ST64 are similarly executed for the next candidate.
[0069]
  When information is decoded by the processes of ST61 to 64 for a predetermined candidate, ST65 becomes “YES” and the decoding process is terminated. That is, when the decoding for the predetermined candidate is successful, even if there is an unprocessed candidate, the processes of ST61 to ST64 for the candidate are canceled.
[0070]
  Returning to FIG. 5, if the decoding is successful by the procedure of FIG. 6, ST7 becomes “YES” and the process proceeds to ST8. In ST8, the decoding result is arranged and the read data in the final format is edited and output to the host system 11 and the monitor 4.
[0071]
  On the other hand, if the information cannot be decoded for any of the candidates for the finder pattern, ST7 is “NO”, and in ST9, the number of candidates N is compared with a predetermined threshold NM (NM> 0). To do. When the number of candidates N is smaller than the threshold value NM, the process proceeds from ST9 to ST10, and the comment 4 is output to the monitor 4. If the number of candidates N is greater than or equal to the threshold NM, the process proceeds from ST9 to ST11, and the comment 5 is output to the monitor 4.
[0072]
  Next, when the finder pattern cannot be extracted in ST4, ST5 becomes “NO” and the process proceeds to ST12, and the brightness L of the processing target image is calculated. This calculation process is performed by a method of averaging the gradations of all pixels in the processing target image. However, the present invention is not limited to this. For example, an average value of the maximum value and the minimum value of brightness may be obtained.
[0073]
  Next, in ST13 and ST14, the brightness L is set to a predetermined threshold value L._max, L_min(L_max<L_min). Here brightness L is L_maxIf larger, ST13 becomes “YES” and proceeds to ST17 to output the comment 1 to the monitor 4To do.The brightness L is L_minIf it is smaller, ST13 is “NO”, ST14 is “YES”, and the process proceeds to ST16 to output the comment 2 to the monitor 4.
[0074]
  The brightness L is L_min≦ L ≦ L_maxIf it is within the range, ST13 and 14 are both “NO”, the process proceeds to ST15, and the comment 3 is output to the monitor 5.
[0075]
  In ST9, when the number of candidates N is compared with the threshold NM, if other candidates are included in the area of the two-dimensional code indicated by a candidate, these candidates are regarded as one candidate. The number of candidates N may be reduced.
[0076]
  In the procedure of FIG. 5, the threshold value NM can be set based on the number N of candidates when it is not possible to secure time for decoding processing for finder pattern extraction. Since a certain processing time T is assigned in advance to reading one two-dimensional code 7, it is considered that the remaining time at the time when the finder pattern extraction processing is completed decreases as the number N of candidates increases. Therefore, for example, the minimum value of N satisfying the following expression (1) can be set as the threshold value NM.
  T−T1 <N × t (1)
[0077]
  In the above equation (1), T1 is a value obtained by adding a predetermined margin time to the time required to extract N finder patterns. T is the time required for the decoding process for one candidate. The value of t is a constant value according to the processing capability of the controller 2, and the value of T1 can be determined according to the number of candidates N. The allocation time T can be fixed to a predetermined value in advance, but is not limited thereto, and can also be set via the host system 11 or the console 5.
  When the allocation time T can be variably set as described above, the threshold value NM varies depending on the value of T. The allocation time T can be set according to the conveyance speed of the component 9.
[0078]
  The brightness threshold L_max, L_minThe setting of is as follows.
  For example, in the case of a device with 256 gradations of brightness, considering that the resolution required for light / dark discrimination is about 10 gradations, L_maxIs set to about 245 gradations. L_minIs set to about 50 gradations in consideration of the brightness level due to the influence of noise and the resolution required for discrimination being about 10 gradations.
  According to the above setting, if the image is too bright due to the halation phenomenon, the brightness L is L_maxBecome bigger. When the image is so dark that the black cell of the two-dimensional code and the black of the background cannot be visually recognized, the brightness L is L_minSmaller.
[0079]
  According to the procedure shown in FIG. 5, if the decoding process fails, it can be determined whether the cause of the failure is in the two-dimensional code or in the background based on the number N of candidates for the finder pattern. If the extraction of the finder pattern fails, it is determined whether the cause of the failure is the illumination or the position of the two-dimensional code with respect to the field of view of the camera 3 based on the brightness of the processing target image. be able to.
[0080]
  In particular, if there is a deficiency in the position of the illumination or the two-dimensional code, the deficiency can be corrected promptly based on the determination result. The adjustment work can be performed efficiently when adjusting the above.
[0081]
  Further, when it is determined that the two-dimensional code state is incomplete, the content of the incompleteness can be more specifically determined depending on which step of the decoding process the failure has occurred.
  For example, when the timing pattern cannot be extracted, it is considered that the contrast is insufficient, so that it can be determined that there is a problem in the illumination state. If the timing pattern can be extracted but cannot be decoded, it can be determined that the marking state of the two-dimensional code is incomplete.
[0082]
  FIGS. 7A and 7B show display examples on the monitor 4 when the reading process of FIG. 5 is completed. FIG. 7A shows a display example when reading is successful. The processing target image 30 including the image 31 of the two-dimensional code is displayed, and the read data 32 and “OK” obtained from the two-dimensional code are displayed. A mark 33 or the like is displayed.
[0083]
  FIG. 7B is a display example when the reading process fails, and comment information 34 (comment 1 in the illustrated example) based on the determination result of the cause of the reading failure is indicated by the corresponding error code 35 or “NG”. "And the mark 36 are displayed. Also in this example, the processing target image is displayed. However, since the two-dimensional code on the image cannot be identified, the processing target image 30 is illustrated as a blank state for convenience.
[0084]
【The invention's effect】
  According to this invention, when the reading process of the two-dimensional code fails, the cause of the failure can be determined based on the number of extractions of the specific pattern in the reading process, and the determination result can be presented to the user. Therefore, the user can accurately grasp the cause of the reading failure and can take an accurate response to the failure. Therefore, even if there are deficiencies in the setting conditions such as illumination, field of view of the imaging means, and the position of the two-dimensional code during adjustments before starting full-scale reading processing, the deficiencies should be removed promptly. Can do.
[Brief description of the drawings]
FIG. 1 is a functional block diagram showing a configuration of a two-dimensional code reader to which the present invention is applied.
FIG. 2 is a diagram illustrating an appearance and a usage state of a two-dimensional code reader.
FIG. 3 is a block diagram showing an electrical configuration of the two-dimensional code reader.
FIG. 4 is a diagram showing a comment information setting table;
FIG. 5 is a flowchart illustrating a procedure of reading processing.
FIG. 6 is a flowchart showing a detailed procedure of decoding processing.
FIG. 7 is a diagram illustrating a display example at the end of reading.
FIG. 8 is an explanatory diagram showing a representative example of a two-dimensional code and a finder pattern and a timing pattern in each code.
[Explanation of symbols]
  1 Two-dimensional code reader
  2 Controller
  3 Camera
  4 Monitor
  7 Two-dimensional code
21 CPU
22 Main memory
23 Input / output control unit
24 Image processing unit
26 Image memory

Claims (6)

A specific pattern formed by a plurality of cells respectively positioned outside the information display area in the two-dimensional code from an image obtained by imaging a two-dimensional code composed of a two- dimensional array of two types of cells having different display modes. In the method of reading the optical information represented by the two-dimensional code in the image by extracting and specifying the information display area based on the extracted specific pattern and performing a decoding process on the area ,
Assigning a predetermined length of processing time to reading one two-dimensional code, and allowing the decoding process to be repeated until the processing time elapses;
As a result of performing decoding processing in accordance with the extraction of the specific pattern, when a reading failure in which optical information cannot be read occurs, the number of extraction of the specific pattern is assigned to the reading of the one two-dimensional code. When the number of extractions exceeds the threshold, it is determined that the number of noises erroneously extracted as the specific pattern is the cause of the reading failure. When the number of extractions of the specific pattern is less than the threshold value, it is determined that the reading failure is caused by the fact that the two-dimensional code in the image cannot be decoded.
A method for reading a two-dimensional code. The method of claim 1, wherein
When reading fails because the specific pattern is not extracted, the brightness of the image is compared with a predetermined numerical range, and if the brightness is within the numerical range, the position of the two-dimensional code at the time of imaging is A method for reading a two-dimensional code, wherein the reading failure is determined to be the cause of the reading failure, and if the brightness is outside the numerical range, the illumination state on the two-dimensional code is determined to be the cause of the reading failure. The method according to claim 1 or 2, wherein
A method for reading a two-dimensional code, wherein information indicating the determined cause is displayed according to the determination of the cause of the reading failure . Image acquisition means for acquiring an image of a two-dimensional code composed of a two- dimensional array of two types of cells having different display modes ,
Specific pattern extraction means for extracting a specific pattern formed by a plurality of cells respectively located outside the information display area in the two-dimensional code from the acquired image;
Decoding processing means for specifying the information display area based on the specific pattern extracted by the specific pattern extraction means, and executing decoding processing using an image in the specified area;
Read data output means for outputting read data indicating the content of the information decoded by the decode processing means;
A discriminating unit for discriminating the cause of the reading failure based on the number of extractions of the specific pattern when a reading failure occurs in which the reading data output unit cannot output reading data;
Output means for outputting a discrimination result by the discrimination means ,
The decoding processing means is set to be able to repeat the decoding processing until the processing time allocated for reading one two-dimensional code has elapsed,
A threshold value determined based on a processing time assigned to reading the one two-dimensional code is set in the determination unit, and a decoding process is executed in accordance with the extraction of the specific pattern. As a result, when the reading failure occurs, the number of extractions of the specific pattern is compared with the threshold value, and when the number of extractions exceeds the threshold value, a large amount of noise erroneously extracted as the specific pattern. Is the cause of the reading failure, and when the number of extractions of the specific pattern is less than the threshold value, the reason for the reading failure is that the two-dimensional code in the image cannot be decoded. A two-dimensional code reading device that determines that there is . The discriminating unit compares the brightness of the image with a predetermined numerical range when the specific pattern is not extracted and fails to read , and if the brightness is within the numerical range, 2 at the time of imaging. determines that the position of the dimension code is responsible for the reading failure, according to claim 4, wherein the brightness of the illumination to the two-dimensional code if outside the value range state is determined to be the cause of the reading failure Two-dimensional code reader. An apparatus as claimed in claim 4 or 5 , wherein
The two-dimensional code reading device, wherein the output means includes display means for displaying information indicating the cause of reading failure determined by the determining means.

Images