Inkjet printer

Abstract

To enhance the accuracy of quality determination while controlling a useless spring.SOLUTION: An inkjet printer 1 includes: a discharge part 22 for discharging ink toward an object to be drawn; an imaging part 3 for imaging the object to be drawn to generate an imaging image; a storage part 41 for storing determination sensitivity showing a level allowing drawing disturbance; and a quality determination part 40d for determining the quality of a drawing state drawn on the object to be drawn by the imaging image on the basis of the determination sensitivity stored in the storage part 41.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to inkjet printing devices. [Background technology]

[0002] For example, a drop-on-demand (DOD) inkjet printing device is known that prints by ejecting ink onto packaging materials such as cardboard (see, for example, Patent Document 1). Also known is a technology that uses a camera to capture an image of a pattern drawn by a DOD inkjet printing device and executes various applications (such as high-precision alignment) using the captured image (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-144574 [Patent Document 2] Japanese Patent Application Publication No. 2020-131141 Summary of the Invention [Problem to be solved by the invention]

[0004] Consider, for example, a system that checks whether the print content read by a camera matches the pre-set print content, i.e., judges whether the print content is good or bad, and outputs a message indicating that the print content is bad if it is judged to be bad. When inspecting the print quality of cardboard using an inkjet printer, black spots due to individual differences between cardboard boxes, dust on the cardboard surface, or slight vibrations in the conveying line can lead to an erroneous judgment that the print quality is bad, even if it is acceptable in terms of print quality. This type of erroneous judgment is generally called a "waste spring." This waste spring can cause cardboard boxes that do not need to be ejected from the conveying line or stop the conveying line, resulting in a decrease in productivity.

[0005] In this regard, Patent Document 2 does not specifically consider how to prevent unnecessary springs, and therefore is unable to avoid the above-mentioned deterioration in productivity.

[0006] The present disclosure has been made in consideration of the above points, and its purpose is to improve the accuracy of pass / fail determination while suppressing unnecessary spring. [Means for solving the problem]

[0007] To achieve the above object, one aspect of the present disclosure can be premised on an inkjet printing device that ejects ink for drawing information on an object and determines the quality of the drawing by capturing an image. The inkjet printing device includes an ink storage unit that stores the ink, a discharge unit that ejects ink supplied from the ink storage unit toward an object to be drawn, an imaging unit that images the object to be drawn and generates a captured image, a storage unit that stores a determination sensitivity that indicates a degree of tolerance for drawing distortion, and a quality determination unit that determines the quality of the drawing on the object to be drawn by using the captured image based on the determination sensitivity stored in the storage unit.

[0008] According to this configuration, when determining the quality of the drawing state drawn on the object to be drawn using the captured image, the determination is made based on the determination sensitivity stored in the memory unit, so that the accuracy of the determination is not reduced and it is possible to prevent objects with drawing states that have no practical effect from being determined to be defective, thereby reducing wasted springs. [Effects of the Invention]

[0009] As described above, the quality of the drawing state drawn on the object to be drawn is determined from the captured image based on the judgment sensitivity stored in the memory unit, so that the accuracy of the quality judgment can be improved while suppressing unnecessary springs. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing the overall configuration of an inkjet printing apparatus according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram of an inkjet printing device. [Figure 3] FIG. 3 is a flowchart showing an example of processing when classifying an image. [Figure 4] FIG. 4 is a flowchart showing an example of secondary determination processing when classifying images. [Figure 5] FIG. 5 is a flowchart showing an example of a process for determining a character string to be read from a captured image. [Figure 6] FIG. 6 is a diagram showing an example of a captured image and a reading result by an inspection machine based on the captured image. [Figure 7] FIG. 7 is a diagram showing the relationship between the captured image and the determination result. [Figure 8] FIG. 8 shows an example of a window for changing the filter level and the allowable number of string matches. [Figure 9] FIG. 9 is a flowchart showing an example of processing for classifying drawing elements. [Figure 10] FIG. 10 is a flowchart showing an example of processing for classifying images obtained by capturing a plurality of drawing elements. [Figure 11] FIG. 11 is a flowchart showing an example of distance notification processing. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that the following description of the preferred embodiments is merely exemplary in nature and is not intended to limit the present invention, its applications, or its uses.

[0012] Fig. 1 is a diagram showing the overall configuration of an inkjet printing apparatus 1 according to an embodiment of the present invention during operation, and Fig. 2 is a block diagram of the inkjet printing apparatus 1. The inkjet printing apparatus 1 includes, for example, a drawing machine 2, an inspection machine 3, a controller 4, and an operation terminal 5.

[0013] The drawing machine 2 is a part that executes drawing processing on an object to be drawn, and is configured separately from the controller 4. The inspection machine 3 is a part that judges whether the drawing state executed by the drawing machine 2 on the object to be drawn is good or bad, and is configured separately from the controller 4. Since the inkjet printing apparatus 1 according to this embodiment is equipped with the drawing machine 2 and the inspection machine 3, it is an inkjet printing apparatus that ejects ink to draw information on an object and judges whether the drawing state is good or bad by capturing an image.

[0014] The controller 4 is a part that controls the drawing machine 2 and the inspection machine 3, and can also be called a control device, control unit, etc. The operation terminal 5 is composed of a personal computer or the like, and is a part where the user inputs drawing information, makes various settings, makes selections, checks, etc. The operation terminal 5 includes a terminal-side display unit 5a, which is composed of, for example, a liquid crystal display, and a terminal-side operation unit 5b, which is composed of a keyboard, mouse, pointing device, etc.

[0015] The above configuration example is merely an example, and the present invention is not limited to the above configuration example. That is, the controller 4 may be incorporated into the operation terminal 5, or may be incorporated into the drawing device 2. Furthermore, the operation terminal 5 may function as the controller 4. Furthermore, FIGS. 1 and 2 show an external device 6 formed of a programmable logic controller (PLC) or the like, which is communicatively connected to the controller 4. A control signal output from the external device 6 is input to the controller 4. The external device 6 may be a device that constitutes a part of the inkjet printing apparatus 1, or may be a device separate from the inkjet printing apparatus 1.

[0016] The inkjet printing device 1 is a device for drawing various types of information on a workpiece W (an example of an object to be drawn) transported by a transport device 10, and is an in-line device that is incorporated into a manufacturing line (also called a production line). The transport device 10 is, for example, a belt conveyor, a roller conveyor, etc. The production line is made up of the transport devices 10. The production line is installed in various factories, warehouses, etc. The drawing device 2 is fixed so that its relative position with respect to the transport device 10 is in a predetermined positional relationship.

[0017] When the inkjet printing device 1 is in operation, multiple workpieces W are placed on the conveying surface 10a of the conveyor 10 and conveyed in sequence in the direction of arrow A shown in Figure 1. Therefore, the upstream side in the conveying direction is the left side in Figure 1, and the downstream side in the conveying direction is the right side in Figure 1. The multiple workpieces W are placed on the conveying surface 10a with spaces between them in the direction of arrow A, which is the conveying direction. The direction perpendicular to the direction of arrow A and along the conveying surface 10a is defined as the width direction.

[0018] The workpiece W is not particularly limited, but may be, for example, packaging materials used to package various products. A typical example of packaging materials is cardboard. The inkjet printing device 1 draws information such as characters including numbers, symbols, barcodes, two-dimensional codes, images, marks, illustrations, or combinations thereof, on the workpiece W being transported by the transport device 10. When only characters are drawn, the inkjet printing device 1 may also be called a printing device, and the drawing device 2 may also be called a printer. Furthermore, drawing information by the inkjet printing device 1 also includes printing images, etc. In this case, the inkjet printing device 1 may also be called a printing device. In the following description, the term "drawing" will be used to refer to both "printing" and "printing."

[0019] In addition, an encoder 7, a timing sensor 8, and the like for detecting the position of the workpiece W are connected to the controller 4. The controller 4 detects the position of the workpiece W based on signals output from the encoder 7 and the timing sensor 8. The controller 4 controls the drawing machine 2 so that drawing begins when the workpiece W reaches a predetermined position.

[0020] (Configuration of Rendering Machine 2) The drawing device 2 is a DOD (Drop On Demand) drawing device that ejects ink only when required for drawing operations, but it may also be a CIJ (Continuous Ink Jet) drawing device that ejects ink even when no drawing operation is being performed. Below, we will explain the case where a DOD drawing device is used as the drawing device 2 of this embodiment.

[0021] As shown in Fig. 2, the drawing machine 2 includes an ink supply unit 20, a print head drive unit 21, and a DOD print head (hereinafter referred to as a print head) 22. The drawing machine 2 also includes a main housing 25 (shown in Fig. 1) in which the print head 22 is provided. The main housing 25 is fixed to the conveyor 10 or the like.

[0022] The print head 22 is fixed in place inside the main housing 25, and the positional relationship between the main housing 25 and the print head 22 is fixed. The ink supply unit 20 and the print head drive unit 21 are also housed inside the main housing 25.

[0023] The print head 22 of the DOD type drawing machine 2 is a component equivalent to a discharge unit, and discharges ink for drawing information on the workpiece W. There are several types of print head 22 structures, and the print head 22 may be any of the following types: thermal ink jet, valve jet, and piezo. In this embodiment, a piezo type print head is used as the print head 22, which is capable of drawing from low resolution to high resolution, has a wide drawing width, and is highly durable.

[0024] The print head 22 has a plurality of ejection ports (not shown) formed, for example, along the vertical direction. As shown in FIG. 1, an opening 25a that is long in the vertical direction is formed on the surface of the main housing 25 that faces the workpiece W. The surface of the print head 22 on which the ejection ports are formed is positioned so as to face the outside through the opening 25a of the main housing 25. Therefore, ink ejected from the ejection ports of the print head 22 flies from the opening 25a of the main housing 25 to the outside of the main housing 25 and adheres to the workpiece W.

[0025] The ink supply unit 20 is a part that supplies ink to the print head 22, and is provided with an ink cartridge 20a as an ink storage unit that stores ink. Instead of the ink cartridge 20a, an ink reservoir (ink storage unit) that can be refilled with ink may be provided.

[0026] The print head driver 21 is controlled by the control unit 40, which will be described later, when performing a drawing operation. The print head driver 21 generates a driving electric waveform for driving the print head 22 and outputs it to the print head 22. The driving electric waveform is a waveform for individually driving the driving elements (piezoelectric vibrators) provided for each ejection port of the print head 22 at a timing based on a drawing command output from the control unit 40. The print head 22 draws on the workpiece W, which is the drawing target, by ejecting ink supplied from the ink cartridge 20a toward the workpiece W.

[0027] (Inspection machine configuration) The inspection machine 3 is a device that judges whether the drawing state of the information drawn on the workpiece W to be drawn based on the drawing data is good or bad by capturing an image of the workpiece W to be drawn, and corresponds to an image inspection unit. The inspection machine 3 can also be called an image inspection device. The inspection machine 3 is installed downstream of the drawing machine 2, and is capable of inspecting the drawing state of the workpiece W drawn by the drawing machine 2.

[0028] The inspection machine 3 includes an illumination unit 30, an imaging unit 31, a control unit 32, and a memory unit 33. The illumination unit 30 is composed of, for example, a light-emitting diode or the like, and is controlled by the control unit 32 to emit light at a predetermined timing to illuminate the drawn portion of the workpiece W. The imaging unit 31 includes an optical system 31a including a lens that receives light emitted from the illumination unit 30 and reflected from the surface of the workpiece W, an image sensor 31b that receives light emitted from the optical system 31a, and an AF motor 31c. The AF motor 31c is a component that automatically focuses on the drawn portion of the workpiece W by driving a focusing lens of the optical system 31a. The autofocus method is not particularly limited, and examples include a contrast method.

[0029] The image sensor 31b is configured with, for example, a CMOS (Complementary Metal Oxide Semiconductor) sensor. The image sensor 31b captures an image of the workpiece W to be imaged and generates a captured image. For example, a light receiving amount signal of a light receiving element of the image sensor 31b is output to an FPGA (Field Programmable Gate Array) (not shown) for processing, and is also output to a DSP (Digital Signal Processor) for processing.

[0030] The control unit 32 has an imaging setting unit 32a. The imaging setting unit 32a is a unit that sets imaging setting parameters and reflects the set imaging setting parameters when capturing an image of the workpiece W. The imaging setting parameters include a plurality of parameters, such as the timing of illumination by the illumination unit 30, brightness (light emission amount), exposure time by the imaging unit 31, and focus (focal position) of the imaging unit 31. Although not shown, the GUI for setting the imaging setting parameters has an area where the illumination brightness, exposure time, focus, etc. can be individually set. When the user inputs each parameter on the GUI, the imaging setting unit 32a accepts the user's setting operation and outputs each parameter to, for example, the storage unit 33 for storage in the storage unit 33. Each parameter can be read from the storage unit 33 as needed.

[0031] When acquiring captured image data of the workpiece W, the imaging setting unit 32a outputs imaging setting parameters to the lighting unit 30 and the imaging unit 31. The lighting unit 30 and the imaging unit 31 operate in accordance with the imaging setting parameters. When the lighting unit 30 receives the imaging setting parameters, it sets the received imaging setting parameters within the imaging unit 31 so that they are reflected when imaging. Furthermore, when the imaging unit 31 receives the imaging setting parameters, it sets the received imaging setting parameters within the imaging unit 31 so that they are reflected when imaging. When the lighting unit 30 and the imaging unit 31 receive an imaging trigger, they illuminate the workpiece W in accordance with the imaging setting parameters and capture an image to acquire captured image data of the workpiece W.

[0032] The control unit 32 may be a part that performs rule-based image inspection as an inspection tool, a part that performs image inspection using a pre-trained model, or a part that performs both rule-based image inspection and image inspection using a pre-trained model. The control unit 32, for example, causes the imaging unit 31 to capture an image of the workpiece W, acquires captured image data of the workpiece W, and performs an inspection based on a preset inspection tool on the acquired captured image of the workpiece W. When performing the inspection, the control unit 32 cuts out the range to be inspected from the captured image. The control unit 32 extracts feature values ​​from the cut-out inspection range and evaluates the extracted feature values ​​using a classifier. If the cut-out inspection range contains characters, the inspection machine 3 uses a dictionary to perform character recognition, enabling character recognition processing (OCR). Furthermore, if a barcode or two-dimensional code is drawn on the workpiece W, the inspection machine 3 can read the barcode or two-dimensional code.

[0033] After the inspection by the inspection machine 3, the control unit 32 creates an inspection result by the inspection tool and compares it with the inspection conditions. The inspection conditions are acquired from the drawing data generation unit 40b of the controller 4. For example, the result of the character recognition process by the control unit 32 is compared with the correct character string acquired from the drawing data generation unit 40b, and if they match, it is judged as good, and if they do not match, it is judged as bad. In the case of a barcode or two-dimensional code, the read result by the control unit 32 is compared with the code information acquired from the drawing data generation unit 40b, and if they match, it is judged as good, and if they do not match, it is judged as bad.

[0034] For example, if there is one configured inspection tool, the inspection results for that configured inspection tool are output as is, but if multiple configured inspection tools are combined, further inspection results are generated and output based on the inspection results of each inspection tool.

[0035] The judgment results by the inspection tool in the inspection machine 3 are not the final judgment results of the inkjet printing device 1, but are collected by the controller 4 described later, and the controller 4 makes a final pass / fail judgment based on the collected judgment results. Details of the pass / fail judgment process by the controller 4 will be described later.

[0036] (Controller 4 configuration) As shown in FIG. 2, the controller 4 includes a control unit 40, a storage unit 41, and an operation display unit 42. The control unit 40 is configured with a microcomputer including, for example, a central processing unit, various memories, etc., and is capable of executing pre-stored software. The control unit 40 includes a drawing setting unit 40a, a drawing data generation unit 40b, an inspection result collection unit 40c, a pass / fail determination unit 40d, and a display screen generation unit 40e. The drawing setting unit 40a, the drawing data generation unit 40b, the inspection result collection unit 40c, the pass / fail determination unit 40d, and the display screen generation unit 40e of the control unit 40 are configured with hardware and software. For convenience, the drawing setting unit 40a, the drawing data generation unit 40b, the inspection result collection unit 40c, the pass / fail determination unit 40d, and the display screen generation unit 40e are described separately, but may be integrated in terms of hardware.

[0037] The operation display unit 42 includes a controller-side display unit 42a consisting of, for example, a liquid crystal display or the like, and a controller-side operation unit 42b consisting of operation keys or the like. The controller-side operation unit 42b is a unit that accepts various operations such as drawing content and settings. Various screens generated by the display screen generation unit 40e are displayed on the controller-side display unit 42a, and the user can operate the controller-side operation unit 42b while looking at the controller-side display unit 42a. The operation state of the controller-side operation unit 42b can be acquired by the control unit 40.

[0038] The controller-side operation unit 42b may be a touch panel. A touch panel is a member capable of detecting operations performed by a user's finger. The type of touch panel is not particularly limited, and examples include a capacitance type and an infrared type. Information about the user's operation of the touch panel is transmitted to the control unit 40.

[0039] The drawing setting unit 40a is a part that sets the drawing content to be drawn on the workpiece W that is the drawing target. The drawing content includes character strings, barcodes, two-dimensional codes, etc. that are to be drawn on the object that is the drawing target. The drawing setting unit 40a makes it possible to set multiple drawing contents. For example, it is possible to draw multiple character strings on one workpiece W, or to draw a combination of character strings and barcodes or two-dimensional codes on one workpiece W. The drawing data generation unit 40b is a unit that defines the ink ejection from the print head 22 and generates drawing data corresponding to the drawing content set by the drawing setting unit 40a. The control unit 40 outputs a control signal to the print head drive unit 21 based on the drawing data generated by the drawing data generation unit 40b, and controls the ejection of ink from the ejection openings of the print head 22.

[0040] The memory unit 41 stores the judgment sensitivity used by the pass / fail judgment unit 40d when making a pass / fail judgment. The judgment sensitivity indicates the degree of tolerance for distortion in the drawing made by the drawing machine 2. Here, as shown in FIG. 1, assume that the object being drawn by the inkjet printing device 1 is, for example, cardboard. The surface of the cardboard may have black dots or dust particles attached thereto, but these foreign objects often pose no practical problems. However, when inspecting the quality of the drawing on the cardboard surface using an image, if the image contains such foreign objects, the inspection machine 3 may erroneously judge the image as defective, even though the foreign objects have no practical impact. This type of erroneous judgment is generally referred to as a "wasteful spring." This wasteful spring can cause cardboard boxes that do not need to be discharged to be ejected from the conveyor line or stop the conveyor line, thereby reducing productivity.

[0041] In contrast to this, in this embodiment, in order to increase the accuracy of the pass / fail judgment while suppressing wasted spring, the inspection result collection unit 40c collects the judgment results of the inspection machine 3, and the pass / fail judgment unit 40d makes a final pass / fail judgment based on the judgment results collected by the inspection result collection unit 40c. Specifically, the inspection machine 3 outputs the judgment results in chronological order, and the inspection result collection unit 40c collects the judgment results output from the inspection machine 3 in chronological order. An image may be captured once for one drawing, and the inspection result collection unit 40c may collect multiple judgment results for the multiple drawings in chronological order, or an image may be captured multiple times for one drawing, and the inspection result collection unit 40c may collect multiple judgment results for one drawing in chronological order.

[0042] The quality determination unit 40d determines the quality of the drawing state drawn on the drawing target object from the captured image, based on the determination sensitivity stored in the storage unit 41. Specifically, the quality determination unit 40d has a primary determination unit 40A that classifies the captured image into a good image in which the drawing state is good, a defective image in which the drawing state is poor, and an intermediate image in which the drawing state is not poor but does not satisfy the criteria for being determined to be good, and a secondary determination unit 40B that classifies the intermediate image into either a good image or a defective image based on the determination result and determination sensitivity of the secondary determination unit 40B.

[0043] Although not shown, the inspection machine 3 may be provided with a pass / fail determination unit. In this case, the pass / fail determination unit can be configured as part of the control unit 32 of the inspection machine 3. The determination process by the control unit 32 of the inspection machine 3 may be executed by the pass / fail determination unit 40d of the controller 4. The part of the control unit 32 of the inspection machine 3 that executes the determination process may be integrated with the primary determination unit 40A and the secondary determination unit 40B of the controller 4 to form the pass / fail determination unit.

[0044] Details of the judgment process by the primary judgment unit 40A of the pass / fail judgment unit 40d will be described below with reference to the flowchart shown in Fig. 3. Steps SA1 to SA4 after the start of this flowchart are executed by the inspection machine 3. In step SA1, the control unit 32 of the inspection machine 3 acquires a correct character string from the drawing data generation unit 40b. In step SA2, the control unit 32 of the inspection machine 3 acquires an inspection target image (for example, a captured image 90 shown in Fig. 4) from the imaging unit 3. In step SA3, an inspection area 91 to which a pre-set inspection tool is to be applied is extracted from the inspection target image 90. In step SA4, the control unit 32 of the inspection machine 3 performs character recognition processing on the character string included in the inspection area 91 extracted in step SA3.

[0045] When the inspection results output from the inspection machine 3 are input to the quality judgment unit 40d, the primary judgment unit 40A of the quality judgment unit 40d refers to the inspection results in step SA5 and judges whether or not a single character can be recognized as a result of the character recognition in step SA4. If a single character cannot be recognized, it is assumed that the drawing quality is poor and the characters are illegible, so the process proceeds to step SA6 and the image is classified as a defective image.

[0046] If the primary judgment unit 40A judges NO in step SA5 and has recognized one or more characters, the process proceeds to step SA7. In step SA7, the primary judgment unit 40A refers to the character recognition result of step SA4 and judges whether the number of recognized characters (number of recognized characters) is outside the allowable range. If the number of recognized characters is outside the allowable range, it is presumed that the drawing condition is not good and there are few readable characters, so the process proceeds to step SA6, where the primary judgment unit 40A classifies the image as a defective image. The number of characters that serves as the judgment threshold in step SA7 is set in advance by the user.

[0047] If the primary determination unit 40A determines NO in step SA5 and the number of recognized characters is within the acceptable range, it is possible that the drawing condition is relatively good, and that some characters are illegible due to the presence of, for example, a foreign object (a black circle indicated by reference numeral 92 in FIG. 4 ) on the surface of the cardboard. In such a case, the captured image 90 is not immediately classified as a defective image, but proceeds to step SA8. In step SA8, the primary determination unit 40A refers to the character recognition results of step SA4 and determines whether the degree of match between the recognized character string and the correct character string is outside the acceptable range. If the recognized character string is outside the acceptable range, the process proceeds to step SA9, where the primary determination unit 40A classifies the captured image 90 as an intermediate image. If it is determined in step SA8 that the degree of match between the recognized character string and the correct character string is within the acceptable range, the process proceeds to step SA10, where the primary determination unit 40A classifies the captured image 90 as a non-defective image. The degree of match that serves as the determination threshold in step SA8 is set in advance by the user.

[0048] In this way, in steps SA7 and SA8, the captured image 90 is judged based on the judgment threshold set by the user, and the captured image 90 is classified into three types: defective image, intermediate image, and good image, thereby enabling inspection of drawing quality according to the user's needs.

[0049] 4, the details of the judgment process for determining the final judgment result by the primary judgment unit 40A and secondary judgment unit 40B of the pass / fail judgment unit 40d regarding the pass / fail of a single image will be described below. Steps SB1 to SB6 of this flowchart are executed by the controller 4. First, in step SB1, a filter level n0 is obtained. The filter level is a judgment threshold value that indicates the degree to which succession of intermediate images is permitted when making judgments by the primary judgment unit 40A, i.e., the allowable number of times. A default value for the filter level is set in advance, and it can also be edited by user input.

[0050] In step SB2, the current count value is acquired. The current count value is a value that depends on the judgment result for the image acquired in time series before the image to be inspected, and the specific counting method will be explained in the steps below.

[0051] In step SB3, as explained above based on the flowchart shown in FIG. 3, the primary judgment section 40A classifies the inspection target images into three types: defective images, intermediate images, and non-defective images. If the image is classified as a non-defective image in the primary judgment, the process proceeds to step SB4, where the count value is set to 0, and the inspection object is classified as a non-defective image as the final judgment result.

[0052] If the image is classified as an intermediate image in the primary judgment, proceed to step SB5, increment the current count value n by 1, and proceed to step SB7, compare the count value n with the filter level n0. If the count value is equal to or less than the filter level, it is determined that the number of consecutive intermediate images is acceptable, and proceed to step SB8, where the image to be inspected is classified as a non-defective image as the final judgment result. On the other hand, if the count value is greater than the filter level, it is determined that the number of consecutive intermediate images exceeds the acceptable number, and proceed to step SB9, where the image to be inspected is classified as a defective image as the final judgment result.

[0053] If the image is classified as a defective image in the primary judgment at step SB3, the process proceeds to step SB6, where the count value is set to filter level value n0, and the process proceeds to step SB9, where the image to be inspected is classified as a defective image as the final judgment result. If an image obtained in time series after an image classified as a defective image in the primary judgment is classified as an intermediate image in the primary judgment, the count value becomes greater than filter level n0, and the image is classified as a defective image as the final judgment result, regardless of the filter level value. Also, by setting the count value for images classified as defective in the primary judgment to the current count value plus 1, as with intermediate images, rather than the filter level value, the filter level can also be used as a degree of tolerance for consecutive intermediate or defective images.

[0054] The following describes in detail a method for determining a read character string obtained from an acquired image, which is used for character string matching and recognition performed by the pass / fail judgment unit 40d, based on the flowchart shown in FIG. 5. Steps SC1 to SC5 of this flowchart are executed by the inspection machine 3. In step SC1, the control unit 32 of the inspection machine 3 acquires an inspection target image (e.g., the captured image 90 shown in FIG. 6) from the imaging unit 3. In step SC2, the control unit 32 of the inspection machine 3 recognizes blocks of print (blobs) from the inspection target image 90 acquired in step SC1, calculates the character similarity (read score) based on the feature amount of the blob, extracts read candidates, and sorts them in descending order of read score (first candidate, second candidate, etc.). In step SC3, the inspection machine 3 acquires a correct character string based on the drawing content set by the drawing setting unit 40a of the controller 4. In step SC4, the extracted read candidates are searched for characters that match the acquired correct character string, and the read score of the matching read candidate is increased. In step SC5, the read candidate with the highest score is finally adopted as the final read character string. In steps SC4 and SC5, instead of increasing the reading score, if a character that matches the correct character string is found in the reading candidates, that character can be unconditionally adopted as the final reading character string.

[0055] In the example shown in FIG. 6, the number of characters constituting the string included in the inspection area 91 of the captured image 90 is five, and the correct string is "12:39." The pass / fail determination unit 40d lists multiple candidates for each of the first, second, third, fourth, and fifth characters as the reading result. At this time, the pass / fail determination unit 40d performs character string matching and recognition based on the drawing content set by the drawing setting unit 40a and the captured image captured by the imaging unit 3, and calculates a reading score that indicates the similarity of the characters in the character string matching and recognition. The candidate with the highest reading score is the first candidate, and the candidate with a lower reading score than the first candidate is the second candidate. A third candidate and / or a fourth candidate may be listed as necessary.

[0056] Normally, the candidate with the highest reading score (first candidate) among multiple candidates is used as the reading result, and the read character string is read as "12:89." Since this reading result differs from the correct character string "12:39," the captured image 90 shown in FIG. 6 is normally classified as a defective product image.

[0057] In contrast to this normal processing, in this embodiment, when a character included in the correct string is listed as a reading candidate, such as the second candidate for the third character, the quality determination unit 40d performs processing to increase the reading score of the second candidate character higher than the first candidate. Also, a first candidate and a second candidate are listed for each of the first and third characters, but the second candidate for the first character is the alphabet "I" and the second candidate for the third character is ";". For the first and third characters, the first candidate is correct.

[0058] The fourth character is partially obscured by a foreign object 92, such as a black dot. Therefore, the first candidate for the fourth character is "8," and the second candidate is "3." Since the correct character string is known to be "12:39," the read score of the second candidate for the fourth character is increased, resulting in the read character string being "12:39." In this way, when the character string matching and recognition yields a second candidate character that is less similar to the first candidate character, if the second candidate character is included in the drawing content set by the drawing setting unit 40a, the pass / fail determination unit 40d assigns a higher score to the second candidate character than to the first candidate character. This reduces wasted space due to individual differences between cardboard boxes and dust adhering to the cardboard surface.

[0059] If the correct character "3" is not listed as a candidate for the fourth character, it is estimated that the drawing condition is poor, and the captured image 90 is classified as a defective image.

[0060] 7 shows the relationship between the read character strings of 12 captured images (images 1 to 12) acquired in chronological order and the primary and secondary judgment results. The primary judgment result is the judgment result by the primary judgment unit 40A, and the secondary judgment result is the judgment result by the secondary judgment unit 40B. Judgment sensitivity is used when making judgments by the secondary judgment unit 40B. This judgment sensitivity includes a filter level that indicates the degree to which successive intermediate images or defective product images are tolerated when making judgments by the primary judgment unit 40A, i.e., the allowable number of times.

[0061] If the number of consecutive intermediate images or defective product images exceeds the allowable number, the secondary judgment unit 40B classifies the intermediate image judged by the primary judgment unit 40A as a defective product image, whereas if the number of consecutive intermediate images or defective product images is equal to or less than the allowable number, the secondary judgment unit 40B classifies the intermediate image judged by the primary judgment unit 40A as a good product image.

[0062] 7, the filter level is set to "2." Therefore, if the number of consecutive intermediate images or defective product images exceeds two, the secondary judgment unit 40B classifies the intermediate image judged by the primary judgment unit 40A as a defective product image, whereas if the number of consecutive intermediate images or defective product images is two or less, the secondary judgment unit 40B classifies the intermediate image judged by the primary judgment unit 40A as a non-defective product image.

[0063] A more detailed explanation follows. Image 1 is classified as a non-defective image in both the primary and secondary judgments because the read character string matches the correct character string. Image 2 is classified as an intermediate image in the primary judgment because a spot stain causes a foreign substance to cover part of the character string, and the read character string differs from the correct character string. However, since the number of consecutive intermediate or defective images in the primary judgment is two or less, it is classified as a non-defective image in the secondary judgment. Image 3 is similar to image 1, so it is classified as a non-defective image in both the primary and secondary judgments.

[0064] Image 4 is classified as an intermediate image in the primary judgment because part of the drawing is missing and the read character string differs from the correct character string, but is classified as a good image in the secondary judgment because the number of consecutive intermediate images or defective images is two or less. Image 5, like image 4, is also missing part of the drawing and the read character string differs from the correct character string, so is classified as an intermediate image in the primary judgment. In the secondary judgment, image 4, which was acquired previously, is an intermediate image, and a second intermediate image was input following this intermediate image as image 5, so the number of consecutive intermediate images is two, and because the number of consecutive intermediate images is two or less, it is classified as a good image in the secondary judgment.

[0065] Like images 4 and 5, image 6 is also missing a portion of its drawing, and the read character string differs from the correct character string, so it is classified as an intermediate image in the primary determination. In the secondary determination, images 4 and 5 acquired previously are both intermediate images, and image 6 is also an intermediate image following images 4 and 5, so the number of consecutive intermediate images exceeds the allowable number of two. In this case, image 6 is classified as a defective image in the secondary determination by secondary determination unit 40B. That is, when image 4 (first intermediate image) and image 5 (second intermediate image) are acquired in chronological order by imaging unit 3, secondary determination unit 40B classifies image 6 (third intermediate image), acquired after image 5, as a defective image.

[0066] Image 7 is similar to image 1, so it is classified as a good image in both the first and second judgments. Image 8 does not contain any characters, so it is classified as a bad image in both the first and second judgments, regardless of the judgment sensitivity. Image 9 is similar to image 1, so it is classified as a good image in both the first and second judgments.

[0067] Image 10 is classified as an intermediate image in the primary judgment because it does not contain two characters and the read string differs from the correct string, but is classified as a good image in the secondary judgment because the number of consecutive intermediate images or defective images is two or less. Image 11 is similar to image 1, so it is classified as a good image in both the primary and secondary judgments.

[0068] The above describes a method for counting the number of consecutive images with poor rendering quality, in which if the image is classified as either an intermediate image or a defective image in the primary judgment, the count of the number of consecutive images is incremented by 1. However, it may also be possible to count the number of consecutive images by incrementing 1 only if the image is classified as an intermediate image in the primary judgment. In this case, intermediate images obtained consecutively after a defective image and any further intermediate images obtained consecutively thereafter can be classified as defective images in the secondary judgment regardless of the filter level.

[0069] When images classified as intermediate images are not consecutive, as in images 2 and 10, possible causes are spot stains or temporary misalignment, which have little impact on the print quality. When spot stains or temporary misalignment occur, the intermediate images are not consecutive, so the intermediate images can be classified as non-defective images in the secondary judgment. On the other hand, as in images 4 to 6, a continuous missing portion of the scanned character string can be caused by a situation that continuously affects the print quality, such as a nozzle failure in the inkjet printing device 1. When a nozzle failure occurs in the inkjet printing device, the intermediate images are consecutive, so the intermediate images can be classified as defective images in the secondary judgment. In this way, by including a filter level in the judgment sensitivity of the pass / fail judgment unit 40d, it is possible to reduce the number of images that are judged as defective due to causes that have little impact on the print quality, thereby reducing unnecessary waste.

[0070] In the example shown in Figure 7, the strictness of the inspection is determined by the number of string matches allowed in the primary judgment and the filter level included in the judgment sensitivity in the secondary judgment. For example, the smaller the filter level value, the stricter the inspection, and conversely, the larger the filter level value, the looser the inspection. Also, the smaller the number of string matches allowed, the stricter the inspection, and conversely, the larger the number of string matches allowed, the looser the inspection.

[0071] The acceptable number of character string matches may be set as an absolute value of the number of mismatched characters, such as "mismatch of a certain number of characters or more," or as a percentage of the degree of mismatch (for example, a percentage), such as "mismatch of a certain percentage or more." Furthermore, if the inspection criteria are to be relaxed beyond just character match / mismatch, the acceptable number may be the "number of characters that can be read." For example, in the case of a drawing containing eight characters, an image may be deemed good if eight blocks (blobs) of print are recognized. Furthermore, the judgment may be based on the number of characters, such as "a good image if the number of characters is 8 characters ± 2 characters," with a tolerance for error in the judgment.

[0072] In short, if you want to make the inspection stricter, you should reduce the filter level value and reduce the number of allowed string matches. Conversely, if you want to make the inspection more lenient, you should increase the filter level value and increase the number of allowed string matches. In this way, being able to change the judgment sensitivity and the number of allowed string matches together increases user convenience. However, it is also possible to make the filter level and the number of allowed string matches changeable separately.

[0073] 8 shows a change window 95 for changing the allowable number of string matches and the filter level. The change window 95 is provided with a character recognition sensitivity change area 95a for changing the character recognition sensitivity. The user can change the character recognition sensitivity to multiple levels, such as "low," "normal," and "strict," by operating the character recognition sensitivity change area 95a. Note that the character recognition sensitivity may be input as a numerical value, as with the allowable number described above.

[0074] The change window 95 is also provided with a filter level change area 95b for changing the filter level. The number input into the filter level change area 95b is accepted as the allowable number of character string matches. The change result in the character recognition sensitivity change area 95a and the input operation into the filter level change area 95b are accepted by the controller-side operation unit 42b. Therefore, the controller-side operation unit 42b is an input unit that accepts the user's input of the filter level and the allowable number of character string matches. The filter level and the allowable number of character string matches accepted by the controller-side operation unit 42b are stored in the memory unit 41. The pass / fail judgment unit 40d obtains the filter level and the allowable number of character string matches stored in the memory unit 41 and applies them during the primary judgment and the secondary judgment.

[0075] When multiple drawing elements are set, the judgment sensitivity can be changed for each drawing element. For example, when the change in judgment sensitivity for a first drawing element is completed, the changed judgment sensitivity is stored in the storage unit 41 in association with the first drawing element. Thereafter, when the change in judgment sensitivity for a second drawing element is completed, the changed judgment sensitivity is stored in the storage unit 41 in association with the second drawing element. In this way, the judgment sensitivity is stored in the storage unit 41 for each drawing element. The quality judgment unit 40d judges the quality of the drawing state for each drawing element based on the corresponding judgment sensitivity. For example, the number indicating the expiration date can be inspected strictly because it is important information, while the lot number can be inspected more loosely because it is for internal identification purposes.

[0076] The flowchart in Figure 9 shows the processing of the inspection tool executed to determine the quality of the drawing state for each drawing element based on the judgment sensitivity. Determining whether the drawing state of a drawing element is good or bad can be achieved in the same manner as described above with reference to the flowchart for determining the quality of an image based on the flowchart in Figure 4. That is, in step SD1, the filter level n0 is acquired. In step SD2, the current count value is acquired. In step SD3, the primary judgment unit 40A classifies the drawing state of the image to be inspected into three types: good drawing state, intermediate drawing state, and poor drawing state.

[0077] If the initial determination results in a good drawing state, the process proceeds to step SD4, where the count value is set to 0, and the inspection object is finally determined to be in a good drawing state.

[0078] If the initial judgment results in classification as an intermediate drawing state, the process proceeds to step SD5, where 1 is added to the current count value n, and then to step SD7, where the count value n is compared with the filter level n0. If the count value is equal to or less than the filter level, it is determined that the number of consecutive intermediate drawing states is acceptable, and the process proceeds to step SD8, where the final judgment result is classification as a good drawing state. On the other hand, if the count value is greater than the filter level, it is determined that the number of consecutive intermediate drawing states exceeds the acceptable number, and the process proceeds to step SD9, where the final judgment result is classification as a poor drawing state.

[0079] If the image is classified as a poor drawing state in the primary judgment at step SD3, the process proceeds to step SD6, where the count value is set to filter level value n0, and the process proceeds to step SD9, where the image to be inspected is classified as a poor drawing state as the final judgment result.If an image obtained in time series after an image classified as a poor drawing state in the primary judgment is classified as an intermediate drawing state in the primary judgment, the count value becomes larger than filter level n0, and the image is classified as a poor drawing state as the final judgment result regardless of the filter level value.

[0080] The image inspection when a plurality of drawing elements are arranged will be described in detail below with reference to the flowchart shown in Fig. 10. The processing of this flowchart is executed by the controller 4.

[0081] In step SE1, an initial value of 1 is set for the count value. In step SE2, an inspection tool corresponding to the drawing element corresponding to the current count value is executed, and in step SE3, the drawing state is judged as good or bad. If the drawing state of the drawing element is bad in step SE3, the process proceeds to step SE4, where the image to be inspected is classified as a defective image. If the drawing state is good in step SE3, the process proceeds to step SE5, where it is judged whether or not it is the last drawing element. If it is not the last drawing element, the process proceeds to step SE6, where the count value is incremented by 1 and the inspection tool is executed on the next drawing element. This process is repeated until the last drawing element is reached, and if the drawing states of all drawing elements are good, the image is classified as a good image in step SE7.

[0082] The judgment sensitivity may include a score indicating the similarity of characters in the matching and recognition of character strings. As described above, when the pass / fail judgment unit 40d performs matching and recognition of character strings based on the drawing content set by the drawing setting unit 40a and the captured image captured by the imaging unit 3, it calculates and stores a score indicating the similarity of characters in the matching and recognition of character strings. By including this score in the judgment sensitivity, it is possible to classify the captured image based on the score.

[0083] When the image to be inspected is classified as a defective image by the determining unit 40d, a method for determining whether the working distance is the cause and notifying the user to adjust the working distance as necessary will be described in detail below based on the flowchart shown in Fig. 11. In step SF1, the control unit 40 of the controller 4 acquires the distance (working distance) between the imaging unit 31 and the workpiece W to be drawn. The distance between the imaging unit 31 and the workpiece W to be drawn may be measured and input by the user, may be set in advance, or may be estimated from the amount of lens drive by the AF motor 31c.

[0084] In step SF2, the control unit 40 determines whether the distance between the imaging unit 3 and the workpiece W is equal to or greater than a predetermined distance set as a judgment threshold. If the result of the judgment in step SF2 is that the distance between the imaging unit 3 and the workpiece W is equal to or greater than the predetermined distance, the process proceeds to step SF3, where the controller-side display unit 42a displays the judgment result of the pass / fail judgment unit 40d together with the fact that the working distance is equal to or greater than the predetermined distance to notify the user. The controller-side display unit 42a is a notification unit that notifies the user that the working distance is equal to or greater than the predetermined distance together with the judgment result of the pass / fail judgment unit 40d.

[0085] The above-described embodiments are merely illustrative in all respects and should not be construed as limiting. Furthermore, all modifications and variations within the scope of the claims are within the scope of the present invention. In other words, in addition to character recognition, the present invention can also be applied to the inspection of barcodes and two-dimensional codes drawn using an inspection device 3. Barcodes and two-dimensional codes have a "grade" that indicates their reading quality. Users may expect a certain level of code print quality to be achieved and set the code inspection threshold to "grade C or higher." In this case, if the read result grade is below a certain level (e.g., grade D), the image is classified as an intermediate image in the initial judgment, just as in the case of character strings. If the code cannot be read at all, the image is classified as a defective product in the initial judgment. For barcodes and two-dimensional codes, the grade is used as the judgment sensitivity. Setting the judgment sensitivity in the same way as for character strings reduces wasted time when drawing and inspecting barcodes and two-dimensional codes. [Industrial Applicability]

[0086] As described above, the present disclosure can be used when drawing various types of information on packaging materials such as cardboard. [Explanation of symbols]

[0087] 1. Inkjet printing device 2 drawing machine 3 Inspection machine (image inspection section) 20a Ink cartridge (ink storage section) 24 Print head (ejection part) 31 Imaging unit 40a Drawing settings section 40b Drawing data generation unit 40d Good / bad judgement section 40A Primary judgment section 40B Secondary judgment section 41 Storage section

Claims

1. An inkjet printing device that ejects ink to draw information on an object and determines whether the drawing state is good or bad by capturing an image, an ink storage section that stores the ink; a discharge unit that discharges the ink supplied from the ink storage unit toward an object to be drawn; an imaging unit that captures an image of the object to be drawn and generates a captured image; a storage unit that stores a judgment sensitivity indicating a degree of tolerance for drawing distortion; and a quality judgment unit that judges the quality of the drawing state drawn on the object to be drawn using the captured image based on the judgment sensitivity stored in the memory unit.

2. 2. The inkjet printing apparatus according to claim 1, The inkjet printing device, wherein the pass / fail judgment unit includes a primary judgment unit that classifies the captured image into a pass image having a good drawing state, a defective image having a poor drawing state, and an intermediate image having a drawing state that is not poor but does not meet the criteria for being judged as good, and a secondary judgment unit that classifies the intermediate image into either a pass image or a defective image based on the judgment result of the primary judgment unit and the judgment sensitivity.

3. 3. The inkjet printing apparatus according to claim 2, An inkjet printing device, wherein when a first intermediate image and a second intermediate image are acquired in chronological order by the imaging unit, the secondary judgment unit classifies a third intermediate image acquired after acquiring the second intermediate image as a defective image.

4. 3. The inkjet printing apparatus according to claim 2, the judgment sensitivity includes an allowable number of times that intermediate images or defective product images are allowed to continue in the judgment by the primary judgment unit, The secondary judgment unit classifies the intermediate image as a defective image if the number of consecutive intermediate images or defective product images exceeds the allowable number, and classifies the intermediate image as a non-defective image if the number of consecutive intermediate images or defective product images is equal to or less than the allowable number.

5. 3. The inkjet printing apparatus according to claim 2, the judgment sensitivity includes an allowable number of times that successive intermediate images are allowed in the judgment by the primary judgment unit, The secondary judgment unit classifies the intermediate image as a defective image if the number of consecutive intermediate images exceeds the allowable number, and classifies the intermediate image as a non-defective image if the number of consecutive intermediate images is equal to or less than the allowable number.

6. 5. The inkjet printing apparatus according to claim 4, a drawing setting unit that sets a character string to be drawn on the object to be drawn as drawing content; the quality determination unit executes a comparison and recognition of a character string based on the drawing content set by the drawing setting unit and the captured image captured by the imaging unit; In the inkjet printing apparatus, the judgment sensitivity further includes a score indicating the similarity of characters in matching and recognition of character strings.

7. 7. The inkjet printing apparatus according to claim 6, When a first candidate character and a second candidate character that has a lower similarity than the first candidate character are obtained by character string matching and recognition, and the second candidate character is included in the drawing content set by the drawing setting unit, the pass / fail judgment unit gives the second candidate character a higher score than the first candidate character.

8. 2. The inkjet printing apparatus according to claim 1, further comprising an input unit for receiving an input of a judgment sensitivity by a user; The storage unit stores the judgment sensitivity received by the input unit.

9. 2. The inkjet printing apparatus according to claim 1, an inkjet printing device further comprising a notification unit that, when the judgment result of the pass / fail judgment unit is a defective image and the distance between the imaging unit and the object to be drawn is equal to or greater than a predetermined distance, notifies a user of the judgment result and that the distance between the imaging unit and the object to be drawn is equal to or greater than a predetermined distance.

10. 2. The inkjet printing apparatus according to claim 1, Further comprising a drawing setting unit capable of setting a plurality of drawing elements; the storage unit stores a judgment sensitivity for each of the drawing elements; The quality determination unit determines the quality of the drawing state for each drawing element based on the determination sensitivity.

Images