JP6665458B2 - Identification device and identification method - Google Patents

Description

  The present invention relates to a substrate identification device and an identification method.

  Electronic devices such as in-vehicle products are required to have a traceability (hereinafter, traceability) mechanism from the viewpoint of reliability. In addition, in a production line of electronic devices, a mechanism for linking data to a product is required for analysis of information related to devices and inspections included in the production line for the purpose of improving quality.

  When manufacturing and selling circuit boards such as printed wiring boards, traceability is required for purposes such as manufacturing process management, quality inspection, shipping inspection, and sales management of board products. Therefore, individual identification information such as a product name, a product number, and a date of manufacture is set for each circuit board, and each circuit board is tracked based on the information.

  As a setting method of the individual identification information, there is a method of attaching a label such as a bar code or a QR code (registered trademark) on which the individual identification information is printed, or an RFID (Radio Frequency Identifier) storing the individual identification information to a circuit board. . There is also a method of printing individual identification information directly on a substrate using a laser marker, an ink jet, or the like. In these methods, in order to identify a circuit board, individual identification information for identifying each circuit board is given to the circuit board.

  By the way, when the individual identification information is given to the substrate by a label or printing, a label for printing the individual identification information or a printing facility for printing the individual identification information on the substrate is required, so that the manufacturing cost increases. There is a problem. In addition, there is a problem in that when an operation of attaching the individual identification information to the substrate or printing it is required, the manufacturing time increases.

  Therefore, there is a need for an identification device, an identification method, and a traceability system that can identify a substrate at any stage after the manufacture of the substrate itself without giving individual identification information for identifying the substrate to the substrate.

  Patent Literature 1 measures the positions of a plurality of measurement targets arranged on a circuit board, acquires a difference between a measurement value of the measurement target and a design value as position information for each measurement target, and acquires the acquired position information. A circuit board identification device for registering a combination of as a substrate identification code is disclosed. According to the identification device of Patent Literature 1, it is possible to identify individual substrates without giving individual identification information to the substrates.

  Further, Patent Literature 2 discloses a technique in which a side surface of a multilayer printed wiring board is imaged by a camera, and a feature amount is extracted from the imaged image data. In Patent Document 2, board identification is performed using a board side pattern.

JP 2013-69838 A JP 2009-140375 A

  The method of identifying a substrate by imaging the side surface of the substrate as in Patent Literature 2 has the following problems when components are mounted on the substrate in a manufacturing line such as an SMT (Surface Mount Technology) line.

  The first problem is that it is necessary to consider the effect on productivity. If a camera is installed in the substrate transport direction of the SMT line, it is necessary to stop the substrate and take an image without fail, resulting in a decrease in productivity. In addition, when stopping the substrate transfer on the SMT line, it is necessary to change the sequence of the existing equipment.

  The second problem is that it is necessary to consider that the imaging position is different between the front surface and the back surface of the substrate. When parts are mounted on both sides of the board, the parts are turned upside down and pass through the SMT line twice. In order to perform board identification, it is necessary to take an image of the same location on the side of the board. However, if the camera is installed at a fixed position, the positional relationship between the front and back sides changes. In addition, since the size of the substrate is different for each product type, the change in the positional relationship on the side surface also differs depending on the size of the substrate. If cameras are installed at a plurality of locations in order to solve these problems, there is a problem that the manufacturing cost is increased and that it takes time and effort to change the camera position depending on the board type.

  An object of the present invention is to provide an identification device that can identify a substrate with high accuracy without impairing productivity.

  The identification device of the present invention is characterized by a registration imaging unit that captures two opposing side surfaces of a registration target substrate from a direction perpendicular to the transport direction, and image data of the registration target substrate captured by the registration imaging unit. After extracting the amount, the extracted feature amount of the registration target board is stored in association with the board information, and the feature amount extracted from the image data of the matching target board is compared with the stored feature amount of the registration target board. And a board identification unit for identifying the board to be checked.

  An identification device according to an aspect of the invention includes a matching imaging unit configured to capture an image of a side surface of a matching target substrate from a direction perpendicular to a transport direction, and a feature amount extracted from image data of the matching target substrate captured by the matching imaging unit. And a board identification unit that identifies the board to be matched by comparing the feature amounts extracted from image data obtained by imaging two opposite side surfaces of the board to be registered from a direction perpendicular to the transport direction.

  In the identification method of the present invention, two opposite sides of the registration target board are imaged in a direction perpendicular to the transport direction, a feature amount is extracted from the image data of the registration target board, and the extracted registration target board is extracted. The feature amount is stored in association with the board information, the side surface of the matching target board is imaged from a direction perpendicular to the transport direction, the feature amount is extracted from the image data of the matching target board, and the extracted feature of the matching target board is extracted. The matching target board is identified by comparing the quantity with the stored feature quantity of the registration target board.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the identification device which can identify a board | substrate accurately, without impairing productivity.

It is a block diagram showing composition of an identification device concerning an embodiment of the present invention. It is a conceptual diagram showing an example of an SMT (Surface Mount Technology) line in which an identification device concerning an embodiment of the present invention is installed. FIG. 2 is a conceptual diagram of a configuration of a registration imaging unit included in the identification device according to the embodiment of the present invention as viewed from above. It is the conceptual diagram which looked at the structure of the imaging part for registration with which the identification device concerning embodiment of this invention is provided from the side. It is a conceptual diagram showing an example of arrangement of a camera in an image pick-up part for registration with which an identification device concerning an embodiment of the present invention is provided. It is a conceptual diagram showing an example of arrangement of a camera in an image pick-up part for registration with which an identification device concerning an embodiment of the present invention is provided. It is a conceptual diagram showing an example of arrangement of a camera and a sensor in an image pick-up part for registration with which an identification device concerning an embodiment of the present invention is provided. FIG. 3 is a conceptual diagram for describing an imaging region when imaging a side surface of a substrate in the embodiment of the present invention. It is the conceptual diagram which looked at the structure of the image pick-up part for verification with which the identification device concerning embodiment of this invention is provided from the side. It is a conceptual diagram showing an example of arrangement of a camera in an image pick-up part for verification with which an identification device concerning an embodiment of the present invention is provided. It is a conceptual diagram showing an example of arrangement of a camera in an image pick-up part for verification with which an identification device concerning an embodiment of the present invention is provided. It is a conceptual diagram showing an example which arranges a camera and a sensor which constitute an identification device concerning an embodiment of the present invention in an SMT line. It is a block diagram showing composition of a board discernment part with which an identification device concerning an embodiment of the present invention is provided. It is a block diagram showing composition of a substrate tracking system provided with an identification device concerning an embodiment of the present invention. 6 is a flowchart illustrating a flow of an imaging process of a side surface of the substrate by the identification device according to the embodiment of the present invention. 6 is a flowchart illustrating a flow of a board registration process by a board identification unit included in the identification device according to the embodiment of the present invention. It is a flowchart which shows the flow of the board | substrate collation process by the board | substrate identification part with which the identification device which concerns on embodiment of this invention is provided. FIG. 3 is a conceptual diagram illustrating a frame constituting image data of a captured side surface of a substrate in the embodiment of the present invention. FIG. 3 is a conceptual diagram for describing a substrate to be registered or collated in an embodiment of the present invention. It is a conceptual diagram for explaining the direction of the board | substrate used as registration or collation object in embodiment of this invention. It is a conceptual diagram for explaining the direction of the board | substrate used as registration or collation object in embodiment of this invention. FIG. 3 is a conceptual diagram for explaining a relationship between a substrate transport speed and a captured image in the identification device according to the embodiment of the present invention. FIG. 3 is a conceptual diagram illustrating an example of a hardware configuration for realizing the identification device according to the embodiment of the present invention.

  An embodiment for carrying out the present invention will be described below with reference to the drawings. However, the embodiments described below have technically preferable limitations for carrying out the present invention, but do not limit the scope of the invention. In all the drawings used in the description of the following embodiments, the same parts are denoted by the same reference numerals unless otherwise specified. In the following embodiments, repetitive description of similar configurations and operations may be omitted.

(Embodiment)
(Constitution)
First, the configuration of the identification device 1 according to the embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a functional configuration of the identification device 1 according to the present embodiment.

  The identification device 1 according to the present embodiment includes a registration imaging unit 10, a collation imaging unit 20, and a board identification unit 30.

  The registration imaging unit 10 images two opposing side surfaces of the registration target substrate from a direction perpendicular to the transport direction. That is, the registration imaging unit 10 captures a captured image for registration (also referred to as a registered captured image).

  The matching imaging unit 20 captures an image of the side surface of the matching target substrate from a direction perpendicular to the transport direction. That is, the collation imaging unit 20 captures a captured image for collation (also referred to as a collated captured image).

  The board identification unit 30 extracts a feature amount from the image data of the registration target board imaged by the registration imaging unit 10 and stores the extracted feature quantity of the registration target board in association with the board information. Then, the board identifying unit 30 extracts a feature amount from the image data of the matching target board imaged by the matching imaging unit 20, and extracts the extracted feature quantity of the matching target board and the stored feature quantity of the registration target board. Is compared to identify the matching target substrate. That is, the board identifying unit 30 extracts a feature amount from the captured image, and registers or matches a board using the extracted feature amount.

  FIG. 2 is an example of an SMT (Surface Mount Technology) line in which the identification device 1 according to the present embodiment is installed. A general SMT line transports a substrate 100 to a plurality of SMT devices 101 (also referred to as surface mounting devices) such as a solder printing device, a post-printing inspection device, a component mounting device, a reflow device, and a visual inspection device. And a transfer device 110 for performing the operation. The transport device 110 includes at least a conveyor belt 111 for transporting the substrate 100 to the SMT device 101, and a conveyor guide 112 for installing the conveyor belt 111. The substrate 100 is transferred to the SMT device 101 by the transfer device 110, and a predetermined process is performed in each SMT device 101. The identification device 1 according to the present embodiment can support both single-sided mounting and double-sided mounting.

  Hereinafter, a specific configuration of the identification device 1 will be described.

(Registration imaging unit)
FIG. 3 is a conceptual diagram of an SMT line on which the identification device 1 is arranged as viewed from above. FIG. 4 is a conceptual diagram of the SMT line viewed from the side. As illustrated in FIG. 3, the registration imaging unit 10 includes two cameras 11, a sensor 13, and a registration control unit 15. In the present embodiment, reference numerals are given without distinguishing between the two cameras 11.

  The two cameras 11 (also referred to as first cameras) are arranged so as to image two opposing side surfaces of the substrate 100 being transported from a direction perpendicular to the transport direction. The two cameras 11 only need to be installed at positions where the side surface of the substrate 100 can be imaged, and may be installed so as to image the side surface of the substrate 100 obliquely.

  As shown in FIG. 5, the two cameras 11 can be installed so as to capture an image of the side surface of the substrate 100 from a gap between the transport device 110 and the SMT device 101. Also, as shown in FIG. 6, the two cameras 11 form a gap (also called a notch) by notching a part of the transfer device 110, and take an image of the side surface of the substrate 100 from the gap. You may comprise.

  Further, the two cameras 11 may be staggered so as not to face each other. As shown in FIG. 7, when the two cameras 11 are installed shifted from each other, it is preferable to install the sensors 12 (12-1 and 2) so as to correspond to each of the two cameras 11. In the case where the lighting is installed on the camera 11, by displacing the camera 11, the imaging can be performed without being affected by the mutual lighting.

  The camera 11 may be installed on the entrance side or the exit side of the SMT device 101. When the SMT line is configured by a plurality of SMT devices 101, the camera 11 used for registering the substrate 100 may be installed at the entrance side of each SMT device 101. Actually, it is not necessary to register each time the substrate 100 is transported to the SMT device 101. Therefore, even if the two cameras 11 are installed only in the first device among the SMT devices 101 constituting the SMT line, Good.

  The sensor 13 (also referred to as a first detection unit) is arranged at a position where the substrate 100 is detected. The sensor 13 is a detection unit that detects the presence or absence of the substrate 100 within the field of view of the camera 11. When detecting the substrate 100, the sensor 13 sends a detection signal for notifying that the substrate 100 has been detected to the registration control unit 15. That is, the sensor 13 outputs a detection signal when the substrate 100 is detected in the imaging range of the camera 11.

  For example, as shown in FIG. 4, the sensor 13 may be disposed at a position where the substrate 100 is imaged from above. Note that the sensor 13 may be arranged so as to detect the substrate 100 not from a position where the substrate 100 is detected from the upper main surface direction, but from a lower position, a side position, or an oblique direction.

  The sensor 13 only needs to detect the substrate 100 that has reached the detection location. For example, the sensor 13 can be realized by an optical sensor that detects an object using light, such as a photomicrosensor, a photoelectric sensor, a proximity sensor, a general camera, or an infrared camera. Further, for example, the sensor 13 may be configured as a sensor that detects an object using a sound wave, an electromagnetic wave, magnetism, or the like. Further, for example, the sensor 13 may be configured by a contact detection sensor.

  In the present embodiment, an example in which the presence or absence of the substrate 100 is performed by the sensor 13 is described. However, the substrate 100 may be detected by performing image processing on image data captured by the camera 11 instead of the sensor 13. Good. Further, a configuration may be adopted in which a moving image obtained by photographing a passing portion of the substrate 100 conveyed on the conveying path of the conveying device 110 or the SMT device 101 is read in real time, and the presence or absence of the substrate 100 is detected using the moving image. In this case, the setting may be made so that the detection of the substrate 100 is used as a trigger and the camera 11 continuously captures the side surface of the substrate 100. When detecting the substrate 100 using the moving image captured by the camera 11, the camera 11 has a function of a detection unit. That is, the registration imaging unit 10 may be configured to detect the substrate 100 by performing image processing on image data captured by the camera 11.

  The registration control unit 15 (also referred to as a first control unit) is connected to the camera 11 and the sensor 13 and controls the camera 11 to capture an image of the substrate 100 when the sensor 13 detects the substrate 100. That is, when the registration control unit 15 receives a detection signal notifying the detection of the substrate 100 from the sensor 13, the registration control unit 15 controls the camera 11 to image the substrate 100. In other words, the registration control unit 15 controls the camera 11 to take an image according to the detection signal from the sensor 13. Then, the registration control unit 15 outputs the image data captured by the camera 11 to the board identification unit 30.

  The registration control unit 15 continuously captures both side surfaces of the substrate 100 while the substrate 100 passes through the imaging range of the camera 11. At this time, the registration control unit 15 acquires an image captured by the camera 11 as image data (moving image). Image data acquired as a moving image is composed of a plurality of frames. In the present embodiment, an example in which a moving image is captured will be described. However, a configuration in which only a still image is captured may be employed. Alternatively, both the moving image and the still image may be acquired, and the moving image and the still image may be combined to identify the substrate 100.

  FIG. 8 is a conceptual diagram illustrating a side surface of the substrate 100. In FIG. 8, it is assumed that the substrate 100 is being transported rightward.

Registration control section 15, when the substrate 100 is detected by the sensor 13, controls the camera 11 to image the imaging range R _1. Then, the registration control unit 15 controls the camera 11 to image the imaging range R ₂ including a part of the imaging range R _1. Thereafter, the registration control unit 15 controls the camera 11 so as to sequentially capture the imaging ranges R ₃ ,..., R _n−2 , R _n−1 , and R _n (n is an arbitrary natural number). . In the present embodiment, the registration control unit 15 controls the camera 11 to image the side surface of the substrate 100 so that adjacent imaging ranges overlap. In other words, the registration control unit 15 causes the camera 11 to continuously capture images such that the same portion of the side surface of the substrate 100 is included in a continuous frame among a plurality of frames constituting the image data captured by the registration imaging unit 10. Control.

Results of a series of imaging processing, the registration control unit 15 acquires the image data (moving image) to the frame images of each imaging range R ₁ to R _n.

(Imaging unit for verification)
Next, the configuration of the matching imaging unit 20 will be specifically described with reference to FIG. The matching imaging unit 20 includes a camera 21, a sensor 23, and a matching control unit 25. The matching imaging unit 20 differs from the registration imaging unit 10 in that the camera 21 can be configured as a single camera. Note that the matching imaging unit 20 may be configured to include a plurality of cameras 21. In addition, the camera 21 of the imaging unit for registration 10 may be diverted for the camera 21 of the imaging unit for verification 20.

  The camera 21 (also referred to as a second camera) is arranged to capture an image of the side surface of the substrate 100 being transported from a direction perpendicular to the transport direction. The camera 21 may be disposed in a gap between the transport device 110 and the SMT device 101 as shown in FIG. Further, the camera 21 may be arranged in a gap formed by cutting out a part of the conveyor guide 112 as shown in FIG.

  The camera 21 may be installed on the entrance side or the exit side of the SMT device 101. When the SMT line is configured by a plurality of SMT devices 101, it is preferable that the camera 21 used for collation of the substrate 100 be installed at the entrance side of each device. However, when there is no need to perform collation between the SMT devices 101 constituting the SMT line, the camera 21 does not need to be installed between the devices. Further, even when the SMT device 101 is configured by a plurality of devices, it is assumed that the collation does not have to be performed every time the substrate 100 is transported to each device. In such a case, the camera 21 may be installed only on the entrance side of the device that performs the first step of the SMT device 101.

  FIG. 12 is an example in which the registered imaging camera 11 and the collation imaging camera 21 are arranged on an SMT line constituted by a plurality of SMT devices 101 (101-1, 2,..., N) (n is 2). Natural numbers above). Two cameras 11 for registered imaging are arranged at the entrance side of the SMT device 101-1 which performs the first step among the SMT devices 101 constituting the SMT line. .., N, which are responsible for the second and subsequent steps, of the SMT device 101 constituting the SMT line, one camera 21 for collation imaging may be arranged at the entrance side.

  Note that two cameras 11 for registered imaging may be arranged at the entrance side of any one of the SMT devices 101 after the second step. Further, the configuration may be such that the collation of the substrate 100 is performed based on an image captured by the camera 11 disposed at the entrance of the first step. When it is not necessary to perform registration or collation at the entrance side of the SMT device 101, the camera 11 or 21 may be omitted.

  Practically, it is only necessary that the registration and collation of the substrate 100 can be performed based on the image captured by the camera 11 or 21 disposed at the entrance of each SMT device 101. Therefore, if the configuration is such that the substrate 100 is registered by the registration imaging unit 10 and the substrate 100 is collated by the collation imaging unit 20, using an image captured by any camera arranged at the entrance of each SMT device 101. Good. That is, the camera arranged on the entrance side of the SMT device 101 functions as the camera 11 (first camera) when used for registering the board 100, and functions as the camera 21 (second camera) when used for collation of the board 100. I just need.

  The sensor 23 (also referred to as a second detection unit) detects the presence or absence of the substrate 100 within the field of view of the camera 21. That is, the sensor 23 outputs a detection signal when the substrate 100 is detected in the imaging range of the camera 21. Note that the arrangement and function of the sensor 23 are the same as those of the sensor 13 of the registration imaging unit 10, and a detailed description thereof will be omitted. Further, the sensor 23 of the imaging unit for registration 10 may be diverted for the sensor 23 of the imaging unit for collation 20. That is, the matching imaging unit 20 may be configured to detect the substrate 100 by performing image processing on image data imaged by the camera 21.

  The matching control unit 25 (also referred to as a second control unit) is connected to the camera 21 and the sensor 23. When the substrate 100 is detected by the sensor 23, the matching control unit 25 controls the camera 21 so as to perform imaging during a period in which the detected substrate 100 passes. In other words, the matching control unit 25 controls the camera 21 to take an image according to the detection signal from the sensor 23. In other words, the matching control unit 25 continuously shoots the camera 21 so that a continuous frame among a plurality of frames constituting the image data captured by the matching imaging unit 20 includes the same portion on the side surface of the substrate 100. Control. Note that the matching control unit 25 may control the camera 21 so as to image the entire side surface of the substrate 100 as in the case of the registration imaging unit 10, or may capture at least a part of the side surface of the substrate 100. May control the camera 21.

(Board identification section)
Next, the configuration of the board identifying unit 30 will be specifically described with reference to FIG. The board identification unit 30 includes a feature amount extraction unit 31, a storage unit 33, and a collation determination unit 35.

  The feature amount extraction unit 31 extracts a feature amount from image data of the side surface of the substrate 100 captured by the registration imaging unit 10 or the comparison imaging unit 20. That is, the feature amount extraction unit 31 extracts a feature amount from image data captured by any of the registration imaging unit 10 and the collation imaging unit 20. The feature amount extraction unit 31 may extract a feature amount for each of a plurality of frames constituting image data (moving image) of the side surface of the substrate 100, or may extract a feature amount composed of continuous frames. Then, the feature amount may be extracted from the entire image data (moving image).

  The feature amount extraction unit 31 extracts a pattern such as a pattern or a shape of a side surface formed by the base material and the fibers constituting the substrate 100 as a feature amount. The feature amount extraction unit 31 may extract the feature amount from the entire side surface of the substrate 100, or may locally extract the feature amount from a part of the side surface of the substrate 100. The feature amount extraction unit 31 may extract the image pattern itself as a feature amount, or may extract a digitized image pattern as a feature amount.

  When extracting the feature amount, the feature amount extracting unit 31 preferably abstracts the image by performing image processing on a captured image used for registration or collation. For example, examples of image processing include binarization and thinning, gamma correction, expansion / contraction processing, labeling, hue extraction, pseudo colorization, and the like. Further, for example, filter processing such as a moving average filter, a Gaussian filter, a median filter, a Sobel filter, an unsharp masking, and a bilateral filter may be performed as image processing. Note that the feature amount extraction unit 31 may perform image processing that is not included in the method described above. Further, the board identifying unit 30 may include an image processing unit that performs image processing when extracting a feature amount, separately from the feature amount extraction unit 31.

  The storage unit 33 stores the feature amount extracted by the feature amount extraction unit 31 from the image data (hereinafter, a registered captured image) captured by the registration imaging unit 10 and the board information of the board 100 in association with each other. The storage unit 33 stores board information such as a product type and a serial number of the board 100 in association with the board 100 from which the feature amount has been extracted.

  The type information may be configured to be input by a user, or may be configured to be received from a higher-level system such as a manufacturing execution system (hereinafter, MES: Manufacturing Execution System) not shown. Further, the serial number may be configured to be received from a higher-level system such as the MES, or may be configured to be automatically generated by the board identification unit 30.

  The collation determination unit 35 performs collation between the characteristic amount extracted by the characteristic amount extraction unit 31 from the image data (collated captured image) captured by the collation imaging unit 20 and the characteristic amount stored in the storage unit 33. The matching determination unit 35 compares the feature amount extracted from the matching data with the feature amount extracted from the registered captured image, and determines whether or not they match. In other words, the collation determination unit 35 collates the feature amounts extracted from each of the plurality of frames constituting the captured image data with the feature amounts stored in the storage unit 33, and determines the substrate having the highest matching rate. 100 are determined to be the same.

  The collation determination unit 35 collates the feature amount extracted from the collated captured image with the feature amount extracted from the registered captured image. Further, the collation determination unit 35 may perform collation by scanning the feature image extracted from the registered captured image on the collated captured image as a template. When matching is performed using template matching, the matching determination unit 35 scans a template extracted from the registered captured image or the matched captured image on the target captured image, and determines the position where the similarity is maximized by the similarity calculation. Just find it. The similarity may be determined based on the normalized mutual phase, the cross-correlation coefficient, the sum of squares and absolute value of the difference, and the like. The method of matching the substrate 100 by the substrate identification unit 30 is not limited to the method described above, and any method can be used.

  The above is the description of the configuration of the identification device 1 according to the present embodiment.

(Substrate tracking system)
Here, FIG. 14 shows a substrate tracking system including the identification device 1 of the present embodiment. As shown in FIG. 14, the board tracking system includes the identification device 1 and the production history server 2 according to the present embodiment. The identification device 1 and the production history server 2 are connected via a local area network or the Internet. Note that the identification device 1 and the production history server 2 may be connected by wire or wirelessly.

  The production history server 2 is a server that manages information such as traceability of the board 100. In order to track the production history of each substrate 100, the production history server 2 records the substrate information of each substrate 100 in association with the passage history of each process constituting the SMT device. The production history server 2 may be constituted by a general server.

(motion)
Next, the operation of the identification device 1 according to the present embodiment will be described.

(Imaging process)
First, an imaging process performed by the registration imaging unit 10 will be described with reference to FIG.

  The registration control unit 15 confirms whether or not a detection signal indicating that the substrate 100 is passing through the imaging range is output from the sensor 13 (Step S11).

  When the detection signal is output (Yes in Step S11), the registration control unit 15 captures an image of the side surface of the substrate 100 (Step S12). If the detection signal has not been output (No in step S11), the registration control unit 15 does nothing (return to step S11).

  After controlling the camera 11 to capture an image of the side surface of the substrate 100, the registration control unit 15 waits for a preset time (hereinafter, a set time) (step S13). The set time may be set in accordance with the transfer speed of the substrate 100, the imaging timing of the camera 11, and the imaging conditions of the side image of the substrate 100. In the present embodiment, the set time may be adjusted so that the imaging ranges of consecutive frames constituting the image data (moving image) overlap.

  When the detection signal has been output after the set time has elapsed (Yes in step S14), the registration control unit 15 images the side surface of the substrate 100 again (return to step S12).

  In the imaging process of FIG. 15, the identification device 1 continuously captures the side surface of the substrate 100 by repeating steps S12 to S14.

  At the stage where the detection signal is no longer output (No in step S14), the registration control unit 15 transfers all frames constituting the image data (moving image) captured continuously to the board identifying unit 30 (step S15). ). In addition, even while the detection signal is being output, the frame that has been imaged may be gradually transferred to the board identifying unit 30 when a predetermined time has elapsed.

  The registration imaging unit 10 has two cameras 11, and these cameras 11 operate at the same timing. When the sensor 13 is provided for each camera 11, the registration control unit 15 controls the corresponding camera 11 according to the detection signal of the sensor 13 corresponding to each camera 11. However, the imaging timings of the two cameras 11 may not be exactly the same time but may be shifted within a predetermined range. The registration imaging unit 10 transfers all frames constituting image data (moving images) captured by the two cameras to the board identification unit 30.

  Next, an imaging process performed by the matching imaging unit 20 will be described. However, since the imaging process by the matching imaging unit 20 is substantially the same as the imaging process by the registration imaging unit 10, detailed description will be omitted.

  The matching imaging unit 20 has a configuration in which the registration imaging unit 10 has one camera 11 (two cameras) (one camera 21). The imaging process performed by the matching imaging unit 20 is performed by a single camera 21 in the same manner as the imaging process performed by the registration imaging unit 10. The matching imaging unit 20 transfers all images (all frames) captured by one camera 21 to the board identifying unit 30.

(Substrate identification processing)
Next, the board identification processing by the board identification unit 30 will be described.

  The board identification unit 30 starts processing after receiving the image data captured by the registration imaging unit 10 or the collation imaging unit 20. At this time, the board identifying unit 30 performs different processing depending on whether the received image data is data acquired from the registration imaging unit 10 or the collation imaging unit 20.

(Substrate registration process)
First, a process (substrate registration process) when the board identifying unit 30 receives the image of the registration imaging unit 10 will be described with reference to FIG.

  After receiving the image data (moving image), the feature amount extracting unit 31 extracts a feature amount for each frame of all images (hereinafter, frames) constituting the received image data (step S21).

  The collation judging unit 35 collates the feature amount extracted by the feature amount extracting unit 31 with the feature amount stored in the storage unit 33 (Step S22).

  For example, when one surface of the imaged substrate 100 is already mounted and components are mounted on the other surface, registration information on the substrate 100 is already registered in the storage unit 33. In this case, if the feature amount extracted in step S21 is compared with the feature amount stored in the storage unit 33, board information (type, serial number, etc.) can be associated with the board 100.

  If the board 100 having the feature amount that matches the feature amount extracted in step S22 is not registered in the storage unit 33 (No in step S23), the board identifying unit 30 stores the feature amount of the board 100 in the board information. And registered in the storage unit 33.

  Then, the board identifying unit 30 transmits the board information and the passage history of the board 100 to the production history server 2 (Step S25).

  On the other hand, when the board 100 having the feature amount that matches the feature amount extracted in step S22 has already been registered in the storage unit 33 (Yes in step S23), the identification device 1 transmits the board information of the board 100 The history is transmitted to the production history server 2 (step S25).

  The above is the description regarding the board registration process of the board identification processing.

(Board collation process)
Next, a process (substrate matching process) when the board identifying unit 30 receives the image of the matching imaging unit 20 will be described with reference to FIG.

  After receiving the image data (moving image), the characteristic amount extracting unit 31 extracts a characteristic amount for each frame of all images (hereinafter, frames) constituting the received image data (step S31).

  The collation determination unit 35 performs collation between the characteristic amount extracted by the characteristic amount extraction unit 31 and the existing characteristic amount stored in the storage unit 33 (Step S32).

  Then, the identification device 1 transmits the board information and the passage history of the board 100 to the production history server 2 (Step S33).

  Here, a method of storing a feature amount of image data composed of a plurality of frames will be described. Here, it is assumed that the image data of the side surface of the target substrate 100 is composed of ten frames. Note that the method described here is an example, and does not limit the processing by the identification device of the present embodiment.

First, as shown in FIG. 18, the 10 frames constituting the image data of the side surface of the substrate A, named respectively _{IA 1, IA 2, IA 3} , ···, and IA _10. The feature amount extraction unit 31 extracts a feature amount from each frame. Each feature value extraction unit 31 stores the feature value of each frame in the storage unit 33 in association with the board information. At this time, the feature values of the frames IA ₁ to IA ₁₀ configuring the same substrate 100 are associated with common substrate information.

Next, a method of collating image data composed of a plurality of frames will be described. Here, the 10 frames constituting the image data of the side surface of the substrate X being a target for verification, respectively _{IX 1, IX 2, IX 3} , ···, and IX _10. The drawing of the substrate X is omitted.

First, the matching determination unit 35 performs the matching IX _1, to obtain a high substrate information most concordance rate from the storage unit 33. Similarly, the collation determination unit 35 also performs collation for IX _{2 to} IX ₁₀ . Then, the collation determination unit 35 determines that the substrate having the highest identification rate among IX _{1 to} IX ₁₀ is the same substrate as the substrate X. It should be noted that a general pattern matching technique may be used for matching the substrates.

  Here, an effect obtained by the identification device 1 according to the present embodiment imaging the side surface of the substrate 100 from two opposite directions will be described.

  FIG. 19 is a perspective view of the substrate 100. Here, the upper surface (main surface) of the substrate 100 being transported is referred to as a first surface, and the lower surface is referred to as a second surface. The direction in which the substrate 100 is transported is referred to as an X direction, a direction perpendicular to the main surface and the transport direction of the substrate 100 is referred to as a Y direction, and a normal direction on the first surface is referred to as a Z direction. Further, the front side of the side surface of the substrate 100 is referred to as a side surface A, and the opposite side facing the side surface A is referred to as a side surface B.

  In an actual SMT line, the substrate 100 faces different directions depending on the manufacturing process. For example, when the substrate 100 is rotated by 180 degrees with respect to the Y axis as shown in FIG. 20, the directions of the first surface and the second surface are switched, although the directions of the side surfaces with respect to the camera are not switched. In addition, for example, as shown in FIG. 21, when rotated by 180 degrees with respect to the X axis, the directions of the first surface and the second surface are switched, and the side surface B faces the near side, and the side surface A faces the opposite side. become.

  In the case as shown in FIG. 20, the notation “side A” on the side of the substrate 100 is located on the left before rotation, but is located on the right after rotation. Therefore, when only one image is captured while the transport of the substrate 100 is stopped, it is necessary to change the position where the substrate 100 is stopped or the position of the camera. In addition, since the length of the substrate 100 in the X-axis direction varies depending on the type, the position at which the substrate 100 is stopped and the position of the camera change depending on the type. On the other hand, in the present embodiment, since the entire side surface of the substrate 100 is continuously photographed, it is not necessary to change the position where the substrate 100 is stopped or the position of the camera.

  That is, in the identification device 1 according to the present embodiment, the registration imaging unit 10 can continuously image the two opposing side surfaces of the substrate 100, so that the entire surface of both side surfaces of the substrate 100 can be imaged. As a result, as shown in FIG. 20 and FIG. 21, it is possible to perform board identification regardless of how the board 100 is turned. Further, according to the identification device 1 according to the present embodiment, it is not necessary to change the setting of the camera position and the like even when the size of the substrate changes.

  In addition, by imaging the side surfaces of the same substrate 100 so that consecutive frames overlap as shown in FIG. 18, the substrate 100 can be identified even when the speed changes during the transportation of the substrate 100.

  FIG. 22 shows an imaging example in which the substrate 100 is transported at a constant speed when registered, and when the substrate 100 is collated, the transport speed changes. The substrate 100 is conveyed rightward, and is divided into a region 1, a region 2, a region 3,... From the right end of the substrate 100.

  First, imaging of a registered captured image will be described. When capturing a registered captured image, the substrate 100 is transported at a constant speed. When the area 1 approaches the imaging range of the camera, the first registered captured image is captured. Hereinafter, the registered captured images are captured in the order of the second, third,.... The registered captured images captured in this manner are captured at equal time intervals.

  Next, imaging of the collation captured image will be described. When capturing the collation captured image, the transport speed of the substrate 100 changes. When the area 1 approaches the imaging range of the camera, the first collation captured image is captured. In the example of FIG. 22, the second to sixth collated captured images are captured almost in a stopped state. Then, the seventh and subsequent collation captured images are captured while the substrate 100 is moving.

  In FIG. 22, since the registered captured image and the collated captured image include images captured in the same region, the board 100 can be identified by collating the images. For example, the second to seventh images of the collated captured image shown in FIG. 22 have a higher coincidence rate with the second registered captured image. That is, according to the present embodiment, board identification can be performed without changing existing equipment control including deceleration and stop timing of board transfer.

(effect)
As described above, in the present embodiment, since the side surface of the substrate can be continuously imaged without stopping the substrate, productivity is not affected. That is, according to the identification device of the present embodiment, it is possible to accurately identify a substrate without impairing productivity.

(hardware)
Lastly, a hardware configuration that enables the identification device according to the present embodiment will be described. FIG. 23 is a block diagram illustrating an example of a hardware configuration of the identification device according to the embodiment of the present invention. FIG. 23 shows a hardware configuration in which a processor 91, a converter 93, a main storage device 95, an input / output interface 96, an image processing circuit 97, and a storage medium 98 are connected by a bus 99. FIG. 23 is a configuration example for realizing the identification device according to each embodiment, and an arbitrary device or circuit may be added or deleted according to the specifications of the device.

  The processor 91 is an arithmetic unit that expands a program stored in the storage medium 98 or the like into the main storage device 95 or the like and executes the expanded program. The processor 91 may be constituted by a general central processing unit (CPU: Central Processing Unit). Some of the functions of the feature amount extraction unit 31 and the collation determination unit 35 of the board identification unit 30 may be assigned to the processor 91.

  The converter 93 is a circuit having a function of converting analog data input from the outside into digital data. The converter 93 has an A / D (Analog / Digital) conversion converter for converting analog data into digital data. The converter 93 may have a function of converting digital data generated internally into analog data.

  The main storage device 95 is a device having a storage area for temporarily expanding programs and data handled by the processor 91. For example, a random access memory (RAM: Random Access Memory) can be configured as the main storage device 95.

  The input / output interface 96 is an interface for exchanging data with an external or internal input / output device. For example, the input / output interface 96 is connected to cameras (11, 21) and sensors (13, 23). The input / output interface 96 may be connected to an input device such as a keyboard and a mouse (not shown), an output device such as a display and a printer (not shown), and a network connected to an external computer or server.

  The image processing circuit 97 is a circuit that performs various processes on image data. The image processing circuit 97 is, for example, an integrated circuit that performs image processing such as dark current correction, interpolation calculation, color space conversion, gamma correction, aberration correction, noise reduction, and image compression on image pickup data. Note that the image processing circuit 97 may be configured to execute processing other than the above-described image processing. When performing the image processing when the feature amount extracting unit 31 processes the image data, a part of the function may be assigned to the image processing circuit 97.

  The storage medium 98 is a medium for storing programs and data. For example, a hard disk or the like can be used as the storage medium 98. Further, a ROM (Read Only Memory) may be configured as the storage medium 98. The function of the storage unit 33 may be assigned to the main storage device 95. The recording medium may be realized by 98, for example, a semiconductor recording medium such as an SD (Secure Digital) card or a USB (Universal Serial Bus) memory. The storage medium 98 may be realized by a magnetic recording medium such as a flexible disk, an optical recording medium such as a CD (Compact Disk) or a DVD (Digital Versatile Disc), or a recording medium using other principles.

  The above is an example of the hardware configuration for enabling the identification device according to the embodiment of the present invention. The hardware configuration in FIG. 23 is an example of a hardware configuration for enabling the identification device according to the present embodiment, and does not limit the scope of the present invention. Further, a processing program that causes a computer to execute processing by the identification device according to the present embodiment is also included in the scope of the present invention. Further, a program recording medium on which a processing program according to the embodiment of the present invention is recorded is also included in the scope of the present invention.

  As described above, the present invention has been described with reference to the exemplary embodiments. However, the present invention is not limited to the above exemplary embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

REFERENCE SIGNS LIST 1 identification device 10 registration imaging unit 11 camera 13 sensor 15 registration control unit 20 collation imaging unit 21 camera 23 sensor 25 collation control unit 30 board identification unit 31 feature quantity extraction unit 33 storage unit 35 collation determination unit 100 substrate 101 SMT device 110 Transport device 111 Conveyor belt 112 Conveyor guide

Claims (10)

A registration imaging unit that images two opposing side surfaces of the registration target substrate from a direction perpendicular to the transport direction;
A feature amount is extracted from each of the image data of the two sides of the registration target board imaged by the registration imaging unit, and the extracted feature quantities of the two side faces of the registration target board are stored in association with board information. And a board that identifies the board to be compared by comparing the feature extracted from the image data of at least one side of the board to be checked with the stored feature of two sides of the board to be registered. An identification device comprising: an identification unit. A matching imaging unit that images at least one side surface of the matching target substrate from a direction perpendicular to the transport direction;
The feature amount of the registration target board extracted from the image data of the registration target board is stored in association with the board information, and is extracted from the image data of at least one side of the matching target board imaged by the matching imaging unit. a feature amount, collates the feature amount extracted from each image data of the two sides of the registration target board captured from a direction perpendicular to two opposite sides of the registration target substrate with respect to the conveying direction And a board identifying unit for identifying the matching target board.   The identification device according to claim 2, further comprising a registration imaging unit configured to image two opposing side surfaces of the registration target substrate from a direction perpendicular to a transport direction. The registration imaging unit,
At least two first cameras arranged to image two opposite side surfaces of the substrate being transported from a direction perpendicular to the transport direction;
A first detection unit that outputs a detection signal when a substrate is detected in an imaging range of the first camera; and a control that causes the first camera to perform imaging in response to a detection signal from the first detection unit. A first control unit that performs
The collation imaging unit,
At least one second camera arranged to image a side surface of the substrate being transported from a direction perpendicular to the transport direction;
A second detection unit that outputs a detection signal when a substrate is detected in an imaging range of the second camera;
A second control unit that controls the second camera to capture an image according to a detection signal from the second detection unit,
The board identification unit,
A feature amount extraction unit that extracts a feature amount from image data captured by any of the registration imaging unit and the matching imaging unit;
A storage unit for storing a feature quantity of the registered substrate which has been extracted by the feature extraction unit in association with the board information,
The image processing apparatus according to claim 1, further comprising: a matching determination unit configured to match the feature amount of the matching target board extracted from the image data captured by the matching imaging unit with the feature amount of the registration target board stored in the storage unit. 3. The identification device according to 3. The first control unit causes the first camera to perform continuous shooting so that a continuous frame among a plurality of frames constituting the image data captured by the registration image capturing unit includes the same portion on the side surface of the substrate. Take control,
The second control unit causes the second camera to perform continuous shooting such that a continuous frame among a plurality of frames constituting the image data captured by the matching imaging unit includes the same portion on the side surface of the substrate. The identification device according to claim 4, which performs control.   The identification device according to claim 4, wherein the feature amount extraction unit extracts a feature amount for each of a plurality of frames constituting image data captured by the first-time registration imaging unit and the comparison imaging unit.   The identification device according to claim 4, wherein the first and second cameras are installed so as to image a side surface of the substrate from a gap between the transport device that transports the substrate and the surface mounting device.   The identification device according to claim 4, wherein the first and second cameras are arranged so as to capture an image of a side surface of the substrate from a notch provided in a transport device that transports the substrate. The collation determination unit,
The feature amount extracted from each of the plurality of frames constituting the image data captured by the registration imaging unit or the matching imaging unit is compared with the feature amount stored in the storage unit, and a match rate is determined. The identification device according to claim 4, wherein the largest substrates are determined to be the same. Two opposite sides of the registration target substrate are imaged from a direction perpendicular to the transport direction,
Extracting a feature amount from each of the image data of the two side surfaces of the registration target substrate;
Storing the extracted feature values of the two sides of the registration target board in association with the board information;
Imaging at least one side surface of the matching target substrate from a direction perpendicular to the transport direction,
Extracting a feature amount from image data of at least one side surface of the matching target substrate;
An identification method for identifying the matching target board by comparing the extracted feature quantity of at least one side face of the matching board with the stored feature quantity of two side faces of the registration target board.

Images