JP5755935B2 - Component pitch measuring apparatus and component pitch measuring method - Google Patents

Description

  The present invention relates to a component pitch measuring apparatus and a component pitch measuring method for measuring a component pitch in a component in which components having the same shape are arranged.

  For example, tape feeders, tray feeders, bulk feeders, and the like are known as feeders that supply components to a component mounting machine. Here, the tray feeder arranges the components on the tray in a matrix (a grid pattern), and sucks the components on the tray with the suction nozzle of the component mounting machine and mounts them on the circuit board. In this case, in order to sequentially pick up the components arranged on the tray with the suction nozzle, the position coordinates of the first picked-up component, the component pitch in the X direction, the component pitch in the Y direction, the number of components in the X direction, Y Tray data including data such as the number of parts in the direction is required.

  Conventionally, data such as component pitch is manually input by an operator with reference to design data obtained from a component supplier, etc., but the dimensions of the tray, component pitch, etc. are strictly controlled. I don't mean. For this reason, there is a possibility that the part pitch input by the operator is deviated from the actual part pitch. As a result, the error in the part position increases cumulatively according to the number of parts in each row on the tray. There is a possibility that component adsorption mistakes may occur.

  In response to this problem, in the tray data creation method disclosed in Patent Document 1 (Japanese Patent Application Laid-Open No. 2001-345595), the camera is moved above the part (suction start point) to be first suctioned on the tray to start the suction. The component of the point is picked up and displayed on the display device, and the operator manually inputs the position of the component at the suction start point on the display screen, and then the component that sucks the camera at the end of the column of the component ( The final suction point) is moved above, the part at the final suction point is imaged and displayed on the display device, and the operator manually inputs the position of the part at the final suction point on the display screen, The part pitch is obtained by dividing the moving amount of the camera from the suction start point to the final suction point by the number of parts in the row.

  Further, in the tray data correction method disclosed in Patent Document 2 (Japanese Patent Laid-Open No. 2008-10594), the amount of suction position deviation of a component sucked by a suction nozzle is recognized as an image, and the component pitch and the inclination of the tray are related to the suction position deviation amount. A correction value is obtained, and preset tray data is corrected.

JP 2001-345595 A JP 2008-10594 A

  However, in the tray data creation method of Patent Document 1, the operator must manually input the suction start point and the final suction point or manually input the number of parts. There is a possibility that the wrong part pitch may be calculated by inputting the wrong number of parts due to misunderstanding or the like.

  Further, since the technique of Patent Document 2 is a technique for correcting tray data such as a component pitch created in advance according to the amount of suction position deviation, it is necessary for an operator to create tray data such as a component pitch in advance. Yes, it takes time.

  Therefore, the problem to be solved by the present invention is to provide a component pitch measuring device and a component pitch measuring method capable of automating the measurement of the component pitch.

In order to solve the above-described problem, the invention according to claim 1 is a component pitch measurement device that measures a component pitch in a component in which components having the same shape are arranged, and relatively moves the camera that images the component and the camera. A moving mechanism that moves the camera and continuously images the camera while relatively moving the camera in the arrangement direction of the parts, and evaluating the periodicity of fluctuations in the captured images and the periodicity of the fluctuations of the camera. A component pitch measuring unit that measures a component pitch based on the relationship with the relative movement amount, and the component pitch measuring unit calculates a correlation value between the captured image and the reference image of the component for each imaging of the camera. The component pitch is measured based on the relative movement amount of the camera between the peak points of the correlation value. Here, “continuous imaging” is to repeat imaging at intervals sufficiently shorter than the component pitch to be measured.

As in the present invention, when images are continuously captured while relatively moving the camera in the arrangement direction of the components, even if only one component is contained in the field of view of the camera, many images including captured images of a plurality of arranged components are included. A captured image can be acquired. These captured images fluctuate periodically with the relative movement of the camera, and fluctuate with the number of times of imaging ( relative movement amount of the camera) from imaging the previous part to imaging the next part as one period. The relative movement amount of the camera for one cycle corresponds to the component pitch. From this relationship, if the periodicity of the fluctuation of the captured image is evaluated, the component pitch can be measured from the relative movement amount of the camera for one cycle. This makes it possible to automate the measurement of the component pitch.

The present invention is, cross-correlation method used as a method of evaluating the periodicity of the variation of the captured image, it calculates a correlation value between the reference image of the captured image and the part for each imaging camera, the fluctuation of the correlation value The component pitch may be measured based on the relative movement amount of the camera between the peak points of the correlation value corresponding to one period. Here, the reference image of the component is an image obtained by capturing the same component as the component for measuring the component pitch in the center (reference position) of the camera field of view. The captured image with the peak correlation value is the captured image with the highest degree of similarity to the reference image of the component, and means that the image is actually captured with the component placed in the center of the camera's field of view. The position of the component is determined from the position of the camera that captured the captured image having the peak value. From this relationship, since the relative movement amount of the camera between the peak points of the correlation value corresponds to the component pitch, the component pitch can be measured from the relative movement amount of the camera between the peak points of the correlation value.

Using such a cross-correlation method, the correlation value peaks even under imaging conditions where the component recognition accuracy is poor due to a small difference in brightness between the component in the captured image and its background. Since the position of the component is determined from the position of the camera that captured the captured image as a point, the component pitch can be measured from the relative movement amount of the camera between the peak points of the correlation value .

  The reference image of the component used in the present invention is stored in advance in the storage means as an image obtained by placing the same component as the component whose component pitch is measured in the center of the field of view of the camera. May be. At this time, it is preferable to capture the reference image under the same imaging conditions (background, illumination conditions, camera position, etc.) as the imaging conditions at the time of component pitch measurement.

Alternatively, as described in claim 2 , when continuously capturing images while moving the camera relatively in the component arrangement direction, the captured image of the leading component may be used as the reference image. In this way, when measuring the component pitch, the captured image of the first component imaged under the same imaging conditions can be used as the reference image, and an accurate reference image can be easily acquired. At this time, if the image of the leading part cannot be recognized with high accuracy, the operator may specify the position of the leading part. For example, the position of the head part may be specified by, for example, the operator manually entering the position of the head part by looking at the captured image displayed on the display device, or design data obtained from a part supplier, etc. The coordinates of the position of the leading part may be manually input with reference to FIG.

The inventions according to claims 1 and 2 described above can be applied to various devices in which parts having the same shape are arranged, and may be applied to, for example, a tray feeder. On the tray to be set in the tray feeder, since the same shaped parts are arranged in a matrix, as claimed in claim 3, the camera are relatively moved in the X direction to measure the X direction of the component pitch, camera May be relatively moved in the Y direction to measure the component pitch in the Y direction. Thereby, it is possible to automatically measure the component pitch in the X direction and the Y direction on the tray.

In addition, claim 4 describes the technical idea substantially the same as the invention of the “component pitch measuring device” described in claim 1 as a “component pitch measuring method”.

FIG. 1 is a block diagram showing a configuration example of a component pitch measuring apparatus according to an embodiment of the present invention. FIG. 2 is a plan view showing the arrangement of components on the tray. FIG. 3 is a diagram (part 1) for explaining a method of measuring the component pitch by continuously capturing images while moving the camera in the component arrangement direction. FIG. 4 is a diagram (part 2) for explaining a method of measuring the component pitch by continuously capturing images while moving the camera in the component arrangement direction. FIG. 5 is a flowchart showing the flow of processing of the component pitch measurement program.

Hereinafter, an embodiment in which a mode for carrying out the present invention is applied to a component mounter will be described.
First, the system configuration of the component mounter will be described with reference to FIG.
The component mounter includes a control device 11 configured by a computer, an input device 12 such as a keyboard and a mouse, a display device 13 such as a liquid crystal display and a CRT, a component mounter control program, and a component pitch measurement shown in FIG. A storage device 14 that stores a program and the like, a mounting head moving device 15 (moving mechanism) that moves a mounting head that holds a suction nozzle in the XYZ directions, and a substrate transfer device that transfers a circuit board on which components are mounted 16, a parts camera 17 that picks up the part sucked by the suction nozzle from below, a mark camera 18 that picks up the reference position mark of the circuit board from above, and the like. The camera 18 is configured to move in the XYZ direction integrally with the mounting head. In the present embodiment, the mark camera 18 is used as a camera for component pitch measurement described later.

  In this component mounting machine, a feeder such as a tray feeder 19, a tape feeder, or a bulk feeder is set, and components are supplied from the feeder. As shown in FIG. 2, on the tray 20 set in the tray feeder 19, parts 21 having the same shape are arranged in a matrix. In this case, in order to suck the components 21 arranged on the tray 20 in order with the suction nozzle, the coordinates (X0, Y0) of the center position of the first component 21 to be sucked first, the component pitch in the X direction, Y Tray data including data such as the component pitch in the direction, the number of components in the X direction, and the number of components in the Y direction is required.

  Therefore, in the present embodiment, the control device 11 executes a component pitch measurement program shown in FIG. 5 to be described later, whereby the arrangement of the components 21 on the tray 20 is continuously captured by the mark camera 18 and the captured images thereof. Is used to automatically count the component pitch and the number of components in the X and Y directions by the cross-correlation method. Hereinafter, a method for counting the component pitch and the number of components will be described.

  With respect to the coordinates (X0, Y0) of the center position of the leading part 21 that is first picked up by the picking nozzle, the outer shape of the leading part 21 is more than a predetermined accuracy from the image of the leading part 21 captured by the mark camera 18. If the image can be recognized by the mark camera 18, an image obtained by capturing the leading component 21 is processed to recognize the outer shape of the leading component 21, and the coordinates (X0, Y0) of the center position of the leading component 21 are recognized. ) May be calculated.

  On the other hand, if the outer shape of the leading component 21 cannot be accurately recognized due to a small luminance difference between the component 21 in the captured image of the mark camera 18 and its background portion, the operator must The center position of the component 21 may be designated. The method of specifying the center position of the first component 21 is, for example, imaging the first component 21 with the mark camera 18 and displaying the image on the display device 13 so that the operator can view the captured image displayed on the display device 13. The center position of the first component 21 may be manually input, or the coordinates (X0, Y0) of the center position of the first component 21 are input by referring to design data obtained from a component supplier or the like. You may enter in.

  Thereafter, as shown in FIG. 3, the mark camera 18 is moved right above the center position of the leading part 21 to capture the leading part 21 in the center of the visual field of the mark camera 18, and the captured image is used as a reference. After being stored in the storage device 14 as an image, the mark camera 18 is continuously imaged while being moved in the arrangement direction (X direction or Y direction) of the components 21 by the mounting head moving device 15. Here, “continuous imaging” is to repeat imaging at intervals sufficiently shorter than the component pitch to be measured.

  As described above, when images are continuously captured while the mark camera 18 is moved in the arrangement direction of the components 21, even if only one component 21 can be accommodated in the field of view of the mark camera 18, captured images of the plurality of components 21 arranged. Many captured images including can be acquired. These captured images periodically change with the movement of the mark camera 18, and the number of times (the amount of movement of the mark camera 18) from when the previous component 21 is captured until the next component 21 is captured is set to 1. The period fluctuates, and the movement amount of the mark camera 18 for one period corresponds to the component pitch. From this relationship, if the periodicity of the fluctuation of the captured image is evaluated, the component pitch can be measured from the amount of movement of the mark camera 18 for one cycle.

  In this case, for example, frequency analysis may be used as a method for evaluating the periodicity of fluctuation of the captured image, or a cross-correlation method may be used. Hereinafter, a method for measuring the component pitch using the cross-correlation method will be described.

  As described above, images are continuously captured while the mark camera 18 is moved in the arrangement direction of the components 21, and the captured image and the reference image (captured image of the leading component 21) stored in the storage device 14 for each imaging. And the correlation value data is stored in the storage device 14 in time series. The correlation value increases as the similarity between the captured image and the reference image increases.

  When the amount of movement of the mark camera 18 exceeds the width of the tray 20, the movement and imaging of the mark camera 18 are finished, and the time series data of the correlation values stored in the storage device 14 are analyzed and correlated. The peak point of the value is searched, the movement amount of the mark camera 18 between the peak points of the correlation value, which is one cycle of the fluctuation of the correlation value (hereinafter referred to as “camera movement amount”), is calculated, The component pitch is stored in the storage device 14.

  Here, the captured image whose correlation value is the peak is the captured image having the highest similarity to the reference image of the component 21, and is an image actually captured by capturing the component 21 in the center of the visual field of the mark camera 18. For this reason, the position of the component 21 is determined from the position of the mark camera 18 that has captured the captured image at which the correlation value is the peak point. From this relationship, since the camera movement amount between the peak points of the correlation value corresponds to the component pitch, the component pitch can be measured from the camera movement amount between the peak points of the correlation value.

  At this time, the camera movement amount between all peak points may be calculated and the component pitch may be stored for each component 21 or the average value of the camera movement amounts between all peak points may be calculated. Then, the average value may be stored as a common component pitch.

Further, since the correlation value has a peak at the position of each component 21, the number of peak points of the correlation value is counted, and the number of peak points is stored in the storage device 14 as the number of components.
The component pitch measurement process of the present embodiment described above is executed by the control device 11 as follows according to the component pitch measurement program of FIG.

  The component pitch measurement program of FIG. 5 is executed each time the tray 20 is pulled out from the tray feeder 19 to a predetermined component supply position, and serves as a component pitch measurement means in the claims. When this program is started, first, in step 101, the mark camera 18 is moved to a position just above the center position of the head component 21, and in the next step 102, the head of the mark camera 18 is displayed at the center position of the field of view. The component 21 is imaged, and in the next step 103, the captured image of the leading component 21 is stored in the storage device 14 as a reference image.

  Thereafter, the process proceeds to step 104 where the mark camera 18 is moved by a predetermined amount in the arrangement direction (X direction or Y direction) of the components 21 by the mounting head moving device 15. Here, the “predetermined amount” is a value for setting a continuous image capturing interval (camera movement amount), and is set to a sufficiently short interval as compared with the component pitch to be measured. Thereafter, the process proceeds to step 05, where the image is captured by the mark camera 18, and in the next step 106, a correlation value between the captured image and the reference image (captured image of the first component 21) stored in the storage device 14 is calculated. Then, the correlation value is stored in the storage device 14.

  Thereafter, the process proceeds to step 107, where it is determined whether or not the camera movement amount from the position of the leading component 21 exceeds the width of the tray 20, and if it is determined that the camera movement amount does not exceed the width of the tray 20. , The above-described steps 104 to 106 are repeated. Thereby, the mark camera 18 is continuously imaged while moving in the arrangement direction of the components 21, the correlation value between the captured image and the reference image is calculated for each imaging, and the correlation value data is stored in the storage device 14. It memorizes in time series.

  Thereafter, when the amount of movement of the mark camera 18 exceeds the width of the tray 20, the process proceeds from step 107 to step 108 where the time series data of the correlation values stored in the storage device 14 is analyzed and the peak value of the correlation value Search for. Thereafter, the process proceeds to step 109, the camera movement amount between the peak points of the correlation value is calculated, and the calculated value is stored in the storage device 14 as the component pitch. At this time, the camera movement amount between all peak points may be calculated and the component pitch may be stored for each component 21 or the average value of the camera movement amounts between all peak points may be calculated. Then, the average value may be stored as a common component pitch.

Thereafter, the process proceeds to step 110, the number of peak points of the correlation value is counted, the number of peak points is stored in the storage device 14 as the number of parts, and the program is terminated.
With this program, the mark camera 18 is moved in the X direction to measure the component pitch in the X direction, the number of components in the X direction is counted, and the mark camera 18 is moved in the Y direction to measure the component pitch in the Y direction. At the same time, the number of parts in the Y direction is counted.

  If the component pitch is measured using the cross-correlation method as in the present embodiment described above, the component recognition accuracy is due to a small difference in luminance between the component 21 in the captured image and its background portion. Even under poor imaging conditions, the position of the component 21 can be determined from the position of the mark camera 18 that captured the captured image with the correlation value being the peak point. Therefore, the amount of movement of the mark camera 18 between the peak points of the correlation value Therefore, the component pitch can be measured with high accuracy.

  Further, as shown in FIG. 4, even if there is a partial luminance change due to a change in illumination, rotation deviation of the component 21, dust or dirt on the upper surface of the component 21, the correlation value has a peak at the center position of the component 21. Therefore, the component pitch can be accurately measured from the amount of movement of the mark camera 18 between the peak points of the correlation value, and the robustness against partial brightness change due to illumination change, rotation deviation, dust / dirt, etc. is improved. can do.

  Further, in this embodiment, when continuously capturing images while moving the mark camera 18 in the component arrangement direction, the captured image of the leading component 21 is used as the reference image, so when measuring the component pitch. In addition, the captured image of the first component 21 imaged under the same imaging conditions can be used as a reference image, and an accurate reference image can be easily obtained.

  However, the reference image is not limited to the picked-up image of the leading component 21 and is obtained, for example, by previously capturing the same component as the component 21 for measuring the component pitch in the center (reference position) of the camera field of view. The stored image may be stored in the storage device 14 as a reference image. At this time, it is preferable to capture the reference image under the same imaging conditions (background, illumination conditions, camera position, etc.) as the imaging conditions at the time of component pitch measurement.

In this embodiment, the cross-correlation method is used as a method for evaluating the periodicity of the fluctuation of the captured image. However, as a reference example related to the present invention, the periodicity of the fluctuation of the captured image is determined using frequency analysis. configured to evaluate is considered, even in this case, since the component recognition accuracy of the captured image is not required, components and recognition accuracy is poor imaging conditions, partial brightness change due to illumination change or the like, the deviation rotary, dust, dirt, etc. is Even in such a case, the component pitch can be measured from the amount of camera movement for one cycle of fluctuation of the captured image.

  In this embodiment, the pitch of the components 21 arranged in a matrix on the tray 20 is measured. For example, the pitch of the components stored in the component storage tape of the tape feeder is measured, or the bulk You may make it measure the pitch of the components supplied from various feeders, such as a feeder.

  In addition, the present invention is not limited to the feeder, and can be widely applied to a system for measuring a component pitch in an apparatus in which components having the same shape are arranged, and can be implemented with various modifications without departing from the gist. .

  DESCRIPTION OF SYMBOLS 11 ... Control apparatus (part pitch measurement means), 14 ... Memory | storage device, 15 ... Mounting head moving apparatus (movement mechanism), 17 ... Parts camera, 18 ... Mark camera, 19 ... Tray feeder, 20 ... Tray, 21 ... Parts

Claims (4)

In a component pitch measuring device that measures a component pitch in an array of components having the same shape,
A camera for imaging the component;
A moving mechanism for relatively moving the camera;
The camera is continuously imaged while being relatively moved in the arrangement direction of the parts by the moving mechanism, and the periodicity of fluctuations of the captured images is evaluated, and the periodicity of the fluctuations and the relative movement amount of the camera are calculated . Component pitch measuring means for measuring the component pitch based on the relationship ,
The component pitch measurement means calculates a correlation value between the captured image and the reference image of the component for each imaging of the camera, and calculates a component pitch based on a relative movement amount of the camera between peak points of the correlation value. parts pitch measurement device, characterized in that the measuring. 2. The component pitch measurement unit according to claim 1 , wherein when capturing images continuously while relatively moving the camera in a component arrangement direction, the component pitch measurement unit uses a captured image of a leading component as the reference image. Component pitch measuring device. The parts are parts arranged in a matrix on a tray set in a tray feeder,
The component pitch measurement means measures the component pitch in the X direction by relatively moving the camera in the X direction, and measures the component pitch in the Y direction by relatively moving the camera in the Y direction. Item 1. The component pitch measuring device according to Item 1 or 2 . In the component pitch measurement method for measuring the component pitch in the arrangement of components of the same shape,
Images are taken continuously while relatively moving the camera in the arrangement direction of the parts, and the periodicity of fluctuations in the captured images is evaluated, and the parts are based on the relationship between the periodicity of the fluctuations and the relative movement amount of the camera. A component pitch measurement method for measuring pitch,
A correlation value between the captured image and the reference image of the component is calculated for each imaging of the camera, and a component pitch is measured based on a relative movement amount of the camera between peak points of the correlation value. Component pitch measurement method.

Images