KR0151178B1 - Camera calibration method for vision surface mounting device - Google Patents

Abstract

The present invention relates to a camera calibration method of a vision surface-mounted component mounter that is suitable for performing the calibration accurately and easily. Conventionally, a calibration point when a component adsorption nozzle is adsorbed differently from a direction in which a calibration frame is assumed When the position in the mounting machine coordinate system is different from the actual position and the direction of the conveyor does not coincide with the assumed direction, a position error occurs in the mounting machine coordinate system. However, in the present invention, the component adsorption nozzle 3 is used for component inspection. The parts inspection camera 2 photographs the parts adsorption nozzle 3 and processes the coordinate data necessary for calibration while changing the position to each part of the camera 2 so as to accurately know the position of the parts adsorption nozzle 3. Since the position of the coordinate system can be set accurately, the reference mark inspection camera (1) is set on the PCB (5). Since the position of the calibration point can be accurately known by photographing at a predetermined interval around the predetermined reference mark 4, it is possible to accurately set the position of the coordinate system, thereby improving the defect.

Description

Camera calibration method of vision surface mount component mounter

1 is a perspective view showing a conventional vision surface mount component mounter.

2 is a perspective view showing a conventional calibration frame.

3 is a perspective view illustrating a calibration point input of a camera for inspecting parts of FIG. 1;

4 is a perspective view illustrating a calibration point input of a camera for inspecting a reference mark of FIG. 1.

5 is a block diagram for explaining the control of FIG.

6 is a perspective view illustrating a calibration point input of a camera for inspecting parts of the present invention;

7 is a perspective view showing a calibration point input of the camera for inspection of the reference mark of the present invention.

8 is a schematic view showing an image of a calibration point input according to FIG.

9 is a schematic diagram showing an image of a calibration point input according to FIG. 7.

10 is a flowchart illustrating the calibration of the present invention.

* Explanation of symbols for main parts of the drawings

1: Camera for inspection of reference mark 2: Camera for inspection of parts

3: nozzle for adsorption of components 4: reference mark

5: PCB 7: Head

8: conveyor

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera calibration method used in a Vision Surface Mounting Device Mounter, and more particularly, to a vision surface mount component mounter that is suitable for accurate and easy calibration. It relates to a camera calibration method.

In general, in the case of a large number of leads such as a quad flat package (QFP) among surface mount components, a vision system is used to improve the read condition inspection and mounting precision of the corresponding component.

In order to perform such a thing, a camera for detecting a lead state and a position of a part, a camera for detecting a fiducial mark position for correcting an actual position, etc. are required, and a calibration operation must be performed in advance for each camera to perform its role. .

Calibration is to find a relationship between the mounting machine coordinate system and the image coordinate system. The relationship is used to convert the part or reference mark from the image coordinate system position to the mounting machine coordinate system position.

In addition, several points of data (Data) for indicating the position in the mounting coordinate system and the image coordinate system are required to obtain the relational expression.

The prior art has a head (7) and a reference mark inspection camera (1) capable of moving the position along each axis of the X, Y, Z, and R axes as shown in FIG. A component adsorption nozzle (3), a reference mark (4), and a printed circuit board (PCB) mounted to the camera 2, the lower side of the head 7, for adsorbing components such as the QFP 6, etc. 5), a conveyor (8), and a tray (9).

In the prior art configured as described above, the head 7 uses the component adsorption nozzle 3 to adsorb components such as the QFP 6 on the tray 9 and then move the QFP to the position of the component inspection camera 2. By moving (6), the state and position of the QFP 6 are examined to determine whether to use the QFP 6 or not.

Next, when the decision to use the QFP (8), after modifying the mounting position of the parts on the PCB (5) according to the position inspection result of the QFP (6), the reference mark (4) with the reference mark inspection camera (1) The position of the component is corrected according to the result of the inspection, and the QFP 6 is mounted at the component mounting position on the modified PCB 5.

For this implementation, the calibration of the component inspection camera 2 and the reference mark inspection camera 1 should be preceded, that is, a calibration tool (10) in which a calibration point 11 as shown in FIG. 2 is drawn. Is absorbed by the component adsorption nozzle 3 as shown in FIG. 3, and the image of the calibration point 11 is taken by the component inspection camera 2 to process the data in the mounting coordinate system and the image coordinate system to perform a calibration algorithm. Algorithm).

Subsequently, a correlation equation between the mounting machine coordinate system and the image coordinate system is obtained, and the calibration tool 10 is fixed to the conveyor 8 as shown in FIG. 4, and the image of the calibration point 11 is fixed with the reference mark inspection camera 1. After performing the data processing and calibration algorithm as described above, the correlation between the real machine coordinate system and the image coordinate system is obtained.

In addition, referring to FIG. 5, when the control of FIG. 1 is performed, the main controller 12 generates each control signal so that the driver 13 is connected to the main controller 12 so that the X-axis, the Y-axis, the Z-axis, and the R-axis are connected. It controls the motor and input / output which determine the position along each axis such as the shaft, and the vision unit 14 is connected to the main control unit 12 so that the camera for component inspection (Charge Coupled Device) ( 2) and a reference couple inspection camera (Charge Coupled Device) 1 is controlled.

However, this conventional technique has the following drawbacks.

First, in the case of FIG. 3, when the component adsorption nozzle 3 is adsorbed differently from the direction in which the tool 10 for calibration is assumed, the calibration point 11 is different from the actual position in the mounting coordinate system.

Second, in the case of FIG. 4, when the direction of the conveyor 8 does not match the assumed direction, a position error occurs in the mounting machine coordinate system.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned drawbacks, and an object thereof is to provide a camera calibration method of a vision surface mount component mounter that can accurately and easily realize calibration without using a calibration tool separately. .

Hereinafter, an embodiment of the present invention for achieving such an object will be described in detail with reference to the accompanying drawings.

First, as shown in FIG. 6, the parts inspection nozzle 2 mounted on the head 7 is moved from the center of the parts inspection camera 2 to the corners of the parts inspection camera 2 so that the parts inspection camera 2 is shown in FIG. After taking the image as shown in (e), the position data of the image coordinate system is obtained, and the position of the component adsorption nozzle 3 is converted into the value of the mounting machine coordinate system (FIG. 10 S1-S11).

Next, since a calibration algorithm is performed, a transformation matrix between an image coordinate system and a mounting coordinate system is obtained (S21 and S22).

Subsequently, the reference mark inspection camera 1 is moved to each corner from the center of the PCB 5 where the reference mark 4 is drawn on the conveyor 8 as shown in FIG. 7. After taking the image shown in Figs. 9A (e) to (e), the position data of the image coordinate system is obtained, and the center position of the reference mark inspection camera 1 is converted into the value of the mounter coordinate system (S12-S20).

In addition, since a calibration algorithm is performed, a transformation matrix between an image coordinate system and a mounting coordinate system is obtained (S21 and S22).

As described above, the present invention has the following effects.

First, in the case of FIG. 6, the component adsorption nozzle 3 changes its position to each part of the component inspection camera 2, and the inspection camera 2 captures the component adsorption nozzle 3 and coordinates necessary for calibration. By processing the data, the position of the component adsorption nozzle 3 can be known accurately, so that the position of the coordinate system can be set accurately.

Secondly, in the case of FIG. 7, the reference mark inspection camera 1 accurately captures the position of the calibration point by photographing the surroundings of any reference mark 4 set on the PCB 5 at regular intervals, thereby accurately determining the position of the coordinate system. Can be set.

Claims (1)

By moving the suction nozzle 3 to each part of the component inspection camera 2, the component inspection camera 2 takes an image to obtain position data of the image coordinate system, and then mounts the position of the component suction nozzle 3. Step (S1-S11) of data conversion of the coordinate system values, and a calibration algorithm using the data to obtain a conversion matrix between the image coordinate system and the mounter coordinate system (S21, S22), and a camera for checking a reference mark. By moving (1) to each part of the PCB (5), the reference mark inspection camera (1) takes an image and obtains the position data of the image coordinate system, and then moves the center position of the reference mark inspection camera (1) to the mounting coordinate system. Since the calibration algorithm is performed using the data (S12-S20) and the data of the steps (S12-S20), the method calculates a transformation matrix between the image coordinate system and the mounting coordinate system (S21, The camera calibration method of the vision surface mount component mounting machine containing S22).