JPH05240625A - Picture recognizing device and electronic parts mounting machine using the same - Google Patents

Abstract

PURPOSE:To recognize acceptable electronic parts by again performing a recognizing process after moving the parts to a position where the contour of the parts can be detected by providing a contour detecting means which detects the contour of the section of the parts left on a screen, center presuming means which presumes the center position of the parts based on the output of the detecting means, and moving means which moves the parts into a visual field. CONSTITUTION:Of the four sides of the visual field 10 of a camera, the side A where the image of a semiconductor parts QFP8 appears is detected and the contour of the section of the image contained in the visual field 10 is detected. Then the length direction of the side contained in the visual field 10 of the sides of the parts QFP8 is detected. Thereafter, the corner C and centers UC and RC of the sides LR and LV of the parts QFP8 are found from these information. Since the outside dimensions of the parts QFP8 are given in advance at the time of preparing data, the intersection PP of vectors is found by drawing straight lines in the vector directions L1 and L2 of the sides from the intermediate points UC and RC of the external dimensions and the intersection PP is decided as the center of the parts QFP8. When the difference between the thus found center PP and center of the visual field 10 is found and a mounting head is moved by the difference, the contour of the parts QFP8 can be accurately detected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image recognition device for performing recognition correction and an electronic component mounting machine using the same.

[0002]

2. Description of the Related Art An electronic component mounter equipped with a conventional image recognition apparatus will be described below. FIG. 3 is a schematic configuration diagram of the electronic component mounter. As shown in FIG. 3, the mounting head 1
The recognition camera 3 captures the image of the electronic component 2 adsorbed in (1) as an image signal. This image signal is sent to the image processing device 5 through the amplifier 4 and subjected to image processing. A correction amount is added to the image signal by the correction processing device 6, and the motor 8 is driven via the motor driving device 7.

FIG. 4 is a flowchart showing the operation of a conventional electronic component mounter. A step number is assigned to each step (hereinafter abbreviated as ST). First, the image is captured in ST1, and the electronic component is recognized in ST2. In ST3, it is determined whether the contour of the electronic component can be detected. If the contour cannot be detected, the electronic component is discarded in ST4. If it is determined in ST3 that the contour of the electronic component has been detected, the lead position is detected in ST5. The number of leads is verified in ST6, and if the number of leads is different, the electronic component is discarded in ST7.
If it is determined in ST6 that the number of leads of the electronic component is the predetermined number, the lead length is verified in ST8. If the lead length is different from the predetermined length, the electronic component is discarded in ST7. If the lead length is determined to be the predetermined length in ST8, the lead pitch is verified in ST9. If the lead pitch is different from the predetermined pitch, the electronic component is discarded in ST7. If it is determined in ST7 that the lead pitch is the predetermined lead pitch, in ST10, inclination correction and X, Y correction for a predetermined axis are performed. The result is sent to the X, Y, θ drive means of the electronic component mounter, and in ST11, the electronic component is moved to the correct position and mounted.

FIGS. 5A to 5C are views for explaining contour detection. In these figures, 8 is a semiconductor component (hereinafter abbreviated as QFP) in which leads are arranged on four sides, 9 is a lead of the QFP 8, and 10 is a camera visual field. Figure 5
(A) shows a state in which the QFP 8 is out of the camera field of view 10. In this case, it is determined in ST3 of FIG. 4 that the contour of the QFP 8 is not detected, so this QFP 8 is discarded in ST4. In FIG. 5B, it is determined that the QFP 8 is within the camera visual field 10 and the contour is detected in ST3 of FIG. 4, so the process proceeds to the lead position detection of ST5 of FIG. Here, as shown in FIG. 5 (c), if part of the lead 9 is deformed, the number of leads becomes insufficient at ST6 of FIG. 4, the lead length becomes defective at ST8, or the lead pitch at ST9. It becomes defective and this QFP 8 is discarded.

[0005]

However, the above-mentioned conventional configuration has a problem that when the contour of the electronic component cannot be detected in ST3 of FIG. 4, the electronic component is discarded although it is a good product. It was

The present invention solves the above-mentioned conventional problems. Based on the image recognition apparatus and the result thereof, a good electronic component can be recognized by moving the electronic component and performing recognition processing again even when the contour cannot be detected. It is an object of the present invention to provide an electronic component mounter that can mount good electronic components without discarding them.

[0007]

In order to achieve this object, the image recognition apparatus of the present invention can detect the contour of an electronic component for which the contour could not be detected by changing the coordinates of the mounting head sucking the electronic component. The electronic component is moved to the position and the recognition processing is performed again.

[0008]

With this configuration, the center position of the electronic component can be determined even if a part of the contour deviates from the field of view of the camera, and the electronic component is moved so that the center position becomes the center position on the screen. , Can recognize electronic components correctly.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a flow chart showing the operation of the image recognition apparatus in one embodiment of the present invention, and only the parts different from the conventional example shown in FIG. 4 are shown. As shown in FIG. 1, first, an image is captured in ST1, and the contour of the electronic component is detected by the contour detecting means in ST2. Next, in ST3, it is determined whether or not the entire contour can be detected. If it is determined that the entire contour can be detected (in the case of FIG. 5B or 5C), the conventional example shown in FIG. Proceed to ST5, then ST6 shown in Figure 4
The following flow will be followed. On the other hand, if it is determined in ST3 that the entire contour could not be detected (in the case of Fig. 5 (a)), ST4 estimates the center position of the electronic component, and ST5 moves the visual field so that the electronic component fits within the camera visual field. The electronic parts are moved by the means, and the process returns to ST1 again and starts from image capturing.

Next, a method of estimating the center of an electronic component in the image processing apparatus of the present invention will be described. FIG. 2A is a diagram for explaining a method of estimating the center of an electronic component, and FIG. 2B is a flowchart of the method of estimating the center of the electronic component.
An example of FP is shown. First, as shown in FIG. 2A, the side A where the image of the QFP 8 extends on the camera visual field 10 is detected in ST1 among the four sides of the camera visual field 10. Then ST2
Then, the contour is detected from the side A within the camera visual field 10. Next, in ST3, the length and direction of the side within the camera visual field 10 of the side of QFP8 (vector symbol is omitted, L
_1, L ₂ and display) for detecting a. Then from these information, corner QFP8 contained in the camera field of view 10 in ST4 (C) and QFP8 sides (L _H, L _V) center U _c of,
Find R _c . In QFP8, each lead 9
The intersection of the lines that connect the tips of is defined as the corner (C). Since the external dimensions of QFP8 are given in advance at the time of data creation, in ST5, draw a straight line from the midpoints (U _c , R _c ) of the external dimensions in the vector direction (L ₁ , L ₂ ) of each side, and in ST6. The intersection point P _p of the vector is obtained, and P _p is determined to be the center of QFP8 in ST7.

P _p and the camera field of view 1 thus obtained
By obtaining the difference between the centers P _c of 0 and moving the mounting head by the difference (X, Y, θ), the entire QFP 8 is set within the camera visual field 10, and the contour can be accurately detected.

By mounting such an image recognition device on an electronic component mounter, accurate alignment can be performed without determining that a non-defective product is defective.

[0013]

As described above, according to the present invention, the contour detecting means for detecting the contour of the portion of the electronic component left on the screen, and the center estimating means for estimating the center position by the output from the contour detecting means. And a configuration including a visual field moving means for moving the electronic component into the visual field based on the result,
There may be some misalignment when picking up electronic parts,
Even if the accuracy of the initial teaching (the operation of inputting information to the recognition device) is not very accurate, it is possible to realize an image recognition device that can detect the outline of the entire electronic component.
Further, by using such an image recognition device in an electronic component mounting machine, it is not necessary to judge a non-defective product as a defective product and discard it as in the conventional case.

[Brief description of drawings]

FIG. 1 is a flowchart showing the operation of an image recognition apparatus according to an embodiment of the present invention.

FIG. 2A is a diagram for explaining a center estimation method of an electronic component, and FIG. 2B is a flowchart of the center estimation method.

FIG. 3 is a schematic configuration diagram of an electronic component mounter including a conventional image recognition device.

FIG. 4 is a flowchart showing the operation of a conventional electronic component mounter.

FIG. 5 is a diagram for explaining contour detection.

[Explanation of symbols]

 ST3 Contour detection ST4 Part center guess ST5 Field of view

Claims (2)

[Claims] 1. When a part of an electronic component to be recognized protrudes from the screen, a contour detecting unit for detecting a contour of a portion of the electronic component left on the screen, and an output from the contour detecting unit are output. An image recognition apparatus having a center estimating means for processing and estimating the center position of an electronic component, and a visual field moving means for moving the electronic component into the visual field in accordance with an instruction from the central estimating means. 2. An electronic component mounter equipped with the image recognition device according to claim 1.