JPH05332723A - Method for detecting position of inspected object - Google Patents

Abstract

PURPOSE:To provide a method capable of precisely detecting the displacement of an inspected object. CONSTITUTION:Prior to the attraction of an inspected object D to a nozzle N, the initial coordinate positions of the nozzle N and a position correcting mark M provided near the nozzle N and integrally with the nozzle N are determined, and the initial coordinate positions are stored in an image processing part VC. The respective attracted state coordinate positions for the inspection target D attracted to the nozzle Z and the position correcting mark M are determined. The coordinate position obtained by correcting the attracted state coordinate position of the inspected object D in reference to the difference (u) between the initial coordinate position and attracted state coordinate position of the position correcting mark M is taken as the coordinate position of the inspected object D.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting the position of an inspection object, and more particularly to a method for detecting the position of an inspection object which enables accurate position detection.

[0002]

2. Description of the Related Art When an electronic component is adsorbed by a nozzle and then mounted at a predetermined position on a substrate, the inspection object such as the electronic component adsorbed by the nozzle is observed by a camera after the adsorption and before mounting. In order to know how much this inspection object is displaced from the ideal position,
Position detection is performed in a camera and an image processing unit that processes the image thereof.

Prior to this position detection, the position of only the nozzle is detected while the object to be inspected is not adsorbed, and the positional deviation when the nozzle is attached to the transfer head is caused by the initialization process. It is supposed to be canceled.

Therefore, in the conventional means, the position of the object to be inspected is detected on the assumption that the nozzle is always positioned accurately at the predetermined coordinates in the camera image. By the way, when the nozzle is sucking the inspection object, the nozzle is in the shadow of the inspection object, and it is difficult to detect the position of the nozzle, and this detection has not been actually performed from the above assumption. ..

[0005]

However, in the electronic component mounting process, when the nozzle that actually picks up the electronic component comes into the field of view of the camera, both the nozzle and its supporting portion side and the camera side vibrate. Ordinarily, the vibration mode differs between the nozzle side and the camera side.

The position detection result of the inspection object (positional deviation of the inspection object from the ideal position) is reflected when the electronic component is mounted on the board, and the electronic component cancels the positional displacement. And then mounted on the board. Therefore, the accuracy of this position detection must be good, especially when mounting a device such as a QFP chip in which a large number of leads extend from the electronic component body via a narrow pitch. However, this accuracy is only slightly reduced, and it causes an implementation error.

Therefore, in the conventional means, the nozzle side and the camera side vibrate differently,
There is a problem in that the accuracy of position detection is reduced and mounting errors are likely to occur.

[0008] Therefore, an object of the present invention is to provide a method for detecting the position of an inspection object which can detect the position of the inspection object with high accuracy.

[0009]

According to the present invention, when an object to be inspected adsorbed by a nozzle is observed by a camera and a position with respect to a transfer head equipped with the nozzle is detected, the object to be inspected is adsorbed by the nozzle. Prior to this, the step of obtaining the initial coordinate position of this nozzle and the position correction mark attached integrally with this nozzle in the vicinity of this nozzle, and storing this initial coordinate position in the image processing unit, and then this nozzle For each of the inspected object and the position correction mark that are adsorbed on the target object, the step of obtaining the adsorption state coordinate position and the difference between the initial coordinate position and the adsorption state coordinate position of this position correction mark are referred to And a step of setting the coordinate position obtained by correcting the coordinate position of the suction state of the object as the coordinate position of the inspection object.

[0010]

With the above-described structure, the initial coordinate positions of both the nozzle and the position correction marker integrally attached to the nozzle are first obtained before the object to be inspected is adsorbed to the nozzle. Here, the nozzle and the position correction marker vibrate integrally.

Next, in the electronic component mounting step, the suction coordinate position of the inspection target and the suction coordinate position of the position correction mark are obtained with the inspection target being sucked by the nozzle.
In this state, the nozzle cannot be observed due to the shadow of the inspection object.

Then, from the difference between the suction coordinate position of the position correction mark and the initial coordinate position, it is possible to know how much the nozzle is displaced from the state before the suction. Then, by referring to this difference, the correct amount of positional deviation of the inspection object is detected.

[0013]

Embodiments of the present invention will now be described with reference to the drawings.

FIG. 1 is a schematic view of an apparatus using a position detecting method according to an embodiment of the present invention, FIG. 2 is a bottom view showing nozzles and position correction markers, and FIGS. 3 and 4 are examples of camera images. FIG. 5, FIG. 5 is a vector diagram, FIG. 6 is a flowchart according to the above method, FIG. 7 is an explanatory view of a position correction mark according to another embodiment, and FIG. 8 is nozzle position detection when the position correction mark of FIG. 7 is adopted. FIG.

In FIG. 1, H is a transfer head and N is a transfer head.
Is a nozzle that is provided below the transfer head H and sucks the electronic component D that is the inspection object, the chain line NC is the central axis of this nozzle N, P is integral with the nozzle N, and in the vicinity of the nozzle N It is a reflector provided in the. The reflector P vibrates integrally with the nozzle N. C is a camera disposed in the recognition station in the electronic component mounting process, L is a light source, and light a emitted from this light source L is reflected by the reflection plate P and enters the camera C. VC is an image processing unit that performs image processing on the image of the camera C according to the flowchart of FIG. 6, and the image processing unit VC is composed of an interface, a computer, and the like. In addition, the reflector P
Outside the nozzle N of the mark M as a position correction mark.
Is attached (see also the bottom view of FIG. 2). This mark M
The shape and the like are arbitrary as long as they can be clearly observed by the camera C. Further, a small light source may be arranged at the above position so that it can be observed as a spot-like bright image in the image of the camera C, and it may be used as a position correction marker.

The apparatus according to this embodiment has the above-mentioned structure. Next, the position detecting method of this embodiment will be described with reference to the flow chart of FIG. 6 with reference to FIGS.

First, prior to adsorbing the inspection object (electronic component D) to the nozzle N, the nozzle N and the position correction mark M are used.
The initial coordinate position of is calculated, and this position is stored in the image processing unit VC (steps 1 and 2). FIG. 3 shows an image of the camera C in this initial state, where V is the visual field, x and y are the horizontal and vertical axes of the visual field V, NS is the image of the nozzle N, and MS is the image of the position correction marker M. Is.

Specifically, similarly to the initialization processing of the electronic component mounting apparatus, first, the offset values Δx and Δy between the origin of the visual field V and the central axis NC of the image NS of the nozzle N are stored, and this offset will be described below. The value Δx. The process is performed with reference to Δy. However, for the sake of simplicity, it is assumed that the offset values Δx and Δy are both 0.
(Step 1). Next, set the marking area MA, scan this marking area MA,
The image MS of the marker M is pattern-matched. Alternatively, the coordinates of the center of the image MS and the rotation angle may be obtained by contour tracing. Here, the broken line indicates the marking area MA before matching, and the solid line indicates the marking area MA at the time of matching. Then, when this matching is obtained, the coordinate position (the initial coordinate position of this marker M) from the central axis NC to the center of the image MS, the shape of this marker, and the like are stored in the image processing unit VC (step 2). ).

Next, the electronic component mounting apparatus is operated to obtain the suction state coordinate position for each of the electronic component D sucked by the nozzle N and the mark M (step 3,).
4).

FIG. 4 shows an image of the camera C in this suction state. In FIG. 4, DS is an image of the electronic component D, D
C is the center of this image DS. Specifically, the coordinate position of the center DS of the image D (coordinate position of the suction state of the inspection object D) and its rotation angle with respect to the x-axis are detected (step 3). Then, pattern matching is performed again on the image MS of the position correction marker M to obtain the adsorption state coordinate position of the marker M. Then, the difference between the initial coordinate position of the sign M stored in the image processing unit VC and the suction coordinate position of the sign M just obtained (that is, the vector u in FIG. 4).
Is calculated (step 5). In this example, the vibration of the nozzle N causes a considerable displacement. Since the marker M and the nozzle N on the reflection plate P vibrate integrally,
The nozzle N is as shown by the broken line in FIG.
The image NS does not appear), and it can be seen that the position is displaced by the vector u.

Next, referring to this vector u, the coordinate position obtained by correcting the already obtained suction state coordinate position of the inspection object (electronic component D) is set as the coordinate position to be obtained (step 6). In FIG. 4, the center DC of the image DS of the inspection object D
Is displaced by the vector v from the original origin. Here, the vector V indicating the desired coordinate position is
(Vector V) = (vector v) − (vector u). Here, in the conventional means, not the correct coordinate position (vector V) but the raw coordinate position (vector v) indicates the displacement. However, as is apparent from FIG. 5, according to the conventional means, when the nozzle N vibrates and the vector u becomes large, the accuracy deteriorates to a nonnegligible level.

Now, FIG. 7 shows a position correction mark according to another embodiment. In the example shown in FIG. 7A, the mark M1 is attached to the cross-shaped line on the lower surface (on the side of the nozzle N) of the reflection plate P, and in the example of FIG. A slit S is formed, a light source L that irradiates light toward the slit S is provided above the reflecting plate P, and cross-shaped slit light (marker) M2 that passes through the slit S is captured by the camera C. It was done like this. The light source L
Is to emit red light and the reflecting plate P is white,
It is possible to obtain a sufficient contrast between the image of the reflection plate P and the image of the slit M2. Such a cross-shaped marker M1,
When M2 is taken into camera C, it is observed as shown in FIG. In this case, in order to obtain the center NC of the image NS of the nozzle N, the following may be done. That is, as shown by the alternate long and short dash line in FIG. 8, the peripheral portion of the visual field V is frame-shaped and line-scanned to find points P1, P2, P3, and P4 in contact with each side of the cross. Then, it is advisable to set the intersection NC of the straight lines passing through P1, P2, P3 and P4 as the center NC.

[0023]

According to the present invention, when an object to be inspected adsorbed by a nozzle is observed with a camera and the position of the nozzle with respect to a transfer head having a lower portion is detected, the object to be inspected is adsorbed by the nozzle. First, the step of obtaining the initial coordinate position of this nozzle and the vicinity of this nozzle and of the position correction marker attached integrally with this nozzle, and storing this initial coordinate position in the image processing unit, and then adsorbing to this nozzle For each of the inspected inspection object and this position correction mark, referring to the difference between the initial state coordinate position and the adsorption state coordinate position of this position correction mark And a step of setting the coordinate position obtained by correcting the suction state coordinate position as the coordinate position of the inspection object. Therefore, even if there is a positional deviation with respect to the camera with the nozzle adsorbing the object to be inspected, this deviation is corrected and accurate position detection can be performed.

[Brief description of drawings]

FIG. 1 is a perspective view of an apparatus using a position detecting method according to an embodiment of the present invention.

FIG. 2 is a bottom view showing a nozzle and the like according to an embodiment of the present invention.

FIG. 3 is an exemplary view of a camera image according to an embodiment of the present invention.

FIG. 4 is an exemplary view of a camera image according to an embodiment of the present invention.

FIG. 5 is an explanatory diagram of correction of a suction state coordinate position according to an embodiment of the present invention.

FIG. 6 is a flowchart according to an embodiment of the present invention.

FIG. 7 is an explanatory view of a position correction sign according to another embodiment of the present invention.

FIG. 8 is an explanatory diagram of nozzle position detection according to another embodiment of the present invention.

[Explanation of symbols]

 N nozzle D inspection object H transfer head M position correction marker M1 position correction marker M2 position correction marker VC image processing unit

Claims (1)

[Claims] 1. An object to be inspected adsorbed by a nozzle is observed by a camera to detect a position with respect to a transfer head equipped with this nozzle. A step of obtaining an initial coordinate position of a position correction mark attached in the vicinity of the nozzle and integrally with the nozzle, and storing the initial coordinate position in the image processing unit; For each of the position correction signs, the step of obtaining the suction state coordinate position and the difference between the initial coordinate position and the suction state coordinate position of the position correction mark are referred to correct the suction state coordinate position of the inspection object in the previous period. The method for detecting the position of an inspection object, the method comprising: using the coordinate position as a coordinate position of the inspection object.