KR20200044750A - Method for determining work inspection result - Google Patents

Abstract

An object of the present invention is to optimally determine whether a workpiece is good or bad. As a means to solve the problem, a method for determining an inspection result of the workpiece comprises the following steps of: inspecting a plurality of inspection items on a plurality of workpieces; obtaining evaluation values for each workpiece by using a combining function combining functions determined for each inspection item based on the inspection result of all inspection items; and arranging evaluation values of each workpiece in a desired space. Based on a distribution status of the evaluation values arranged in the desired space, a boundary dividing the desired space into regions representing a plurality of classes is set in the desired space.

Description

How to judge the inspection result of a work {METHOD FOR DETERMINING WORK INSPECTION RESULT}

The present invention is used in an appearance inspection apparatus for a work that carries out a workpiece such as an electronic component and images each surface by a work imaging means to perform an appearance inspection, and determines the inspection result of the work that makes the determination based on the inspection result. It's about how.

Conventionally, a workpiece such as an electronic component having a hexahedron shape is mounted on an upper surface of a rotating disk-shaped transparent glass table, and the workpiece is adsorbed to the upper surface of the glass table and conveyed, and each surface of the workpiece is imaged A visual inspection device for a workpiece that is imaged by means to perform visual inspection is known (see Patent Document 1).

In the apparatus for inspecting the appearance of such a workpiece, the workpiece is inspected for a plurality of inspection items. Then, the threshold value of OK / NG is set for each inspection item, and the work determined as OK for all inspection items is judged as a good product.

However, it is not easy to determine this threshold, since it is determined by case by case how to determine whether a good or bad product can be optimally set by setting the threshold of each inspection item.

Japanese Patent Application No. 2011-133458

The present invention has been made in consideration of these points, and an object of the present invention is to provide a method for judging an inspection result of a work that can optimally judge a good or bad product.

The present invention relates to a plurality of workpieces in an inspection result judging method for inspecting a plurality of workpieces on a plurality of inspection items and determining a work grade based on inspection results of all inspection items. A process of inspecting a plurality of inspection items and a process of obtaining evaluation values for each work using a combination function combining functions determined for each inspection item based on inspection results of all inspection items, Based on the process of arranging the evaluation values of each work in the desired space and the distribution state of the evaluation values arranged in the desired space, a boundary is defined in the desired space to divide the desired space into regions representing a plurality of classes. It is a method for judging the inspection result of a work, characterized by comprising a step.

In the present invention, the inspection is an inspection result judging method characterized in that the inspection is an image inspection.

In the present invention, in the step of obtaining the evaluation value, the function determined in the inspection result item includes a coefficient determined by the importance distribution of the inspection item, and the inspection result determination method of the work is characterized by the above-mentioned.

In the present invention, in the step of setting the boundary, a region indicating a plurality of grades is a method for judging the inspection result of a work, characterized in that it includes a good product area and a bad product area.

In the present invention, in the step of obtaining the evaluation value, the evaluation value obtained for each work includes an OK score and an NG score, and is a method for judging the inspection result of the work.

The present invention is a method for judging the inspection result of a work, characterized in that, after the step of setting the boundary, each work is judged as a good product or a defective product.

As described above, according to the present invention, it is possible to appropriately and easily determine the good / bad product with respect to the work to be inspected.

1 is a plan view showing a visual inspection device for a work.
2 (a) and 2 (b) are views showing a state in which the workpiece placed on the upper surface of the conveying table is conveyed.
3 is a flowchart showing a method for judging the inspection result of a work according to the present invention.
4 is a diagram showing a two-dimensional space used in a method for determining a result of inspection of a work according to the present invention.
5 is a diagram showing a test result determination method as a comparative example.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

1 to 5 are diagrams showing an embodiment of a method for judging an inspection result of a work according to the present invention.

First, the outline of the external appearance inspection device of the work as the image inspection device in which the method for determining the inspection result of the work is used will be described with reference to FIG. 1.

As shown in FIG. 1, the visual inspection device 30 for a work includes a transparent glass transport table 2 and a transport table 2 for transporting a work W such as an electronic component including a hexagonal electrode. 2) the side camera unit 8, the inner camera unit 9, the upper camera unit 10, the lower camera unit 11, and the front camera unit 12 that image the six surfaces of the work W conveyed by the And a rear camera unit 13.

Among these, the conveying means is constituted by the conveying table 2, and the side camera part 8, the inner camera part 9, the top camera part 10, the bottom camera part 11, and the front camera The work imaging means is configured by the unit 12 and the rear camera unit 13.

In the visual inspection apparatus 30 for the work shown in Fig. 1, first, the work W is arranged in a line by the vibration action of the linear linear feeder 1 descending with a slight inclination, and the arrow N is Direction. Next, the workpiece W is a transparent glass conveying table provided horizontally below the linear feeder 1 through the vibration-free vibration part 4 at the downstream end of the linear feeder 1. It is displaced on the upper surface of (2). Then, the ionizer 6 provided slightly below the position of the non-vibration portion 4 and immediately below the transport table 2 ejects positive ions toward the underside of the transport table 2. For this reason, the bottom surface of the conveying table 2 is positively charged, and the work W is adsorbed on the upper surface of the conveying table 2 by the action of electrostatic induction or dielectric polarization generated thereby.

The adsorbed work W is conveyed by the rotation of the conveying table 2 (direction of the arrow X around the central axis 3), and thereafter the conveying table 2 by the action of the alignment guide 7 It is aligned on the work conveying arc 5, which is assumed to be near the outer edge of the upper surface of the. And, while the work W is conveyed in its aligned state, the side camera unit 8, the inner camera unit 9, the top camera unit 10, the bottom camera unit 11, the front camera unit 12, and the rear camera In the part 13, each surface is imaged by the corresponding imaging means, and visual inspection is performed.

The work W which has completed the visual inspection is discharged from the work conveying arc 5 onto the storage box (not shown) by the discharge unit 14 in response to the result of the visual inspection.

Here, the state in which the workpiece W placed on the upper surface of the conveying table 2 is conveyed is shown in Fig. 2 (a) as a perspective view. The workpiece W has a hexahedron shape, and the length of each side is WL in the longitudinal direction along the rotational direction of the conveying table 2 indicated by an arrow X, as shown in Fig. 2A. One side in the width direction along the radial direction from the central axis 3 (FIG. 1) of the transfer table 2 toward the outer edge portion is Ww, and one side in the height direction upward from the top surface of the transfer table 2 is WH. Here, the lengths WL, Ww, and WH of each side are all approximately 1 mm to 10 mm.

2 (b) is a plan view of the work W of FIG. 2A seen from the top surface (direction of the arrow C) of the transfer table 2. In Fig. 2 (b), the work W is placed on the upper surface of the transport table 2 such that the side Ws facing the outside of the transport table 2 becomes the tangent line of the work transport arc 5. . The inner surface Wi facing the side Ws faces the central axis 3 (FIG. 1) of the conveying table 2. Then, the front Wf is directed toward the rotational direction of the conveyance table 2 indicated by the arrow X, and the rear Wr is directed toward the opposite direction to the rotational direction.

As shown in Fig. 2 (a), the length WL faces the longitudinal direction of the work W. As described above, in the visual inspection apparatus 30 for the work, the lengthwise direction of the hexahedron-shaped work W is conveyed by the action of the linear feeder 1 and the alignment guide 7 in FIG. 1. It is mounted on the upper surface of the conveying table 2 so that it follows the rotation direction of (2) and the side surface Ws is the tangent of the work conveying circular arc 5 toward the outside of the conveying table 2.

Here, the correspondence relationship between each surface of the workpiece | work W and each camera part which images these is described below. First, in the side camera part 8 of FIG. 1, the side surface Ws of FIG. 2B is imaged from the direction of the arrow A of FIG. 2A. Next, in the inner surface camera unit 9 of Fig. 1, the inner surface Wi of Fig. 2B is imaged from the direction of the arrow B in Fig. 2A. Next, in the top camera unit 10 of FIG. 1, the top surface Wt of FIG. 2A is imaged from the direction of the arrow C in FIG. 2A. Next, in the bottom camera part 11 of FIG. 1, the bottom surface Wb of FIG. 2 (a) is imaged from the direction of the arrow D of FIG. 2 (a). Next, in the front camera unit 12 of Fig. 1, the front surface Wf of Fig. 2B is imaged from the direction of the arrow E in Fig. 2A. Finally, in the rear camera portion 13 of FIG. 1, the rear surface Wr of FIG. 2B is imaged from the direction of the arrow F of FIG. 2A.

Next, the imaging results captured by the side camera unit 8, the inner camera unit 9, the top camera unit 10, the bottom camera unit 11, the front camera unit 12, and the rear camera unit 13 are It is sent to the control unit 20.

In the control unit 20, a method for judging the inspection result of the work is performed according to the flowchart shown in FIG. 3 with the following tips.

Specifically, first, the inspection method is started, the side camera unit 8, the inner camera unit 9, the top camera unit 10, the bottom camera unit 11, the front camera unit 12 and the rear camera unit ( 13), the six surfaces of each work W are imaged, and the control unit 20 performs inspection on a plurality of inspection items based on the imaging results of the six surfaces of the captured work W.

Examples of these inspection items include (1) the size of defects in the workpiece, (2) the size of the electrodes, and (3) the state of contamination.

Next, in the control unit 20, evaluation is performed for each work W using a combining function combining functions determined for each inspection item based on the inspection results of the entire inspection items (1) to (3). Get the value.

Next, a coupling function in which a function determined for each inspection item is combined will be described below.

Here, the following equation (1) can be used as the coupling function F1.

F1 = Afa (OK, NG) + Bfb (OK, NG) + Cfc (OK, NG) + ... (1)

The combining function F1 is a function Afa () generated from coefficients A, B, C ... indicating the importance distribution of each test item when determining the score when determining the evaluation value of each work, and the threshold value of each test item It is generated as a function consisting of a combination of OK, NG), Bfb (OK, NG), and Cfc (OK, NG) (first order in (1) equation). In the formula (1), "OK" represents the OK score, and "NG" represents the NG score.

Moreover, the coupling function F1 is applied when description by an equation is easy. On the other hand, when it is difficult to describe the expression of the coupling function F1, a function F2 corresponding to F1 may be generated in AI (artificial intelligence) having a learning function.

In the equation (1), as described above, F1 denotes a combining function, and functions Afa (OK, NG) are functions determined for each inspection item generated from inspection items, for example, the threshold of defects in the work, and the function Afa (OK, NG) has parameters of OK (good) or NG (bad) based on the threshold. In addition, the functions Afa (OK, NG) have a coefficient A indicating the importance distribution of the test item, and when the weight of the test item is low, the coefficient A takes a small value, and when the weight is high, the coefficient A has a large value. To take.

In addition, in equation (1), the function Bfb (OK, NG) is a function determined for each test item generated from a test item, for example, a threshold of the size of the electrode of the work, and the function Bfb (OK, NG) is a threshold It has the parameters of OK or NG as a reference. In addition, the function Bfb (OK, NG) has a coefficient B indicating the importance distribution of the inspection item, and when the weight of the inspection item is low, the coefficient B takes a small value, and when the weight is high, the coefficient B has a large value. To take.

In equation (1), the function Cfc (OK, NG) is a function determined for each test item generated from a test item, for example, a threshold value of the contamination state of the work, and the function Cfc (OK, NG) is based on the threshold value. It has one OK or NG parameter. In addition, the functions Cfc (OK, NG) have a coefficient C indicating the importance distribution of the test item, and when the weight of the test item is low, the coefficient C takes a small value, and when the weight is high, the coefficient C has a large value. To take.

In this way, the evaluation value for each work is determined using the combining function F1 which combines the functions determined for each inspection item shown in equation (1).

Next, the control unit 20 arranges the evaluation values for each work in the desired two-dimensional space shown in FIG. 4.

In this case, the evaluation value for each work includes an OK score (good score) and an NG score (bad score) as shown in FIG. 4. Here, in Fig. 4, the vertical axis represents the OK score, and the horizontal axis represents the NG score.

As shown in Fig. 4, if the value of the OK score is large among the evaluation values for each work, evaluation of the work becomes good, and when the value of the NG score is large, the evaluation of the work becomes poor.

Next, on the basis of the distribution state arranged in the secondary space, boundaries A and B are set in the secondary space to partition the secondary space into regions α, β, γ, and δ representing a plurality of classes.

In Fig. 4, the secondary space is divided into four regions α, β, γ, and δ by linear boundaries A and B, of which the region α is a good product region, and the work entering the region α is a good product It is judged as.

The area δ is a defective product area, and a work entering the area δ is determined as a defective product.

The regions β and γ are gray regions, and the work entering the regions β and γ is subjected to visual inspection (gray product).

In this way, an evaluation value including the OK score and the NG score of each work is placed in the secondary space, and the secondary space is divided into four regions α, β, γ, and δ based on the distribution state of the evaluation value. Set boundaries A and B to divide. Next, the workpieces entering these four regions α, β, γ, and δ can be judged as good products, defective products, or gray products.

Next, a comparative example of the present invention will be described with reference to FIG. 5. In the comparative example shown in Fig. 5, for each work, inspection item (1): defect size is obtained, inspection item (2): electrode size is obtained, inspection item (3): contamination state is obtained .

In the comparative example shown in Fig. 5, the thresholds 1 to 3 are set for each of the inspection items (1) to (3) for each work, and the thresholds 1 to 3 are standard for each inspection item (1) to (3). Thus, good (OK) and bad (NG) are judged. And when all the inspection items 1-3 of each work are judged as good, the work is judged as a good product, and when any one of the inspection items 1-3 is defective, the work is regarded as a defective product. Is judged.

However, in the case of this comparative example, it is unclear how to set the thresholds of each inspection item to properly judge whether good or bad items are appropriate, and the thresholds of each inspection item are determined independently of the thresholds of other inspection items. It is difficult to judge whether a good product or a defective product is considered in consideration of the overall item balance.

On the other hand, according to the present invention, the evaluation value is obtained using a combining function in which a function determined for each test item is combined, the evaluation value is placed in the secondary space, and the secondary space is based on the distribution state of the evaluation value. Is divided into four regions α, β, γ, and δ, and the grades (good, defective, and gray) of the work entering the four regions α, β, γ, and δ are determined. For this reason, it is possible to judge good and bad products in consideration of the balance of the entire inspection items. In addition, since each function for each inspection item includes a coefficient determined by the importance distribution of the inspection item, it is possible to judge whether a good product or a defective product is important by considering the inspection result of a highly important inspection item. Good and bad products can be judged appropriately.

Moreover, in the said embodiment, although the evaluation value of each work contains the two values which have an OK score and an NG score, the example in which the evaluation value was arrange | positioned on the two-dimensional space was shown, but it is not limited to this. Has three values, and accordingly, the evaluation value may be arranged in a three-dimensional space.

In addition, while placing the evaluation values in the two-dimensional space and showing the example of dividing the two-dimensional space using the two straight-line boundaries A and B based on the dispersion state of the evaluation values, the boundaries A and B have a straight line shape It is not limited to that of the thing, and may be a curved boundary.

In addition, although an example of dividing the two-dimensional space into regions α, β, γ, and δ representing four grades using two straight-line boundaries A and B is shown, it is not limited thereto. You may divide into each area | region which shows a dog or more grades.

In addition, in the above description, the present invention is supposed to be applied to an appearance inspection device, but the present invention can be applied to an image processing device of a taping device or the like other than the appearance inspection device.

1: Linear feeder 2: Transfer table
6: ionizer 7: alignment guide
8: Side camera part 9: Inner camera part
10: top camera section 11: bottom camera section
12: front camera unit 13: rear camera unit
20: control unit 30: appearance inspection device of the work

Claims (6)

A method for judging a test result of a work in which a plurality of work items are inspected for a plurality of work items and a work grade is determined based on the test results of all the test items,
A step of inspecting a plurality of inspection items on the plurality of workpieces;
For each work, based on the inspection results of all inspection items, a process of obtaining the evaluation value using a combination function combining functions determined for each inspection item,
The process of placing the evaluation value of each work in the desired space,
And determining a boundary for dividing the desired space into regions representing a plurality of grades in the desired space based on the distribution state of the evaluation values arranged in the desired space. Way. According to claim 1,
The inspection is an image inspection, characterized in that the inspection result determination method of the work. The method of claim 1 or 2,
In the step of obtaining the evaluation value, the function determined for the inspection result item includes a coefficient determined by the importance distribution of the inspection item, and the method for judging the inspection result of the work. The method of claim 1 or 2,
In the step of setting the boundary, the method for determining the inspection result of a work, characterized in that the region indicating a plurality of grades includes a good product area and a bad product area. The method of claim 1 or 2,
In the step of obtaining the evaluation value, the evaluation value determined for each work includes an OK score (score) and an NG score. The method of claim 4,
A method of judging the inspection result of the work, characterized in that, after the step of setting the boundary, each work is judged as a good product or a defective product.

Images