JPH0719290B2 - Appearance inspection device - Google Patents

Description

Detailed Description of the Invention [Industrial applications]

The present invention relates to a visual inspection apparatus that determines whether or not the shapes of one or more target portions belonging to an object are acceptable based on a captured image of the object.

[Prior art]

Conventionally, the following method has been used to image an object and inspect the shape of the object based on the image. For example, FIG. 7 shows an image of an object in the imaging field of view, and in the figure it is assumed that the image 30 of the object is circular. 31,32
Are all areas for measurement, and are for determining whether or not the shape of the object image 30 is acceptable. That is, the region 31 is a ring shape that is larger than the target image 30 by a predetermined amount, and the region 32 is a circle that is smaller than the target image 30 by a predetermined amount. region
By measuring the area of the object image 30 at 31, 32, the pass / fail of the shape of the object is determined. That is, if any one of the area values is equal to or larger than the predetermined value, it is determined to be rejected, and if any of the area values is less than the predetermined value, it is determined to be pass. By the way, in order to accurately determine the shape, the position of the object image 30 must be correctly associated with the positions of the regions 31 and 32, that is, concentric. Incidentally, in FIG. 7, the object image 30 is at the reference position, that is, the areas 31, 3
It is located concentrically with 2. However, in general, the object image 3
The position of 0 does not become concentric with the areas 31 and 32, and is usually slightly shifted. Therefore, at that time, as described below, a measure is taken to find the deviation of the object image 30 from the reference position and correct the positions of the regions 31 and 32 accordingly. FIG. 8 is an explanatory diagram for obtaining the positional deviation of the object image 30, that is, the deviation from the reference position. In FIG. 8, the rectangular regions 33 and 34 are located at the right end portion and the upper end portion of the object image 30. It is fixed so as to include. Based on the object image 30 in the regions 33 and 34, the positions X1 and Y1 of the right end portion and the upper end portion thereof are obtained. As is well known, how to obtain this position is based on the projection of the object image 30 on the X axis and the Y axis. Thus, the reference positions Xo and Yo in the X-axis and Y-axis directions at these points are set.
Deviations from (see Fig. 7) Xd (= X1-Xo), Yd (= Y1-Y)
o) Next, according to this deviation Xd, Yd, the regions 31, 32
Is corrected to be concentric with the object image 30.

[Problems to be Solved by the Invention]

The conventional technique described above causes inconveniences in the following cases. Now, when a portion belonging to the object, such as a hole, is the object of shape inspection, even if a correction area corresponding to the areas 33 and 34 in the conventional example is provided for this hole, a relatively small amount of the object Due to the position change, the correction area is displaced from the target hole, and the position of the measurement area based on the displacement cannot be corrected. In other words, when the position change (the degree of movement) of the object is large compared to the size of the target portion, visual inspection of the target portion by the conventional method becomes impossible. More generally, the object moves slightly from the reference position within the imaging field of view, the object size changes slightly from the reference position, and the mutual positional relationship when there are multiple target parts changes slightly from the reference. If so, the correct pass / fail judgment of the shape of the target portion cannot be made. Moreover, the change from the reference position or the reference size of the object as described above is not a special matter, but rather can occur normally. The problem to be solved by the present invention is to solve the above problems of the conventional technique, and even if there are some changes in the object position and the object size, there are some changes in the mutual positional relationship when there are a plurality of target parts. However, it is still another object of the present invention to provide an appearance inspection device capable of accurately determining whether the shape of a target portion is acceptable or not.

[Means for Solving the Problems]

In order to solve this problem, the present invention relates to an apparatus for determining the pass / fail of the shapes of a plurality of target portions belonging to an object based on an imaged image of the object. A measurement area generating means for generating a plurality of measurement areas, a fixed correction area including the entire image of the object, an image of each target portion and a predetermined relationship portion of the object having a predetermined positional relationship with the target portion are displayed. A correction area generation unit that generates a plurality of correction areas that can move in the fixed correction area so as to include a center of gravity detection for the entire image of the object in the fixed correction area The eccentric position is obtained, the generation positions of the plurality of movable correction areas are corrected according to the first eccentric position, and then the images included in each of the plurality of movable correction areas are corrected. Then, the end point detection is performed to obtain a deviation position of the predetermined relational part of the object from the reference position, and position correction means for correcting the generation positions of the plurality of measurement regions according to the second deviation position, And a feature amount measuring unit that obtains a feature amount of the image of the target portion in each measurement region, and the generation of the plurality of measurement regions based on the first and second offset positions obtained by the position correction unit. After correcting the respective positions, it is characterized in that the pass / fail of the shape of each target portion is determined based on the result output from the feature amount measuring means.

[Action]

The measurement area generation means generates a measurement area corresponding to the image of the target portion, and the correction area generation means generates a fixed correction area including the object, each target partial and an object in a predetermined positional relationship with the fixed correction area. Each movable correction area including a predetermined related portion is generated. Then, the position correction means corrects the position of each movable correction area according to the deviation of the predetermined position of the object included in the fixed correction area from the reference position, and the predetermined relationship included in each movable correction area. The position of each measurement region is corrected according to the deviation of the location from the reference position. Next, the feature amount measuring means obtains the feature amount of the image of the target portion in the measurement region, and based on this feature amount value, the pass / fail of the shape of the target portion is determined.

【Example】

An embodiment of a visual inspection apparatus according to the present invention will be described below with reference to the drawings. By the way, the point of this embodiment is to provide a correction region corresponding to the target portion in order to correct the position of the measurement region corresponding to the image of the inspection target portion belonging to the object in accordance with the change in the position of the target portion. not,
It is provided corresponding to a predetermined relation part of the object having a predetermined positional relationship with the target portion. FIG. 1 is a block diagram of a configuration of this embodiment.
Description will be given with reference to the drawings. In the figure, 1 is a TV camera for imaging, 2 is an A / D converter for digitizing the video signal of the TV camera 1, and 3 is binarization of the output of the A / D converter 2 based on a predetermined threshold value. The binarization circuits 4 and 4 are image memories for storing the output data of the binarization circuit 3. Reference numeral 5 denotes a measurement area generation circuit, which generates a measurement area corresponding to the target portion in order to measure the shape of the target portion. Reference numeral 6 denotes a correction area generation circuit, which generates two types of correction areas corresponding to a predetermined location in order to measure a change in the position of the predetermined location of the object, as will be described later in detail. A position correction circuit 7 corrects the position of the measurement region according to the change in the position of the target portion obtained based on the correction region described above. An area measuring circuit 8 measures the area of the target portion in the measurement area. A determination circuit 9 determines whether or not the shape of the target portion is acceptable based on the area value output from the area measurement circuit 8. Next, the operation of this embodiment will be described mainly with reference to the flowchart of FIG. 4 and supplementarily referring to FIGS. In FIG. 4, in step S1, a region W1 is generated, and in step S2,
The coordinates X1, Y1 of the point G, which is the center of gravity of the object image at the time of measurement in the imaging visual field, are obtained. In steps S3 and S4, the area W
Generates 2, W3. In step S5, the Y coordinate Y2 of the point P2, which is the upper end of the object image, is obtained. In step S6, the X coordinate X3 of the point P3, which is also the right end of the object image, is obtained. Now, the above contents will be supplementarily described with reference to FIG. 2 and FIG. FIG. 2 is an explanatory view of an imaging visual field when the object image 10 is at the reference position in this embodiment, and FIG.
The figure is an explanatory view of the imaging field of view when the object image 10 is deviated. 11 is an image of one target portion, and 12 is an image of another target portion. W6 and W7 are measurement areas corresponding to the target partial image 11, and correspond to areas 31 and 32 in the conventional example (see FIG. 7). Similarly, W8 and W9 are measurement regions corresponding to the target partial image 12, respectively. A point G is the center of gravity of the object image 10 and is a predetermined position in the invention. The point P2 is the upper end of the object image 10 and is one of the predetermined related places in the invention. Further, the point P3 is the right end of the object image 10 and is another one of the predetermined related parts in the invention. W1 is a fixed correction area provided corresponding to the center of gravity G and including the object image 10 with a margin, and W2 is a movable correction area provided corresponding to the point P2 and including an upper end portion of the object image 10 with a margin. , W3 is another movable correction area provided corresponding to the point P3 and including the right end portion of the object image 10 with a margin. Now, in FIG. 2, the coordinates of the center of gravity G, the point P2, and the point P3 at the reference position are (Xs1, Ys1), (Xs1, Ys2), (Xs3,
Ys1). The position correction of the measurement regions W6, W7 is performed according to the deviation of the X, Y coordinates of the point P2 from the reference position, and similarly, the position correction of the measurement regions W8, W9 is performed for the X, Y points of the point P3. It is done according to the deviation of the coordinates from the reference position. In this case, the X coordinate of the point P2 is substituted by that of the center of gravity G, and the Y coordinate of the point P3 is substituted by that of the center of gravity G. In FIG. 3, the object image 10 is slightly enlarged and deviated in a similar manner to that at the reference position shown in FIG. At this time, the coordinates of the center of gravity G, point P2, and point P3 are (X1, Y
1), (X1, Y2), and (X3, Y1). Returning to FIG. 4, at step S7 and step S8, at point P2,
Deviation from the reference position in the X direction (X1-Xs1), deviation from the reference position in the Y direction (Y2-Ys2), and deviation of the point P3 in the X direction (X3-Xs3), Y direction deviation Rank (Y1-Ys1) is required.
It should be noted that the coordinates of the center of gravity G, the point P2, and the point P3 at the reference position are obtained in advance, although the step for obtaining them is omitted in FIG. In step S9, the measurement areas W6 and W7 are set to the position correction circuit 7 (first
The position is corrected in accordance with the deviations (X1-Xs1) and (Y2-Ys2) of the point P2 from the reference position in the X and Y directions. Similarly, in step S10, the measurement areas W8 and W9 are positioned by the position correction circuit 7 in correspondence with the deviations (X3-Xs3) and (Y1-Ys1) of the point P3 from the reference positions in the X and Y directions. to correct. In steps S11, S12, S13, S14, the area measurement value A6 for the measurement region W6, the area measurement value A7 for the measurement region W7,
The area measurement value A8 is obtained for the measurement region W8, and the area measurement value A9 is obtained for the measurement region W9. Note that each area measurement is performed by the area measurement circuit 8 in FIG. The following is the determination process based on each area measurement value, and step S15
The pass / fail of the shape of the target partial image 11 is determined in steps S18 to S18, and the pass / fail of the shape of the target partial image 12 is determined in steps S19 to S22. Note that this determination processing is performed by the determination circuit 9 in FIG. 1, and a is a predetermined allowable value set for the area measurement value. Next, the shape determination of the target portion belonging to another object will be described below with reference to the drawings. FIG. 5 is an explanatory view of an imaging visual field when another object is at the reference position,
FIG. 6 is an explanatory diagram of the imaging visual field when the other object is deviated. 20 is an image of this other object, 21,22,23,24
Is an image of the target part, and corresponds to the holes formed near the side edges of the object. Each hole is roughly second
This is the same as in FIGS. In this case, the measurement area is W6, W7; W8, W9; W10, W11; W12, W13 provided corresponding to the target partial images 21, 22, 23, 24. The predetermined position is the center of gravity G of the object image 20, and the predetermined related positions are the upper end P2, the right end P3, the lower end P4, and the left end P5. The fixed correction region is a region W1 provided corresponding to the center of gravity G, and the movable correction region is a region W2 provided corresponding to the upper end P2, the right end P3, the lower end P4, and the left end P5. , W
They are 3, W4, W5. The position correction of the measurement areas W6, W7; W8, W9 is the same as that in FIGS. Also, the measurement area W10, W11; W12, W13
Is corrected in accordance with the deviation of the X and Y coordinates of the points P4; P5 from the reference position. In this case, the X coordinate of the point P4 is substituted with that of the center of gravity G, and the Y coordinate of the point P5 is substituted with that of the center of gravity G. Therefore, the pass / fail judgment of the shapes of the target partial images 21, 22, 23, and 24 is made based on the position-corrected measurement regions described above.
It will be performed according to steps S15 to S22 of the flowchart in the figure. Further, in the above description of the embodiments, the area value is shown as the characteristic amount, but it goes without saying that another characteristic amount, for example, a perimeter or the like may be used. As is apparent from the above description, the pass / fail of the shape of each target portion belonging to the object in FIGS. 2 and 3 and FIGS. 5 and 6 is determined by the positional relationship between the target portion and the predetermined relationship part of the object. As long as the above condition is maintained, even if there is a slight change from the reference in the position of the object or the size of the object, it is accurately determined. Further, when there are a plurality of target portions, for example, two or four, the shapes of the respective target portions are accurately inspected even if the positional relationship between these target portions changes slightly from the reference. By the way, the above has described the apparatus as an appearance inspection apparatus for the shape of the target portion. However, in general, the device of the present invention serves as a defect inspection device based on the appearance of the target portion when the target portion is defective or scratched, and serves as a target recognition device based on the shape of the target portion when the target portion is a character or an unknown object.

【The invention's effect】

The present invention has the following excellent effects as compared with the conventional technology. (1) Even if the object position or the object size is slightly changed from the reference, the shape of the target portion is accurately inspected. (2) When there are a plurality of target portions, the shape of each target portion can be accurately inspected even if the positional relationship between the target portions changes slightly from the reference.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment according to the present invention, FIG. 2 is an explanatory view of an imaging visual field when one object is at a reference position in this embodiment, and FIG. 4 is a flow chart showing the operation of this embodiment, FIG. 5 is an explanatory view of the imaging field of view when another object is at the reference position, and FIG. 7 is an explanatory view of an imaging visual field when the object is displaced, FIG. 7 is an explanatory view of an imaging visual field when the object is at the reference position in the conventional example, and FIG. It is an explanatory view of how to obtain the deviation when there is. Code Description 1: TV camera, 2: A / D converter, 3: Binarization circuit, 4: Image memory, 5: Measurement area generation circuit, 6: Correction area generation circuit, 7: Position correction circuit, 8: Area measurement Circuit, 9: Judgment circuit.

Claims (1)

[Claims] 1. An apparatus for determining whether a shape of a plurality of target portions belonging to the object is acceptable based on an image of an imaged object, wherein a plurality of measurement regions corresponding to the respective shapes of the respective target portions are generated. Measurement area generating means, a fixed correction area including the entire image of the object, the fixed correction area so as to include an image of each target portion and a predetermined relational portion of the object in a predetermined positional relationship with the target portion. A correction area generating means for generating a plurality of correction areas movable in the inside, and a center of gravity detection for the entire image of the object in the fixed correction area to obtain an offset position from the reference position of the object, Correcting the generation position of each of the plurality of movable correction areas according to the deviation amount of 1, and then detecting the end point of the image included in each of the plurality of movable correction areas. From each of the measurement regions, a position correction unit that obtains the displacement amount from the reference position of the predetermined relational location of the object and corrects the generation positions of the plurality of measurement regions according to the second displacement amount. And a feature amount measuring unit that obtains a feature amount of the image of the target portion, and corrects the generation positions of the plurality of measurement regions based on the first and second displacement amounts obtained by the position correcting unit. After that, the appearance inspection apparatus is characterized in that whether or not the shape of each of the target portions is acceptable is determined based on the result output from the feature amount measuring means.