JPH09147111A - Threshold value determining device and inspecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To speedily and accurately determine a threshold value and easily inspect a product. SOLUTION: Plural inspection ranges are set across the edge of the mold surface of an IC, projections of the inspection ranges are generated, and threshold values of the projections of the respective inspection ranges are found (ST1). One point having the same density value with the threshold value is set for each inspection range (S12). A straight line is drawn passing the one point, the range encircled with the straight line is determined as an image processing range, and specific inspection is performed in the image processing range (ST3). Two inspection ranges are generated across each side of the mold surface of the IC. The density difference between the mold surface of the IC and its background is >=30 gradations and when the density difference is considered to be 100%, (n)% density is decided as threshold value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for determining a projection threshold value for defining a boundary between an inspection object such as a mold surface of an IC and its background, and an inspection apparatus for the inspection object.

[0002]

2. Description of the Related Art Conventionally, an image processing technique is often used for inspecting an inspection object such as a mold surface of an IC. However, for example, the range of the IC mold surface (image processing range)
In determining the above, conventionally, an operator visually confirms the range of the image taken in the image processing apparatus.

Therefore, as shown in FIGS. 11 and 12,
In the case where the density difference between the mold surface 21 of the IC and the background 22 (or the lead 23) in the image 20 is small,
The boundary between the mold surface 21 and the background 22 thereof, that is, the image processing range Sg, is largely deviated from the range S1 + S2 of the actual mold surface of the IC.

[0004]

As described above, conventionally, since the image processing range includes a personal error of the operator, it is necessary to inspect an object to be inspected such as an IC molded surface. I can't do it accurately. Therefore, conventionally, there is a drawback in that it is difficult to accurately and promptly determine the boundary between the inspection object and its background.

The present invention has been made to solve the above-mentioned drawbacks, and its purpose is to provide a boundary between an inspection object and its background,
That is, the image processing range is accurately and promptly determined, and the inspection target is inspected in the image processing range.

[0006]

In order to achieve the above object, a projection threshold value determining device of the present invention creates a projection of an image in which an object to be inspected and its background are projected, and determines the density of the object to be inspected. A means for obtaining the density of the background, and a density difference between the inspection object and the background of 1
When the density is set to 00%, the density of n% is used as a threshold value, and the density difference between the inspection object and the background is 30 gradations or more.

The inspection apparatus of the present invention creates means for setting a plurality of inspection ranges straddling the edges of the inspection target and an projection of the plurality of inspection ranges for an image showing the entire inspection target. , Means for obtaining a projection threshold value of each inspection range, means for setting at least one point having the same density value as the threshold value in each inspection range, and a straight line passing through the at least one point, An area surrounded by a straight line is defined as an image processing area, and means for performing a predetermined inspection in the image processing area is provided.

The inspection object has a quadrangle, and each side of the inspection object is covered with at least two inspection ranges. The image processing range is determined by drawing a connecting straight line.

The density difference between the inspection object and its background is 3
When the gradation is 0 or more and the density difference between the inspection object and the background is 100%, the density of n% is set as the threshold value. The inspection apparatus according to the present invention comprises means for setting the image processing range obtained by the inspection apparatus according to claim 2 as a first image processing range for an image in which the entire first inspection target object that is a non-defective item is projected. A means for setting an image processing range obtained by the inspection device according to claim 2 as a second image processing range for an image showing the entire second inspection object, the first image processing range, and the second image processing range. Compare the image processing range,
If both are the same size, a predetermined inspection is performed in the second image processing range of the second inspection object, and if they are different sizes, the second inspection object is a defective product. I have it.

The density difference between the inspection object and its background is 3
When the gradation is 0 or more and the density difference between the inspection object and the background is 100%, the density of n% is the first or second threshold value.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a projection threshold value determining device and an inspection device of the present invention will be described in detail with reference to the drawings. 1 to 3 show a projection threshold value determining device and an inspection device according to an embodiment of the present invention. Note that FIG. 1 is a diagram showing the appearance of the threshold value determining device and the inspection device of the present invention, and FIG.
Are flowcharts showing the operation of the image processing apparatus of FIG.

The threshold value determining device and the inspection device of the present invention are composed of a camera 11 and an image processing device 12 for taking in an image projected by the camera 11 and performing a predetermined process. The camera 11 can be composed of a CCD camera, a video camera, a digital camera, or the like. In addition, 13 has shown the IC package used as an inspection object.

Next, the operation of the threshold value determining device and the inspection device of the present invention will be described with reference to the flowcharts of FIGS. An example will be described below in which the boundary between the IC mold surface and its background is determined and a predetermined inspection is performed with the IC mold surface as the image processing range.

First, a normal IC (non-defective product) is placed at a predetermined position, and an image of the entire IC is taken into the image processing apparatus by the camera. Here, in the image, as shown in FIG. 4, for example, one pixel is represented by 256 gradations (8 bits), and Px ×
Suppose it has a size of Py pixels. Also, IC
The density of the mold surface is high (dark) and the density of the background is low (bright).

Further, as shown in FIG. 4, one of the four sides (boundary) of the range representing the mold surface of the IC in the image having the size of Px × Py pixels is defined as one side. Multiple (for example, eight) inspection ranges 1 to 8 are set so as to straddle.

Here, it is preferable that at least two inspection ranges extend over each side of the range representing the mold surface of the IC (steps ST1 and ST11). Next, FIGS. 5 and 6
As shown in, the projection in the direction (X direction or Y direction) perpendicular to the side (boundary) is created in each inspection range k (1 to 8) (steps ST1 and ST12). Then, the threshold value of each projection is obtained based on the shape of each projection (steps ST1, ST1).
3).

The projection threshold determination method is as follows:
It is possible to use a method of using a derivative (a method of setting a portion having the maximum projection inclination as a threshold value), a method of setting a predetermined density as a threshold value based on the density of the mold surface and the background.

A method for determining the threshold value based on the density of the mold surface and the density of the background will be described below. In this method, as shown in FIG. 6, the density difference between the density of the mold surface (highest (dark) density) and the density of the background or the lead (lowest (light) density) is set to 100%. , N% concentration is used as the threshold value.

That is, the threshold value Nth can be determined by the equation (1). Nth = (A−B) × (H ÷ 100) (1) (where A is the density of the mold surface (highest density), and B is the density of the background (or lead) ( (Lowest concentration)
And H is a correction factor. Note that the value of H is arbitrarily determined from empirical rules and the like. Further, the difference (AB) between the density of the mold surface and the density of the background thereof is preferably 30 gradations (HEX) or more in order to obtain an accurate threshold value.

As a result, as shown in FIG. 7, the projection threshold values for all the inspection ranges k (1-8) are obtained. Therefore, the coordinates of a point (pixel) Pk (x, y) having a threshold value is obtained in the center of each inspection range. However, the origin of the coordinates is one of the four corners of the image 20 (for example, point O). Further, k is 1 to 8, and it is sufficient to obtain one point in each inspection range.

Then, when the points of the two inspection ranges extending over the same side are connected to each other by a straight line, the intersection points PA to PD of the respective straight lines are obtained, and the image processing range S1 is determined. Since the image processing range S1 is a reference for the inspection of the mold surface of the IC (product) performed later,
It is registered in the image processing apparatus (step ST2).

As described above, if at least two inspection ranges are arranged so as to extend over each side of the range representing the mold surface of the IC, a more accurate image processing range S1 can be determined. Next, an IC to be inspected is placed at a predetermined position, and an image of the entire IC is captured by the camera into the image processing apparatus. Here, it is assumed that the shape of the IC to be inspected matches the shape of the reference IC from which the image processing range S1 is obtained.

Further, as shown in FIG. 4, a plurality of (e.g., eight) inspection ranges 1 are arranged so as to straddle a predetermined one of the four sides (boundaries) of the range representing the mold surface of the IC. Set ~ 8. It should be noted that at least two inspection ranges are preferably arranged on each side of the range representing the mold surface of the IC (steps ST3 and ST11).

Next, as shown in FIGS. 5 and 6, within each inspection range k (1 to 8), a projection in the direction (X direction or Y direction) perpendicular to the side (boundary) is created (step). ST3, ST12). Then, the threshold value of each projection is obtained based on the shape of each projection (steps ST3 and ST13).

As a method of determining the projection threshold, a method using differentiation (a method in which the projection has the largest inclination) is used, or a predetermined density is set as the threshold based on the density of the mold surface and the density of the background. The method of doing is used.

For example, in the method of determining the threshold value based on the density of the mold surface and the density of the background, as shown in FIG.
The density difference between the density (highest (dark) density) of the mold surface and the density of the background or lead (lowest (lightest) density) is 100%, and n% density is the threshold value.

That is, the threshold value Nth can be determined by the above equation (1). As a result, as shown in FIG. 7, projection threshold values for all inspection ranges k (1 to 8) are obtained.

Therefore, the coordinates of a point (pixel) Pk (x, y) having a threshold value is obtained in the center of each inspection range. However, the origin of the coordinates is one of the four corners of the image 20 (for example, point O). Further, k is 1 to 8, and it is sufficient to obtain one point in each inspection range.

Then, when the points of the two inspection ranges extending over the same side are connected to each other by a straight line, intersections PA to PD of the respective straight lines are obtained, and the image processing range S2 is determined (step ST4).

Next, it is checked whether the image processing ranges S1 and S2 are equal. If the image processing ranges S1 and S2 are equal, it is inspected whether or not there are defects or stains in the image processing range (mold surface) S2 (step ST6).
If there is no defect or stain in the image processing range S2, the IC (product) is determined to be a good product, and the process for the next IC is started. If the image processing range S2 has defects or stains, the IC (product) is determined to be defective and the next I
The process moves to C (step ST7).

A case where the image processing ranges S1 and S2 are not equal will be described. Such a state occurs when a chip 24 exists on the edge of the mold surface of the IC as shown in FIG. In this case, in the projections of the inspection ranges 1 and 2, the boundary between the mold surface and the background becomes ambiguous as shown in FIG.

Therefore, as shown in FIG. 10, the boundary S2 between the mold surface and its background obtained by the image processing is much more inside than the boundary between the actual mold surface and its background.

Therefore, the image processing range S2 is largely deviated from the range of the mold surface and does not coincide with the image processing range S1. If the image processing ranges S1 and S2 are not equal, the IC (product) is determined to be defective and the next IC
Move on to processing.

As described above, according to the present invention, the inspection range is provided so as to extend over each side of the mold surface of the IC, which is the inspection object, and the projection of the inspection range is created.
The threshold is calculated and the boundary between the mold surface of the IC and its background,
That is, the image processing range is determined.

Therefore, as compared with the case where the image processing range is visually determined, the boundary between the inspection object (mold surface of the IC) and the background thereof, that is, the image processing range can be determined more accurately and promptly. it can.

Further, according to the present invention, the image processing range S of non-defective products is obtained.
After obtaining 1, the image processing range S2 of the product to be inspected is obtained, and both are compared to distinguish good products from defective products.

That is, the inspection of the product (in particular, the inspection of whether or not there is a chip on the edge) can be performed at the same time when the image processing range required for the inspection within the mold surface of the IC is obtained. Therefore, the simplification of the inspection can be achieved.

The present invention can be widely applied to the case of determining the projection threshold value such as the boundary between the inspection object (not limited to the IC) and its background. For example, the border between the IC lead and background, or the IC lead and mode.
It is also possible to define the boundaries of the rudder surface. Further, it can be widely applied not only to the chipping of the edge of the mold surface of the IC but also to the chipping of the edge of the inspection object.

[0039]

As described above, the projection threshold value determining device and the inspection device of the present invention have the following effects. An inspection range is provided so as to extend over each side of the inspection object, a projection of the inspection range is created, and a threshold is obtained to determine the boundary between the mold surface of the IC and its background, that is, the image processing range. doing. Therefore,
It is possible to determine the boundary between the inspection object and the background thereof, that is, the image processing range, more accurately and more rapidly than the case where the image processing range is visually determined.

After obtaining the image processing range S1 of the non-defective product, the image processing range S2 of the product to be inspected is obtained,
By comparing the two, the good product and the defective product are distinguished. That is, it is possible to perform a predetermined one of the product inspections (in particular, the inspection of whether or not there is a chip on the edge) at the same time when the image processing range is obtained, and to achieve the ease of the inspection. You can

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of a threshold value determination device and an inspection device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the image processing apparatus of FIG. 1;
G.

FIG. 3 is a flowchart showing the operation of the image processing apparatus of FIG. 1;
G.

FIG. 4 is a diagram showing an image taken in the image processing apparatus.

5 is an enlarged view showing an inspection range of the image of FIG.

FIG. 6 is a view showing a projection within the inspection range of FIG.

FIG. 7 is a diagram showing an image processing range S1 registered in advance.

FIG. 8 is a diagram showing an image taken in the image processing apparatus.

9 is a diagram showing projections within inspection ranges 1 and 2 of the image of FIG.

FIG. 10 is a diagram showing an image processing range S2 of an IC to be inspected.

FIG. 11 is a diagram showing a relationship between a conventional image processing range and a range of a mold surface.

FIG. 12 is a diagram showing a case where the IC of FIG. 11 is viewed from the side.

[Explanation of symbols]

 11: camera, 12: image processing device, 13: IC, 20: image, 21: mold surface, 22: background 23: lead, 24: lack of mold.

Claims (6)

[Claims] 1. A means for generating a projection of an image in which an inspection object and its background are projected to obtain the density of the inspection object and the density of the background; and a density difference between the inspection object and the background of 100. A threshold value for the density of n% when it is defined as%, and the density difference between the inspection object and the background is 30 gradations or more. 2. A means for setting a plurality of inspection ranges straddling the edges of the inspection target for an image in which the entire inspection target is projected, and projections of the plurality of inspection ranges are created, and respective inspection ranges are created. A means for obtaining a projection threshold value of, a means for setting at least one point having the same density value as the threshold value in each inspection range, a straight line passing through the at least one point, and surrounded by the straight line. An inspection apparatus comprising: a range as an image processing range, and means for performing a predetermined inspection in the image processing range. 3. The inspection object has a quadrangular shape, and each side of the inspection object is covered with at least two inspection ranges, and each of the at least two inspection ranges straddling the same side. The inspection apparatus according to claim 2, wherein the image processing range is determined by drawing a straight line connecting the points. 4. The inspection apparatus according to claim 2, wherein the density difference between the inspection object and its background is 30 gradations or more, and the density difference between the inspection object and the background is 100%. In addition, the inspection apparatus is characterized in that the concentration of n% is used as the threshold value. 5. A means for setting an image processing range obtained by the inspection device according to claim 2 as a first image processing range for an image in which the entire first inspection target object which is a non-defective product is projected, A means for setting an image processing range obtained by the inspection device according to claim 2 as a second image processing range for an image in which the entire inspection object is projected, the first image processing range and the second image processing. The ranges are compared, and if the two have the same size, a predetermined inspection is performed in the second image processing range of the second inspection target, and if the two have different sizes, the second inspection target is a defective product. An inspection apparatus comprising: 6. The inspection apparatus according to claim 4, wherein the density difference between the inspection object and its background is 30 gradations or more, and the density difference between the inspection object and the background is 100%. In addition, the inspection apparatus, wherein the concentration of n% is the first or second threshold value.