JP3480482B2 - Image inspection apparatus and image inspection method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image inspection apparatus and an image inspection method for inspecting defects on the surface of an inspection target by imaging the surface of the inspection target and performing image processing on the image. .

[0002]

2. Description of the Related Art An image inspection apparatus using an image processing technique is applied to quality inspection of caps of food and drink containers. By using such a device, a large amount of products can be automatically inspected, and the inspection process can be made efficient. The image inspection apparatus according to this method captures an image of the upper surface of the cap, which is the inspection object, with a CCD camera or the like, digitizes this image signal, and binarizes the image with an appropriate threshold value. If there are defects such as stains and scratches on the upper surface of the cap, such defects are displayed as connected regions (hereinafter, also referred to as “islands”) in this binary image. By finding this connection area, we inspect for the presence of defects,
The size and shape of the defect can be known from the area (number of pixels) and the shape of the connection region.

[0003]

By the way, the conventional image inspection apparatus automatically inspected whether or not there is a defect in the product and what kind of defect the defect was, but a defect was found. In some cases, it simply removed the product from the line. However, by accumulating detailed information on the found defects, it can be used as a means for knowing a problem in the manufacturing process by utilizing this.

The present invention has been made based on the above circumstances, and provides an image inspection apparatus that can be used to find problems in the manufacturing process by using information about defects found by inspecting products. The purpose is to do.

[0005]

According to another aspect of the present invention, there is provided an image inspection apparatus for capturing an image of an object to be inspected, which is output from the image capturing means. Based on the inspection result obtained by the image processing means for performing the predetermined image processing on the image signal and inspecting the presence or absence and the type of the defect in the inspection object, a plurality of the inspection objects Based on the statistical result obtained by the inspection result totaling means and the inspection result totaling means for totaling the inspection results to obtain the statistical result,
For each lot, the total number of inspection objects and the types of defects
Numbers including numbers are displayed on the screen, and based on the above inspection results.
Then, the inspection object is displayed on the screen in numbers for each lot,
An image display unit having a function of displaying a number corresponding to the inspection object in which the defect is found in a color corresponding to the defect.

According to a sixth aspect of the present invention, an image inspection method performs predetermined image processing on an image of an inspection target captured by an image pickup unit to inspect whether there is a defect in the inspection target and its type. Then, based on the inspection result, the inspection results for the plurality of inspection objects are totaled, and the total number of inspection objects and the number of each defect type are calculated for each lot based on the statistical result obtained by the totalization. Including
Display the number on the screen, and based on the inspection result,
It is characterized in that the inspection object is displayed in numbers on the screen for each lot, and the number corresponding to the inspection object in which the defect is found is displayed in the color corresponding to the defect.

The present invention collects inspection results for a plurality of inspection objects based on inspection results obtained by image processing to obtain statistical results, and further determines a necessary part of the statistical results by a predetermined value. By displaying in a format, it becomes possible to easily examine the statistical results. As a result of this examination,
Information such as the types of defects that often occur and where many defects occur in the inspection object can be obtained.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of an image inspection apparatus according to this embodiment, and FIG. 2 is a flowchart showing a procedure for determining the presence / absence of a defect and its type in the image processing apparatus. The image inspection device of the present embodiment, on the upper surface of the cap of the PET bottle used as a beverage container,
The inspection is performed for defects such as dirt, scratches, defective shapes, and protrusions that may occur in the manufacturing process. As shown in FIG. 1, the image inspection apparatus according to the present embodiment includes a CCD camera 11 for capturing an image of the upper surface of the cap,
An image processing apparatus 12 that performs a predetermined image process on an image signal obtained by the camera 11 to determine the presence or absence of a defect and its type, and a monitor 1 that displays an image of the upper surface of the cap.
3. The operation of the image processing device 12 is controlled to manage the entire inspection, the results of the inspection performed in the image processing device 12 are totaled to obtain statistical results, and the inspector operates this to obtain statistical results. The computer 14 can refer to detailed contents. The image processing device 12
Has a built-in magneto-optical (MO) disk,
The image data of the cap can be stored here, and the image can be displayed on the monitor 13 as needed.

In FIG. 1, when a cap 10 as an inspection object is automatically conveyed to an image pickup area of a CCD camera 11, a sensor (not shown) detects it. CCD
The camera 11 starts the imaging operation at the same time when it receives the signal to that effect. In the present embodiment, the caps are transported at a speed of about 1700 per minute, and the imaging operation is also a high-speed operation corresponding thereto. The CC of the present embodiment
The D camera 11 has an imaging field of view of 34 mm and a resolution of 0.07.
mm / pic is used.

Image signals of the upper surface of the cap captured by the CCD camera 11 are sequentially sent to the image processing device 12. The image signal sent to the image processing device 12 is subjected to the following processing for each image signal of each cap. First, the image signal is converted from an analog signal to a digital signal by A /
D-converted into a digital image signal having 256 gradations, for example. Then, after noise is removed by the image filter, it is binarized with a predetermined threshold value. Image processing device 1
2 further performs particle size measurement based on the binarized image signal. Specifically, the connected area (or “island”) is labeled, the number of labeled islands is calculated, and the area value of each island is calculated. Here, the area of an island is obtained as the number of pixels forming the island.

The image processing apparatus 12 determines the presence or absence of a defect and the type thereof according to the procedure shown in FIG. 2 based on the result of the grain size measurement. In FIG. 2, “image processing result calculation” in step 1 corresponds to the above-described granularity measurement. When the image processing result is calculated, first, check whether the cap is clean (step 2), whether there is no flaw (step 4), whether the shape is not defective (step 6), and whether there is no protrusion (step 8). Sequential determination is made in this order. When it is determined that there is a defect, the defect is counted according to its type (step 3, step 5, step 7, step 9). If any one of these defects is found, the original image of the cap, that is, the image captured by the CCD camera 11 is recorded on the MO disk in the image processing device 12. At this time, the information on which position of the cap each defect was found is also recorded.

On the other hand, if none of "dirt", "scratch", "defective shape", and "protrusion" is counted (step 10), it is counted as a normal product (step 1).
1). When the above processing is completed, the inspection result is displayed on the CRT screen of the computer 14 (step 12). 3 to 8 are diagrams showing screen configurations for displaying the above inspection results. The above-mentioned inspection is conducted in units of 50 caps per lot. FIG. 3 shows an outline of the inspection results of 50 caps of a certain lot, and each number from 1 to 50 corresponds to each cap included in one lot. The shaded number means that a defect was found in the cap corresponding to that number, and on the actual CRT screen, the color corresponding to each defect (for example, red, blue, yellow, green 4 colors). When a plurality of types of defects are found for one cap, a color matching the first detected defect name is displayed.

In FIG. 3, the numbers 6, 23, 24 and 26 are shown.
Numbers # 49 are shaded, indicating that defects were found in the caps of these numbers. Figure 3
On the screen, the cap for the "TOTAL" column is 50
The number of normal caps is 50, and the total number of caps is 50 and the number of normal caps is 45. Also, "dirt", "scratch", "shape"
Each column of "protrusion" (meaning defective shape) indicates that the number of defects of each type found was two "dirt" and one other defect. . When the inspection results of lots before and after this lot are to be displayed, the "F1" key or the "F2" key is pressed on the keyboard of the computer 14.

By selecting the number of the cap in which the defect is found on the screen of FIG. 3, the details of the defect found in the cap can be displayed. The number can be selected with a mouse or a cursor, but a transparent touch panel may be provided on the surface of the CRT so that the details of the defect can be displayed only by touching the desired number with a finger.

FIG. 4 shows a screen displaying the details of defects for the sixth cap out of the 50 inspected caps. In the screen of FIG. 4, the inspection time is first displayed in the uppermost column, and the defect name of the defect found for the cap is displayed below it. To the right of each defect name, information related to that defect is displayed. That is,
As for “dirt”, the set threshold value for binarization is 256.
There are 20 of the gradations, and the number of found stains is 2. As for “flaw”, the set threshold value is 20 gradations, the length of the flaw is 2 pixels, and the number of flaws is 1
It is an individual. As for “shape defect”, the set threshold value is 20 gradations and the degree of deformation is 2%. The threshold value set for the “protrusion” is 20 gradations, the angle is 2 °, and the number is 2. If necessary, an actual image of the cap No. 6 recorded on the MO disk inside the image processing apparatus 12 is displayed on the monitor 13.
It can also be displayed above.

When more detailed inspection data than that shown in FIG. 4 is required, the defect name is selected on the screen of FIG. 4 by the same method as described above. FIG. 5 shows a screen when “dirt” is selected on the screen of FIG. On the screen of FIG. 5, for each of the two “dirt” found, the “upper left coordinate”,
Specific numerical values indicating "lower right coordinates", "center coordinates", and "area" are displayed. This makes it possible to know the position and size of the stain and its approximate shape. For other defects, a similar screen can be displayed by selecting a desired defect type on the screen of FIG.
In the screen of FIG. 5, "upper left coordinate", "lower right coordinate", "center coordinate", and "area" are displayed, but "lower left coordinate" and "upper right coordinate" are also displayed. You may do it.

FIG. 6A is a screen in which the display contents of FIG. 5 are graphically displayed. That is, the two stains shown in FIG. 5 are displayed so that it can be seen at what position and in what size and shape the actual cap upper surface exists. Then, when the No. 1 or No. 2 of the "number" in FIG. 5 is selected, the position display cursor (indicated by the "x" mark in FIG. 6B) is displayed on the image of the corresponding defect as shown in FIG. 6B. .) Moves. Alternatively, the color of the defective portion may be changed by selecting the number.
If the type of defect is a defective shape, you can see which part is deformed from the original shape and to what extent.
It is also possible to superimpose and display the portion in which the shape defect has occurred by superimposing it on the image of the standard product.

FIG. 7 shows a screen for displaying a statistical result in which the results of inspection of a certain number of caps are summarized and summarized. The screen of FIG. 7 shows that there were one defect each of “dirt”, “scratch”, “defective shape”, and “projection”. By looking at this, it is possible to obtain information as to what kinds of defects are predominant in the manufacturing line. FIG. 8 is an image in which all the found defects (indicated by “x” marks in the figure) are displayed on the cap by type, based on the statistical results of FIG. 7. By displaying in this way, it is possible to know at which position of the cap the defect has occurred. This means that if, for example, the same type of defect occurs at a specific position of the cap,
It is possible to predict that there may be a problem at a specific place in the manufacturing process, which can be useful for improving the manufacturing process.

The present invention is not limited to the above embodiment, and various modifications can be made within the scope of the invention. In the above embodiment, the present invention is applied to the defect inspection of the upper surface of the PET bottle cap, but the inspection object of the present invention is not limited to the PET bottle cap and can be applied to various inspection objects. As an example, if the present invention is applied to a defect inspection of a semiconductor wafer immediately before final shipment after polishing, physical / chemical inspection, etc., a problem in the manufacturing process of the semiconductor wafer is discovered and its improvement is performed. Can be useful for

[0020]

As described above, according to the present invention, the inspection results of a plurality of inspection objects are totaled based on the inspection results obtained by the image processing to obtain the statistical results, and the statistical results By displaying the necessary parts of them in a predetermined format, it is possible to easily examine the statistical results. As a result of this examination, the types of defects that often occur, and where many defects occur in the inspection target Or, since it is possible to obtain information such as the time at which a defect occurred, it is possible to obtain a clue as to whether or not there is a problem in the manufacturing process of the inspection object based on this information, and if so, where. Therefore, it is possible to provide an image inspection device and an image inspection method that are useful for improving the manufacturing process.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of an image inspection apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a procedure for determining the presence / absence of a defect and its type in the image processing apparatus.

FIG. 3 is a diagram showing a screen configuration for displaying an inspection result of caps for one lot in the image processing apparatus.

FIG. 4 is a diagram showing a screen configuration for displaying the content of a defect found for a certain cap.

FIG. 5 is a diagram showing a screen configuration for displaying the position and area of a specific type of defect found for a certain cap.

6 is a diagram of a screen in which a specific position of the defect displayed in FIG. 5 is graphically displayed by superimposing an image of the cap.

FIG. 7 is a diagram showing a screen configuration for displaying statistical results of defects found for a fixed number of caps.

FIG. 8 is a diagram of a screen in which the defect occurrence positions found by performing a certain number of caps on the cap image are displayed in a graphic form.

[Explanation of symbols]

10 Inspection object (cap) 11 CCD camera 12 Image processing device 13 monitors 14 Computer

Claims (10)

(57) [Claims] 1. An image pickup means for taking in an image of an inspection object, and an image processing means for performing a predetermined image processing on an image signal output from the image pickup means to inspect the presence or absence of a defect in the inspection object and its type. And, based on the inspection result obtained by the image processing means, the inspection result totaling means for totaling the inspection results of the plurality of inspection objects to obtain a statistical result, and the statistical result obtained by the inspection result totaling means. Based on the total number of inspection objects and defect types for each lot based on
Display numbers including the number of each screen, and the test formation
Based on the result, the inspection object is displayed on the screen in numbers for each lot.
And an image display unit having a function of displaying a number corresponding to the inspection object in which the defect is found, in a color corresponding to the defect. Wherein said image display means, the test results to be based <br/>, for each test object, and time of testing, the defect name of defects found, detailed information of the defect and characterized in that the also have to display the functions according to claim 1, wherein
Image inspection apparatus. 3. The image display means inspects one or more defects found for each defect type on the basis of statistical results of a plurality of inspection objects obtained by the inspection result totaling means. The image inspection apparatus according to claim 1 or 2, wherein the image inspection apparatus has a function of graphically displaying the image of the object in a corresponding position on the image. 4. The image inspection apparatus according to claim 1, wherein the inspection object is a cap of a food and drink container. 5. The image inspection apparatus according to claim 1, 2 or 3, wherein the inspection object is a semiconductor wafer. 6. An image of an inspection object taken in by an image pickup means is subjected to predetermined image processing to inspect the presence or absence of a defect in the inspection object and its type, and a plurality of the inspection results are obtained based on the inspection result. A number that includes the total number of inspection objects and the number of each defect type for each lot based on the statistical results obtained by totaling the inspection results for inspection objects
The characters are displayed on the screen, and the lock is
An image inspection method characterized in that the inspection object is displayed as a number for each image on the screen, and the number corresponding to the inspection object in which the defect is found is displayed in the color corresponding to the defect. For each lot on the basis of the method according to claim 7, wherein the test results, and the time of the inspection, a defect name of defects found, more of the defect
Image inspection method according to claim 6, wherein the displaying the fine information. 8. group statistical result obtained by the totalization
For each type of the one or more defects found by Zui defect
The image inspection method according to claim 6 or 7 , wherein the image is superposed on a corresponding position of the image of the inspection object and displayed graphically. 9. The image inspection method according to claim 6 , wherein the inspection object is a cap of a food and drink container. 10. The image inspection method according to claim 6 , wherein the inspection object is a semiconductor wafer.