JP2003279331A - Appearance examining apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To register dictionary data and a decision parameter with a high recognition rate without depending upon an ability in a short time without stopping a production to regulate a decision registering means. <P>SOLUTION: An appearance examining apparatus comprises an image processing unit 103 for processing an imaged image of a mounting component 2 imaged by a camera 6, an image storage unit 201 for storing the image, a decision parameter section 203 having a decided threshold value, a standard dictionary section 204 and a user dictionary section 205 having feature data of a character, an examined result storage unit 207 for storing a decided result, and a data processing unit 102 for forming examined data for examining and processing a decision of good or not. The apparatus automatically stores the image in the storage unit, conducts an examining simulation using the image stored in the storage unit by the data processing unit, processes an examination simulation of the recognition rate after the change of the parameter for deciding the good or not by using the imaged image stored in the storage unit and change of the dictionary data registering the feature of the component, and verifies the rate. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a visual inspection apparatus, and more particularly to a visual inspection apparatus for mounting components on a printed circuit board on which electronic components such as semiconductor components are mounted.

[0002]

2. Description of the Related Art Conventionally, a visual inspection apparatus of this type has a function of determining whether or not an electronic component, which is an object to be inspected and mounted on a printed circuit board, is missing, is reversely mounted, has a constant difference, or is misaligned. Have. As a conventional technique, Japanese Patent Laid-Open No. 9-126
In Japanese Patent No. 727, the image processing unit captures an image of a component on a board from the image captured by the camera, and the dimension measuring unit measures inspection information, for example, the dimensions of the pad and the positional relationship by image processing. The data processing unit that receives the inspection information from the measuring unit searches the component library unit for the inspection information of the component that matches the data.
Further, in Japanese Patent Application Laid-Open No. 8-254501, an inspection apparatus is provided with a normal inspection mode and a teaching mode.
In the teaching mode, discriminant analysis is performed using sample data as pre-processing to create a criterion, then the inspection mode is entered, and the inspection is performed according to the criterion. Here, a method is described in which a component that causes an erroneous determination is taught, added to sample data, and a determination criterion is created again. However, any of the techniques is a method of calculating an optimum judgment value within the range of the actual board, and for the final judgment reference value, an inspection of a certain amount of the inspection target component is tried,
Tuning based on empirical factors is necessary.

[0003]

In the above prior art, the dictionary data and the judgment parameters used in the inspection are set by the verification and tuning using the actual board or the actual parts, and the past boards and the parts are also set. Cannot be verified. Therefore, in order to register a dictionary with a high recognition rate, a certain amount of test trials and judgment based on empirical factors must be used, which requires individual competence, effort, and time. was there. In addition, in order to register a dictionary with a high recognition rate for the inspection target that was erroneously recognized,
Since the dictionary data is tuned using the actual inspection target that was erroneously recognized, there was a problem of stopping the production and verifying it.

An object of the present invention is to provide an appearance inspection apparatus capable of registering dictionary data and determination parameters having a high recognition rate without depending on the ability in a short time without stopping production.

[0005]

In order to solve the above problems, a picked-up image obtained by picking up an object to be inspected is automatically stored in the image storage unit, and the picked-up image stored in the image storage unit by a data processing unit is used. The image recognition rate is stored in the image storage unit without using the actual board or actual component after the inspection simulation is performed, and the recognition rate after the change of the judgment parameter for judging the quality and the change of the dictionary data in which the feature of the part is registered are changed. Verification is performed by using inspection simulation processing using images.

[0006]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows the configuration of a visual inspection apparatus according to an embodiment of the present invention. In FIG. 1, a mounting component 2 which is an object to be inspected is mounted on the upper surface of a printed circuit board 1,
The mounting component 3, which is the object to be inspected, is also mounted on the lower surface.
The annular illumination device 4 is arranged downward between the CCD camera 6 and the printed circuit board 1 on the upper surface, and the CCD camera 10 and the annular illumination device 8 are similarly arranged upward on the lower surface.
On the upper surface side, the appearance inspection apparatus moves the CCD camera 6 and the annular illumination device 4 along the X-axis moving axis 5 and the Y-axis moving axis 7 by the XY robot control unit 101 while annularly illuminating the mounted component 2 to be inspected. The device 4 illuminates and the CCD camera 6 captures an image. Further, on the lower surface side, the CCD camera 10 and the annular illumination device 8 are moved by the XY robot control unit 101 independently of the upper surface side while moving the X-axis moving axis 9 and the Y-axis moving axis 11 while mounting components to be inspected. 3 is illuminated by the annular illumination device 8 and an image is taken by the CCD camera 10. Further, the appearance inspection apparatus of the present embodiment has an image processing unit 103 that extracts the characteristics of a captured image by image processing, an image storage unit 201 that stores the captured image, and data such as the character string direction of parts. A component library unit 202, a determination parameter unit 203 having a determination threshold, a standard dictionary unit 204 and a user dictionary unit 205 having character feature data, a component placement unit 206 having component mounting position information, and a determination result saved. It has an inspection result storage unit 207 that performs inspection, and a data processing unit 102 that creates inspection data for inspection and performs quality determination processing. The captured image is subjected to image processing by the image processing unit 103 based on the data of the component library unit 202 having information such as the direction of the character string and the color of the character by the data processing unit 102, and the features are extracted. Next, the standard dictionary unit 204 and the user dictionary unit 2 having dictionary data having character characteristics as data.
The image from which the features were extracted was
R (Optical Character Read
The image processing unit 103 collates characters according to the r) method.
Next, the data processing unit 102 performs a quality determination process based on the collation result and the data of the determination parameter unit 203. The determination parameter unit 203 has information regarding the storage condition of the captured image, and stores the image in the image storage unit 201 based on the information and the quality determination result. In addition, when the captured image is stored, the board ID which is information for managing the captured image,
The mounting surface, device, and image file No. are saved in the inspection result storage unit 207 as image data management data 501 (FIG. 4). The board ID, the mounting surface, and the device are information to be associated with the quality determination result stored in the inspection result storage unit 207, and the image file NO is information to be associated with the file name of the stored captured image.

FIG. 2 shows a flow of saving the captured image and the inspection result saved in the image storage unit 201 in the inspection result storage unit 207. The inspection is started (301), and the data processing unit 10
When it is determined that the inspection item 2 is defective (302), the data processing unit 102 temporarily stops. At this point, the data processing unit 1
Since it is unclear whether the determination result of 02 is a correct report or a false report, the captured image is displayed on the screen and the operator is called by a buzzer or the like (303). The operator compares the image pickup screen with the collation character information to make a final judgment as to whether the product is a good product or a defective product (3
04), the determination result is input to the data processing unit 102 from an input unit (not shown) (305). The input result is stored in the inspection result storage unit 207, and the captured image is stored in the image storage unit 201 based on the image storage condition of the determination parameter unit 203. In the determination parameter 203, as the storage condition of the captured image, three conditions are stored: all images are stored, images determined by the data processing unit 102 to be defective, and images whose final result input by the operator is defective are stored. It has for each component ID that distinguishes the type. The above conditions are set in advance. The inspection result is stored as the inspection result data 502 (FIG. 4) in the inspection result storage unit 207 after the inspection of one substrate is completed (306).

FIG. 3 shows a flow of the inspection simulation using the saved image saved in the image storage unit 201.
The purpose here is to register dictionary data and determination parameters with a high recognition rate, and the method is to determine the final registered data by verifying the recognition rate by inspection simulation. There are two types of dictionary data for character recognition, standard dictionary data in which the characteristics of characters are registered as default values, and user dictionary data additionally registered by the user. The determination is made as these two types of dictionary data. The parameters are processed by the data processing unit 102 and the image processing unit 103. The standard dictionary data has the information in the standard dictionary unit 204 and cannot be added, changed or deleted. On the other hand, the user dictionary data has the information in the user dictionary unit 205 and can be added, changed, or deleted.
Therefore, the recognition rate is improved by changing the determination parameter of the determination parameter unit 203 that sets the pass / fail threshold value for the addition, change, and deletion of the user dictionary data and the character matching. Regarding the flow of the inspection simulation, after the user dictionary is registered or the judgment parameter is changed (401), the simulation range is set (402). Here, which part type is to be targeted, which image is to be stored in the period from what time to what time, all images, or only the image which the data processing unit 102 has erroneously recognized. Set conditions such as target. Next, an inspection simulation process is executed (403). The inspection simulation is processed by the data processing unit 102 and the image processing unit 103 according to the captured image stored in the image storage unit 201 and the inspection algorithm. As for the processing method, the processing after the image capturing is performed by the same processing as the normal inspection, and the inspection (quality judgment) result is output (404). In the inspection simulation, processing is executed based on the data such as the inspection simulation data 503 (FIG. 4), and the inspection simulation result 504 (FIG. 4) is output and stored in the inspection result storage unit 207. In the inspection simulation result 504, the state of the false alarm is confirmed, and it is confirmed that there is no oversight that the defective product is judged as a good product (405). If an oversight has occurred, the user dictionary is registered or the determination parameter is changed and the inspection simulation is performed again. In addition, regarding false reports that determine non-defective products as defective products, user dictionary registration or change of determination parameters and inspection simulation are performed multiple times, and the inspection simulation result summary 505 (FIG. 4)
Thus, the final setting data is determined (406) and the changed content is registered (407).

As described above, in the present embodiment, the picked-up image picked up by the inspection is automatically stored in the image storage unit, and the inspection simulation is performed using the picked-up image stored in the image storage unit to make a judgment for judging the quality. By changing the parameters and changing the dictionary data in which the features of the parts are registered, the recognition rate after the change of the judgment registration condition can be obtained by using the captured image saved in the image storage section without using the actual board or the actual parts. Since inspection simulation processing is performed for verification, it is possible to register dictionary data and judgment parameters with a high recognition rate, and because it is possible to adjust judgment registration conditions when erroneous recognition occurs without stopping production, equipment operation is possible. The rate can be improved.

FIG. 5 shows an application example of the inspection simulation. This is an example in which the component 601 mounted on the board is inspected, and an example in which the stamp character of the component 601 is “HG1B” and the non-defective product is “HGLB” of the component 602 is shown.
OCR is a method in which matching is performed using an index called a certainty factor, which indicates a matching ratio with a character. Here, "l" is replaced with "L".
If you recognize it, you will miss it. In the inspection here, “l” of “HG1B” is “l” with a certainty factor of 86%,
An example result of "L" is shown with a certainty factor of 34%, and this character is recognized as "l". As a result, this part is "l"
Is not “L”, it is determined to be a defective product. Next, an example in which the component 603 is recognized as a non-defective product but a defective product will be described. In this example, "LV42" is "L" when the certainty factor is 46%, and "L" when the certainty factor is 35%. As a result, "LV42" is judged to be "LV42",
Although it is a non-defective product, it is a false alarm to recognize it as a defective product. In order to increase the recognition rate of the component 603, it is necessary to make adjustment so that the character "L" can be recognized as "L" without fail. Here, the characteristics of the character "L" are registered as a user dictionary. Here is an example. User dictionary unit 2 to increase the recognition rate
When a new feature of the character “L” is registered in 05, the registered result is, for example, “L” with a certainty factor of 97%, “L”, and a certainty factor of 1
7% may result in "l". In this case, since "LV42" is determined to be "LV42", it is not a false alarm. Next, as an effect of registering this dictionary, when the component 601 in which the defect has been detected is inspected, an oversight may occur. Since the dictionary data has changed since the last check, the result may change to "L" for "HG1B" with a certainty of 57% and "l" with a certainty of 47%. Yes, "HGIB" is erroneously recognized as "HGLB", and as a result, it is overlooked.
These verifications need to be performed retroactively to the past inspection parts, but in the present embodiment, the inspection simulation process is performed using the captured image stored in the image storage unit 201 without using the actual board or the actual parts. Since such a method is used for verification, it is possible to prevent such an oversight.

[0011]

As described above, according to the present invention,
There is an effect that the dictionary data having a high recognition rate and the determination parameter can be registered in a short period of time and with a small labor without individual difference. Further, since the determination registration condition can be adjusted when the erroneous recognition occurs without stopping the production, there is an effect that the facility operation rate can be improved.

[Brief description of drawings]

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

FIG. 2 is a flow chart for explaining storage of captured images and inspection results according to the present invention.

FIG. 3 is a flow chart of an inspection simulation using a saved image of the present invention.

FIG. 4 is an explanatory diagram showing an example of inspection data of the present invention.

FIG. 5 is an explanatory diagram illustrating an application example of the inspection simulation of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Printed circuit board, 2 ... Top surface mounting component, 3 ... Bottom surface mounting component, 4 ... Top surface side annular illuminating device, 5 ... Top surface side Y-axis movement axis,
6 ... Top side camera (CCD), 7 ... Top side X-axis movement axis,
8: lower surface side annular illumination device, 9: lower surface side X-axis movement axis, 10
... lower surface side camera (CCD), 11 ... lower surface side Y-axis movement axis,
101 ... XY robot control unit, 102 ... Data processing unit,
103 ... Image processing unit, 201 ... Image storage unit, 202 ... Parts library unit, 203 ... Judgment parameter unit, 204 ...
Standard dictionary part, 205 ... User dictionary part, 206 ... Component placement part, 207 ... Inspection result storage part, 501 ... Image data management data, 502 ... Inspection result data, 503 ... Inspection simulation data, 504 ... Inspection simulation result, 505 ... Inspection simulation summary, 601
… Defective parts mounted on the board, 602… Non-defective parts

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Magari Masatoshi             5-2-1 Omika-cho, Hitachi City, Ibaraki Prefecture             Information Control Systems Division, Hitachi, Ltd.             Within (72) Inventor Hirokazu Hirakawa             5-2-1 Omika-cho, Hitachi City, Ibaraki Prefecture             Inside the Hitachi Information Control System F-term (reference) 2F065 AA01 AA07 AA14 AA17 AA20                       AA31 AA51 BB02 BB05 BB27                       CC01 CC28 DD06 FF04 GG17                       JJ03 JJ26 MM07 PP03 PP25                       QQ21 QQ24 QQ25 QQ28 QQ31                       RR05                 2G051 AA65 AB14 AC11 AC21 CA04                       DA03 EA12 EA14                 5B057 AA03 BA02 DA03 DA12 DB02                       DC33 DC39                 5E313 CC04 FG08

Claims (2)

[Claims] 1. An electronic component, which is an object to be inspected and is mounted on a printed circuit board, missed mounting, reverse mounting, difference in constant,
In an appearance inspection apparatus for determining the quality of mounting deviation, an image processing unit that performs image processing on a captured image of the inspection target object, an image storage unit that stores the captured image, and a determination parameter having a determination threshold value. Unit, a standard dictionary unit having a character feature data and a user dictionary unit, an inspection result storage unit that stores the determination result, and a data processing unit that creates inspection data for inspection and performs quality determination processing, The captured image captured is automatically stored in the image storage unit, and the data processing unit performs inspection simulation using the captured image stored in the image storage unit to change the determination parameter for determining the quality and the characteristics of parts. Characterized in that the recognition rate after the change of the registered dictionary data is verified by performing an inspection simulation process using the captured image stored in the image storage unit. That the visual inspection apparatus. 2. The appearance inspection apparatus according to claim 1, wherein the dictionary data and simulation results of each inspection before and after the change of the determination parameter value are stored, and the results are compared before and after.