KR20150104763A - Method for automatic determination of test model for optical inspection - Google Patents
Method for automatic determination of test model for optical inspection Download PDFInfo
- Publication number
- KR20150104763A KR20150104763A KR1020140026494A KR20140026494A KR20150104763A KR 20150104763 A KR20150104763 A KR 20150104763A KR 1020140026494 A KR1020140026494 A KR 1020140026494A KR 20140026494 A KR20140026494 A KR 20140026494A KR 20150104763 A KR20150104763 A KR 20150104763A
- Authority
- KR
- South Korea
- Prior art keywords
- inspection
- substrate
- model
- reference mark
- inspection model
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
- H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
- H01L21/67005—Apparatus not specifically provided for elsewhere
- H01L21/67242—Apparatus for monitoring, sorting or marking
- H01L21/67288—Monitoring of warpage, curvature, damage, defects or the like
Abstract
The present invention relates to a substrate optical inspection method capable of inspecting a substrate defect through an image of a camera, and more particularly, to a method of automatically determining an inspection model for optical inspection, Recognizing a reference mark at a predetermined position on the substrate, performing a continuous inspection when the reference mark is recognized at the position, and performing an inspection when the reference mark is not recognized at the position by changing the inspection model Step.
According to the present invention as described above, the optical inspection is immediately performed by activating the specific inspection model among the stored inspection model storage units, and only when the reference mark does not coincide with the inspection target substrate image data in the reference mark recognition step, It is possible to shorten the optical inspection time by activating other inspection models, and it is unnecessary for the operator to individually reset the inspection model as in the prior art, which simplifies the process and has a high inspection efficiency.
Description
The present invention relates to a substrate optical inspection method capable of inspecting a defect of a substrate through an image of a camera, and more particularly, to a substrate optical inspection method for automatically determining an inspection model such as an upper / And more particularly, to a substrate optical inspection system and method capable of quickly inspecting defective substrates.
Printed circuit boards (PCBs) are widely used in semiconductors, multimedia devices, communication devices, various electronic products, automobiles, and related industries, which constitute the final circuit by mounting various electronic components on the surface.
A manufacturing process of a printed circuit board is briefly described by exposing a surface of a photosensitive resin film to a photosensitive pattern using a wiring pattern film using a laminated board made of copper laminated inside and outside and etching the copper into a desired pattern using the characteristics of the photosensitive resin Thereby forming a wiring pattern (plating pattern). Thereafter, the pattern side and the insulating layer are laminated by using a press machine, holes are formed for electrical connection between the pattern layers, and the layers stacked through the holes are electrically connected. Then, it is covered with the wiring part etc. by using a photo-imageable solder resist (PSR part) to prevent copper from being oxidized by contact with the air or to prevent copper from being crushed by external impact or to prevent shorting due to metal.
Thereafter, a step of inspecting the printed circuit board is required. This is a step of inspecting various bonds such as the interval between fine line widths of the circuit pattern, short circuit to the circuit pattern, short circuit, protrusion, denting, foreign matter adhesion, appearance of the solder resist, .
That is, it is necessary to grasp the existence of the defective printed circuit board and filter it.
Conventionally, a method of inspecting the surface appearance of a substrate, such as a circuit, a terminal, a bonding pad, and a gold tab on a printed circuit board, includes a method of inspecting electric current using a probe, There is a way for the inspector to visually inspect the appearance, such as the circuit diagram and terminal, of the printed circuit board.
However, since the method of inspecting the printed circuit board using the probe merely checks the electrical connection state of the circuit, it is impossible to check whether the integrated circuit or the terminal printed on the substrate meets the standard, Since there is no such equipment, equipment is required separately.
When the inspector visually inspects the circuit diagram on the printed circuit board, there is a difference in the reliability of the inspection depending on the skill of the operator, the circuit not visible on the printed circuit board can not be inspected, As the substrate size increases, the accuracy of the inspection becomes lower.
As described above, effective inspection of the printed circuit board is a key to improving the quality and productivity, and accordingly, an optical inspection technique through camera photographing is used.
In the optical inspection technique through camera photographing, the reference image of the substrate photographed without error is compared with the photographed image of the substrate to be inspected, and the defect is judged. The inspection model is activated in the library in which the inspection reference model of the inspection target substrate is stored by checking the model, and the substrate is sequentially matched by judging whether the defect is defective or not.
Here, in order to align the substrate with the substrate, the reference mark written on the substrate is searched to align the substrate with the reference position, and then the optical image and the reference image are aligned with each other.
In addition, in order to inspect the top and bottom surfaces of the substrate, the operator first determines whether the substrate to be inspected is on the top surface in advance and sets the mode to the top surface inspection mode or the bottom inspection mode. .
However, in such a process, before the reference mark is inspected, the operator sets the inspection model in advance and the inspection is performed through the reference mark matching, resulting in a problem that the inspection speed is slow.
Accordingly, there is an increasing demand for a substrate optical inspection system and method that can overcome the unreasonable points of the conventional optical inspection method of the substrate, shorten the optical inspection time, and simplify the work process.
SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and it is an object of the present invention to provide an inspection model in which an image of a substrate to be inspected is activated in a reference mark sorting process for a substrate optical inspection, The present invention also provides a method for automatically determining an inspection model for an optical inspection, which can determine whether or not the inspection model is matched, and continuously conduct inspection or automatically activate another inspection model.
According to an aspect of the present invention, there is provided a method of automatically determining an inspection model for an optical inspection, comprising: activating a specific inspection model for inspection of a substrate; recognizing a reference mark at a predetermined position on the substrate; Performing a continuous inspection when the reference mark is recognized at the corresponding position, and changing the inspection model when the reference mark is not recognized at the corresponding position, thereby performing inspection.
Here, the inspection model is provided including a top surface inspection model for inspection of the top surface of the substrate and a bottom inspection model for the bottom inspection of the substrate.
In addition, the inspection model includes a plurality of inspection models for inspection of different types of substrates.
Further, when there are three or more inspection models, the inspection model is continuously changed until a reference mark corresponding to the inspection model is recognized.
According to another aspect of the present invention, there is provided a method of automatically determining an inspection model for an optical inspection, comprising: activating a specific inspection model for inspection of a substrate; recognizing a specific identification mark at a predetermined position on the substrate; And performing a reference mark recognition operation corresponding to the changed inspection model after changing the inspection model if the reference mark is not recognized at the corresponding position.
According to the present invention as described above, the optical inspection is immediately performed by activating the specific inspection model among the stored inspection model storage units, and only when the reference mark does not coincide with the inspection target substrate image data in the reference mark recognition step, It is possible to shorten the optical inspection time by activating other inspection models, and it is unnecessary for the operator to individually reset the inspection model as in the prior art, which simplifies the process and has a high inspection efficiency.
1 is a block diagram showing a main configuration of a substrate optical inspection system according to an automatic determination of inspection models according to the present invention.
2 is a flowchart showing a procedure of an automatic determination method for an inspection model for an optical inspection according to the present invention.
3 is a flowchart showing another embodiment of a method for automatically determining an inspection model for an optical inspection according to the present invention.
Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.
1 is a block diagram showing a main configuration of a substrate optical inspection system according to an automatic determination of inspection models according to the present invention.
The substrate optical inspection system according to the present invention includes a
The
The
The inspection target
The inspection
The inspection
Here, the first inspection model sets the inspection reference based on the top surface reference image of the specific substrate, and the second inspection model sets the inspection reference based on the bottom reference image of the specific substrate.
Alternatively, the inspection
In particular, when three or more different types of substrate inspection models are used as described above, the inspection model is continuously changed until a reference mark corresponding to the inspection model is recognized.
The
Hereinafter, a method for automatically determining an inspection model for an optical inspection according to the present invention will be described in detail.
2 is a flowchart showing a procedure of an automatic determination method for an inspection model for an optical inspection according to the present invention.
Referring to the drawings, a method of automatically determining an inspection model for an optical inspection according to the present invention includes capturing an image of a specific substrate to be inspected through a camera to obtain inspection substrate image data (S210).
In this optical inspection, the specific inspection model among the inspection models stored in the inspection
The optical inspection compares the reference image stored in the inspection
If the reference mark is recognized at a predetermined position on the substrate during the matching process, the optical inspection is continuously performed in accordance with the inspection reference order of the activated inspection model (S240) to determine the defective state of the substrate.
However, if the reference mark is not recognized at a predetermined position on the substrate, the activated inspection model is deactivated and another inspection model not activated among the inspection models stored in the inspection
For example, the inspection model stored in the inspection
However, if the specific substrate image data and the reference mark of the top surface inspection model storage unit do not match in the optical inspection step, the substrate image data is recognized as image data, the previously activated first inspection model is deactivated, So that the reference mark is matched.
In addition to the top and bottom surface inspection models, if a plurality of inspection models are not matched with the reference marks of any one of the first inspection model and the second inspection model, .
In addition, when there are three or more inspection models, the inspection model may be continuously changed until a reference mark corresponding to the inspection model is recognized.
As described above, the optical inspection time is shortened by automatically activating the second inspection model on the opposite side only when the reference mark is not coincident with the inspection target substrate image data, The worker does not have to reset the upper surface or lower surface inspection model storage unit of the substrate in the conventional manner, thereby simplifying the process.
3 is a flowchart showing another embodiment of a method for automatically determining an inspection model for an optical inspection according to the present invention.
In another embodiment of the present invention, the inspection model is not changed in the reference mark matching step when the upper and lower inspection model reference marks are the same on the surface inspection model of the substrate, In order to make the change.
More specifically, in another exemplary embodiment, after capturing an image of a specific substrate (S310), the inspection
Then, a specific identification mark is recognized at a predetermined position on the substrate to be inspected (S330). If the identification mark is recognized at the position, the reference mark is inspected on the image of the specific substrate, (S340).
Subsequently, the substrate optical inspection is continuously performed in the inspection order of the activated inspection model (S350).
That is, when the initial top surface inspection model is activated and the identification mark in the top surface inspection model is recognized in the substrate image to be inspected, the substrate to be inspected is determined as the top surface image, and the inspection is performed through matching of the reference marks.
However, if the identification mark of the activated inspection model is not recognized in the inspection target substrate image, the inspection target substrate image is judged as the image of the other surface, the activated inspection model is inactivated and the inspection model of the inactive surface is activated ) Again to recognize the identification mark.
Although the present invention has been described in connection with the above-mentioned preferred embodiments, it is possible to make various modifications and variations without departing from the spirit and scope of the invention. Accordingly, the scope of the appended claims should include all such modifications and changes as fall within the scope of the present invention.
10: CPU 20: Memory
30: Inspection substrate pickup section 40: Inspection model storage section
50: Optical inspection part
Claims (5)
Recognizing a reference mark at a predetermined position on the substrate;
Performing a continuous inspection when the reference mark is recognized at a corresponding position, and performing an inspection by changing an inspection model when the reference mark is not recognized at the corresponding position. Automatic determination method.
Wherein the inspection model includes a top surface inspection model for top surface inspection of the substrate and a bottom inspection model for bottom surface inspection of the substrate.
Wherein the inspection model includes a plurality of inspection models for inspection of different types of substrates.
Wherein when the inspection model is three or more, the inspection model is continuously changed until a reference mark corresponding to the inspection model is recognized.
Recognizing a specific identification mark at a predetermined position on the substrate;
Performing a reference mark recognition operation when the identification mark is recognized at a corresponding position, performing a reference mark recognition operation corresponding to the changed inspection model after changing the inspection model if the reference mark is not recognized at the position, And determining whether or not the inspection model is automatically determined.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020140026494A KR20150104763A (en) | 2014-03-06 | 2014-03-06 | Method for automatic determination of test model for optical inspection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020140026494A KR20150104763A (en) | 2014-03-06 | 2014-03-06 | Method for automatic determination of test model for optical inspection |
Publications (1)
Publication Number | Publication Date |
---|---|
KR20150104763A true KR20150104763A (en) | 2015-09-16 |
Family
ID=54244326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020140026494A KR20150104763A (en) | 2014-03-06 | 2014-03-06 | Method for automatic determination of test model for optical inspection |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR20150104763A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102016113522A1 (en) | 2015-07-24 | 2017-01-26 | Hyundai Autron Co., Ltd. | Method and apparatus for controlling tire pressure using a zero crossing |
KR20230027500A (en) | 2021-08-19 | 2023-02-28 | 주식회사 크레셈 | Board inspection method using a learning model |
-
2014
- 2014-03-06 KR KR1020140026494A patent/KR20150104763A/en not_active Application Discontinuation
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102016113522A1 (en) | 2015-07-24 | 2017-01-26 | Hyundai Autron Co., Ltd. | Method and apparatus for controlling tire pressure using a zero crossing |
KR20230027500A (en) | 2021-08-19 | 2023-02-28 | 주식회사 크레셈 | Board inspection method using a learning model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101118210B1 (en) | Visual inspection system of printed circuit board and method of the same | |
CN105510348A (en) | Flaw detection method and device of printed circuit board and detection equipment | |
CN108648175B (en) | Detection method and device | |
JP2006017474A (en) | Printed circuit board tester, printed circuit board assembling and testing line system and program | |
JP5045591B2 (en) | Method for creating area setting data for inspection area and board appearance inspection apparatus | |
KR20150084623A (en) | Test apparatus, test system and method of defects in blind vias of printed circuit board | |
CN110132960B (en) | Detection method of circuit board assembly | |
KR100609107B1 (en) | Installation for detecting pattern error of p.c.b. and method thereof | |
CN107592910A (en) | Method for checking electronic device | |
KR101745883B1 (en) | Apparatus and method for inspecting printed circuit boards | |
KR20150104763A (en) | Method for automatic determination of test model for optical inspection | |
KR20160105082A (en) | Apparatus for testing board and testing method thereof | |
TW202104876A (en) | Printed circuit board repair method and system thereof | |
KR101410218B1 (en) | Method of detecting an unmounted chip on a substrate | |
JP3225067B2 (en) | Lead measurement method | |
KR20120022272A (en) | Apparatus for inspecting surface of substrate | |
KR101570023B1 (en) | Mehtod of automatically classifying PCB | |
TWM583616U (en) | Printed circuit board inspection system | |
KR101148669B1 (en) | The apparatus for testing the substrate | |
JP4087973B2 (en) | Inter-component inspection method and apparatus for mounted components | |
KR101495323B1 (en) | Flip chip alignment check apparatus using vent hole and method thereof | |
JP2006080216A (en) | Quality evaluation method and testing apparatus of substrate | |
KR102450824B1 (en) | Inspection method for substrate warpage of board using a 3d learning model | |
JP2009176826A (en) | Inspecting method for printed wiring board, and device therefor | |
KR19980014808A (en) | Inspection system of printed circuit board and its inspection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
E902 | Notification of reason for refusal | ||
E601 | Decision to refuse application |