Printed circuit board inspecting apparatus, method for determining fault type of screen printer and computer readable recording medium

• US 10,674,651 B2
• Filed: 12/22/2017
• Issued: 06/02/2020
• Est. Priority Date: 12/21/2017
• Status: Active Grant

• Current Assignee: Koh Young Technology Inc.
• Original Assignee: Koh Young Technology Inc.
• Inventors: Kim, Jae Hyung, Noh, Jeong Kyu, Lee, Jae Hwan, Lee, Duk Young
• Primary Examiner(s): Carter, Aaron W
• Application Number:
  US15/852,646
• Publication Number:
  US 20190200494A1
• Time in Patent Office: 893 Days
• CPC Class Codes: B41F 15/26 for articles with flat surfaces
• CPC Class Codes: B41F 33/0036 Devices for scanning or checking the printed matter for quality control
• CPC Class Codes: B41F 33/0045 for automatically regulating the ink supply
• CPC Class Codes: G01B 11/25 by projecting a pattern, e.g. one or more lines, moiré fringes on the object G01B11/255 takes precedence ; image analysis for depth or shape recovery G06T7/50
• CPC Class Codes: G01N 2021/95646 Soldering
• CPC Class Codes: G01N 21/956 Inspecting patterns on the surface of objects contactless testing of electronic circuits G01R31/308; testing currency G07D; manufacturing processes per se of semiconductor devices implementing a measuring step H01L22/10
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: H05K 13/08 Monitoring manufacture of assemblages
• CPC Class Codes: H05K 3/1216 by screen printing or stencil printing
• CPC Class Codes: H05K 3/3485 Applying solder paste, slurry or powder thick film methods for applying conductive paste or ink patterns H05K3/12

Abstract

A printed circuit board inspection apparatus generates at least one image indicating whether an anomaly associated with a plurality of solder pastes printed on a first printed circuit board is detected, if an anomaly in at least one solder paste of the plurality of solder pastes is detected by using an image of the first printed circuit board, obtains at least one value indicating relevance between at least one fault type and the generated at least one image, using the machine-learning based model, and determines a fault type associated with the at least one solder paste in which the anomaly is detected from at least one fault type, based on the obtained at least one value.

Claims

1. A printed circuit board inspection apparatus comprising:

a memory configured to store a machine-learning based model that derives at least one value indicating relevance between at least one fault type of a screen printer and an image indicating whether an anomaly associated with a plurality of solder pastes printed on a printed circuit board is detected; and

a processor electrically connected with the memory,

wherein the processor is configured to;

generate at least one image indicating whether an anomaly associated with a plurality of solder pastes printed on a first printed circuit board is detected, if the anomaly in at least one solder paste of the plurality of solder pastes is detected by using an image of the first printed circuit board,

obtain at least one value indicating relevance between at least one fault type and the generated at least one image, using the machine-learning based model, and

determine a fault type associated with the at least one solder paste in which the anomaly is detected from the at least one fault type, based on the obtained at least one value,

wherein the at least one solder paste in which the anomaly is detected and at least one solder paste in which the anomaly is not detected among the plurality of solder pastes printed on the first printed circuit board are distinguished from each other in the at least one image generated by the processor.

2. The apparatus of claim 1, wherein the processor detects whether the anomaly in the at least one solder paste of the plurality of solder pastes occurred using first measured information about the plurality of solder pastes obtained from the image of the first printed circuit board.

3. The apparatus of claim 2, wherein the first measured information includes at least one of a volume, an area, a height, a shape, or a position of a solder paste.

4. The apparatus of claim 3, wherein the memory further stores second measured information about a plurality of solder pastes respectively printed on a plurality of second printed circuit boards that have been inspected by the apparatus before the first printed circuit board is inspected, and

wherein the processor detects whether the anomaly in the at least one solder paste of the plurality of solder pastes occurred by using the first measured information and second measured information.

5. The apparatus of claim 4, wherein the second measured information is generated by combining sequentially measured information about the plurality of solder pastes respectively printed on the plurality of second printed circuit boards in accordance with an inspection sequence of the plurality of second printed circuit boards.

6. The apparatus of claim 5, wherein the processor determines changes in the measured information about the plurality of solder pastes respectively printed on the plurality of second printed circuit boards according to the inspection sequence of the plurality of second printed circuit boards on the basis of the second measured information, and determines the fault type associated with the at least one solder paste in which the anomaly is detected from the at least one fault type based on the determined changes in the measured information.

7. The apparatus of claim 1, wherein the at least one fault type includes a fault of a squeegee blade, a fault in fixing of the squeegee blade, a fault of a support part, a fault of a table, a fault due to poor setting of a grid lock, a fault due to negligence and poor kneading of a solder paste, a fault due to a deficit of a solder paste, and a fault due to poor contact of a stencil.

8. The apparatus of claim 1, wherein the machine-learning based model is a CNN (Convolution Neural Network) model.

9. The apparatus of claim 1, wherein the processor performs learning on the machine-learning based model through a plurality of images indicating whether an anomaly in a plurality of solder pastes printed on a printed circuit board is detected, classified in accordance with at least one fault type.

10. The apparatus of claim 9, wherein the processor performs learning on the machine-learning based model through the generated at least one image, after the fault type associated with the at least one solder paste in which the anomaly is detected is determined.

11. The apparatus of claim 1, wherein the processor determines the fault type corresponding to value over a predetermined threshold of the obtained at least one value as the fault type associated with the at least one solder paste in which the anomaly is detected.

12. A non-transitory computer-readable recording medium having a program to be executed by a computer, wherein the program includes executable instructions that instruct a processor, when the program is executed by the processor, to perform:

generating at least one image indicating whether an anomaly associated with a plurality of solder pastes printed on a first printed circuit board is detected, if an anomaly in at least one solder paste of the plurality of solder pastes is detected by using an image of the first printed circuit board;

obtaining at least one value indicating relevance between at least one fault type and the at least one image using a machine-learning based model that derives at least one value indicating relevance between at least one fault type of a screen printer and an image indicating whether an anomaly in the plurality of solder pastes printed on a printed circuit board is detected; and

determining a fault type associated with the at least one solder paste in which the anomaly is detected from the at least one fault type, based on the obtained at least one value,

wherein the at least one solder paste in which the anomaly is detected and at least one solder paste in which the anomaly is not detected among the plurality of solder pastes printed on the first printed circuit board are distinguished from each other in the at least one image generated by the processor.

13. The computer-readable recording medium of claim 12, wherein the executable instructions further instruct the processor to perform detecting whether the anomaly in the at least one solder paste of the plurality of solder pastes occurred, using first measured information about the plurality of solder pastes obtained from the image of the first printed circuit board.

14. The computer-readable recording medium of claim 13, wherein the first measured information includes at least one of a volume, an area, a height, a shape, or a position of a solder paste.

15. The computer-readable recording medium of claim 13, wherein the detecting whether the anomaly in the at least one solder paste of the plurality of solder pastes occurred includes detecting whether the anomaly in the at least one solder paste of the plurality of solder pastes occurred by using the first measured information and second measured information,

wherein the second measured information is generated by combining sequentially measured information about a plurality of solder pastes respectively printed on a plurality of second printed circuit board in accordance with inspection sequence of a plurality of second printed circuit boards that have been inspected by a printed circuit board inspection apparatus before the first printed circuit board is inspected.

16. The computer-readable recording medium of claim 12, wherein the machine-learning based model is a CNN (Convolution Neural Network) model.

17. The computer-readable recording medium of claim 12, wherein the executable instructions further instruct the processor to perform learning on the machine-learning based model through a plurality of images indicating whether an anomaly associated with a plurality of solder pastes printed on a printed circuit board is detected, classified in accordance with at least one fault type.

18. The computer-readable recording medium of claim 12, wherein the determining the fault type includes determining the fault type corresponding to value over a predetermined threshold of the obtained at least one value as the fault type associated with the at least one solder paste in which the anomaly is detected.

19. A method for determining a fault type of a screen printer by a printed circuit board inspection apparatus, the method comprising:

generating at least one image indicating whether an anomaly associated with a plurality of solder pastes printed on a first printed circuit board is detected, if an anomaly in at least one solder paste of the plurality of solder pastes is detected by using an image of the first printed circuit board;

obtaining at least one value indicating relevance between at least one fault type and the generated at least one image, using a machine-learning based model that derives at least one value indicating relevance between at least one fault type of the screen printer and an image indicating whether an anomaly associated with the plurality of solder pastes printed on a printed circuit board is detected; and

determining a fault type associated with the at least one solder paste in which the anomaly is detected from the at least one fault type, based on the obtained at least one value,

wherein the at least one solder paste in which the anomaly is detected and at least one solder paste in which the anomaly is not detected among the plurality of solder pastes printed on the first printed circuit board are distinguished from each other in the at least one image generated by the processor.