CN-122016645-A - FPC defect detection method, device, system and storage medium

CN122016645ACN 122016645 ACN122016645 ACN 122016645ACN-122016645-A

Abstract

The invention relates to an FPC defect detection method, device and system and a storage medium. The method comprises the steps of controlling an encoder to synchronously send high-frequency line trigger signals to an optical camera and a light source controller, enabling the light source controller to control a high-uniformity coaxial light source and a multi-angle high-directivity line light source of a multi-spectrum combined light source to work according to the received high-frequency line trigger signals, enabling the optical camera to collect a time-sharing line scanning image of an FPC to be detected, carrying out algorithm correction positioning on the obtained time-sharing line scanning image to obtain a correction scanning image, fusing a traditional machine vision detection algorithm and an AI deep learning detection algorithm, comparing the correction scanning image with a Gerber image of a product of the FPC to be detected, and obtaining defect information of the FPC to be detected from the correction scanning image. The invention can solve the problems of easy missed detection and insufficient compatibility caused by the limitation of the traditional light source and easy missed detection or incapability caused by the defect of the traditional algorithm.

Inventors

HUANG LIMING
HU ZONGMIN
HUANG WEIYE

Assignees

盐城维信电子有限公司

Dates

Publication Date: 20260512
Application Date: 20260114

Claims (13)

1. The FPC defect detection method is applied to an optical detection system, wherein the optical detection system comprises an optical camera, an encoder connected with the optical camera, a light source controller connected with the encoder, and a multispectral combined light source which is connected with the light source controller and consists of a high-uniformity coaxial light source and a multi-angle high-directivity line light source, and the optical camera is correspondingly matched with the multispectral combined light source; Characterized in that the method comprises: the method comprises the steps that a control encoder synchronously sends a high-frequency line trigger signal to an optical camera and a light source controller, so that the light source controller controls a high-uniformity coaxial light source and a multi-angle high-directivity line light source of a multispectral combined light source to work according to the received high-frequency line trigger signal, and the optical camera collects a time-sharing line scanning image of an FPC to be detected; Carrying out algorithm correction positioning on the obtained time-sharing line scanning image to obtain a correction scanning image; And (3) fusing a traditional machine vision detection algorithm and an AI deep learning detection algorithm, comparing and analyzing the correction scanning image with the Gerber image of the product of the FPC to be detected, and acquiring defect information of the FPC to be detected from the correction scanning image.
2. The FPC defect detection method of claim 1, wherein said causing the light source controller to control the operation of the high-uniformity coaxial light source and the multi-angle high-directivity line light source of the multi-spectrum combined light source according to the received high-frequency line trigger signal while causing the optical camera to collect a time-sharing line scan image of the FPC to be detected includes: The light source controller is controlled to drive the high-uniformity coaxial light source and the multi-angle high-directivity line light source of the multi-spectrum combined light source to work according to the received high-frequency line trigger signal and a preset scanning period, so that the multi-spectrum combined light source respectively emits uniform light, or/and low-angle light, or/and middle-angle light, or/and high-angle light to the FPC to be detected; And controlling the optical camera to acquire a time-sharing line scanning image of the FPC to be detected under the action of uniform light, or/and low-angle light, or/and middle-angle light, or/and high-angle light of the multispectral combined light source.
3. The FPC defect detection method according to claim 1, wherein the multispectral combined light source includes the high-uniformity coaxial light source and the multi-angle high-directivity line light source, and a light source connection frame connecting the high-uniformity coaxial light source and the multi-angle high-directivity line light source; The second light outlet of the multi-angle high-directivity line light source is used for obliquely corresponding to the FPC to be detected, and the second light return port of the second light outlet is positioned at the middle position of the multi-angle high-directivity line light source; the first light outlet of the high-uniformity coaxial light source is correspondingly arranged above the second light return port of the multi-angle high-directivity line light source, and the first light return port of the high-uniformity coaxial light source corresponds to the lens of the optical camera.
4. The FPC defect detection method of claim 3, wherein said highly uniform coaxial light source includes a first light source housing having a first light source cavity, and a first light-emitting lamp panel, a diffusion plate, a half-mirror, and a half-mirror, which are sequentially disposed in said first light source cavity, said first light-emitting lamp panel being disposed on one side surface of said first light source cavity, corresponding in parallel to said diffusion plate, and connected to said light source controller, said half-mirror being disposed obliquely to the other side surface of said first light source cavity, corresponding in inclination to said diffusion plate; the bottom surface of the first light source shell is provided with a first light outlet corresponding to the bottom surface of the semi-transparent semi-reflective spectroscope, and the top surface of the first light source shell is provided with a first light return port corresponding to the top surface of the semi-transparent semi-reflective spectroscope and corresponding to the lens of the optical camera.
5. The method of claim 4, wherein the high uniformity coaxial light source comprises a first water-cooled heat sink structure disposed at a side opening of the first light source housing, and the first light-emitting lamp panel (disposed at a side of the first water-cooled heat sink structure.
6. The FPC defect detection method of claim 4, wherein said high uniformity coaxial light source includes a window plate having an exit window provided at said first light return port, and a window mirror provided at said exit window on said window plate, said window mirror corresponding to a top surface of said half mirror; The first light-emitting lamp panel comprises a first light-emitting circuit board connected with the light source controller and a plurality of first LED lamp beads arranged on the first light-emitting circuit board in an array mode.
7. The FPC defect detection method according to claim 3, wherein the multi-angle high-directivity line light source includes a light source mounting frame, and each of high-directivity line light source groups symmetrically provided on both sides of the light source mounting frame, each of the high-directivity line light source groups including at least three high-directivity line light source units arranged in an arc shape on the light source mounting frame in turn; each high-directivity line light source unit is provided with a second light outlet corresponding to the inclination of the FPC to be detected, a light source gap is arranged between the two high-directivity line light source groups, and the light source gap forms a second light return opening corresponding to the first light outlet of the high-uniformity coaxial light source.
8. The FPC defect detection method of claim 7, wherein each of said high-directivity line light source units includes a second light source housing having a second light source cavity provided on said light source mounting bracket, a second light-emitting lamp panel provided on a top side of said second light source cavity and connected to said light source controller, and a fresnel lens provided at a bottom side opening of said second light source housing, said second light-emitting lamp panel being disposed opposite to said fresnel lens, an outer side of said fresnel lens forming said second light outlet for corresponding to an FPC tilt to be detected.
9. The FPC defect detection method of claim 8, wherein said high directivity line light source unit includes a second water-cooling heat dissipation structure provided at a top side opening of said second light source housing, said second light-emitting lamp panel being provided on said second water-cooling heat dissipation structure; The second light-emitting lamp panel comprises a second light-emitting circuit board which is arranged on the side face of the second water-cooling heat dissipation structure and connected with the light source controller, and a plurality of second LED lamp beads which are arranged on the second light-emitting circuit board in a row shape and have different wavelengths.
10. The FPC defect detection method of claim 1, wherein said fusing a conventional machine vision detection algorithm and an AI deep learning detection algorithm, comparing and analyzing said corrected scan image with a Gerber image of said FPC to be detected, obtaining defect information of said FPC to be detected from said corrected scan image, comprises: Preprocessing and registering the correction scanning image and the Gerber image of the product of the FPC to be detected; Comparing the registered correction scanning image with the product Gerber image by adopting a traditional machine vision detection algorithm, determining basic defects in the correction scanning image, and positioning all suspected defect areas deviating from design standards; Detecting the suspected defect area by adopting an AI deep learning detection algorithm, and determining the real defect in the suspected defect area; And outputting a judging result of the defect information of the FPC to be detected by combining the detection results of the traditional machine vision detection algorithm and the AI deep learning detection algorithm.
11. The FPC defect detection device is applied to an optical detection system, wherein the optical detection system comprises an optical camera, an encoder connected with the optical camera, a light source controller connected with the encoder, and a multispectral combined light source which is connected with the light source controller and consists of a high-uniformity coaxial light source and a multi-angle high-directivity line light source, and the optical camera is correspondingly matched with the multispectral combined light source; characterized in that the device comprises: The image acquisition module is used for controlling the encoder to synchronously send high-frequency line trigger signals to the optical camera and the light source controller at the same time, so that the light source controller controls the high-uniformity coaxial light source and the multi-angle high-directivity line light source of the multispectral combined light source to work according to the received high-frequency line trigger signals, and the optical camera acquires a time-sharing line scanning image of the FPC to be detected; the image correction module is used for carrying out algorithm correction positioning on the obtained time-sharing line scanning image to obtain a correction scanning image; And the defect acquisition module is used for fusing a traditional machine vision detection algorithm and an AI deep learning detection algorithm, comparing and analyzing the correction scanning image with the Gerber image of the product of the FPC to be detected, and acquiring defect information of the FPC to be detected from the correction scanning image.
12. An optical detection system, comprising: an optical camera; An encoder connected to the optical camera; the light source controller is connected with the encoder; The multispectral combined light source comprises a high-uniformity coaxial light source and a multi-angle high-directivity line light source which are connected with the light source controller, and the optical camera is correspondingly matched with the high-uniformity coaxial light source and the multi-angle high-directivity line light source; the control processor is connected with the optical camera, the encoder, the light source controller and the multispectral combined light source; wherein the control processor is configured to implement the FPC defect detection method of any one of claims 110.
13. A computer-readable storage medium, wherein computer-executable instructions are stored in the computer-readable storage medium, which when executed by a processor, are adapted to carry out all or part of the method steps of the FPC defect detection method as claimed in any one of claims 1 to 10.

Description

FPC defect detection method, device, system and storage medium Technical Field The invention relates to the technical field of optical detection, in particular to an FPC defect detection method, an FPC defect detection device, an FPC defect detection system and a storage medium. Background In recent years, along with the development of electronic technology, FPC (Flexible Printed Circuit) flexible circuit boards are playing more and more important functions, and compared with the traditional PCB (Printed Circuit Board), the flexible circuit boards have obvious advantages in light weight and flexibility, and can be used for bending high-rise parts such as hinges, mechanical joints and the like. Therefore, for some high-end customization applications, the effect and the advantages of the FPC flexible board product are remarkable, the added value and the profit margin of the product are very high, the quality control in the FPC production and manufacturing process is very strict, the defects in the process and the defects before FQC (Final Quality Control) shipment are checked and controlled in time, and the defective products are prevented from flowing out to the client side to cause larger loss. Therefore, in the FPC manufacturing process, accuracy and defective product interception capability of the post-etching circuit AOI (Automated Optical Inspection) inspection, and the inspection of the appearance AVI (Automatic Visual Inspection) of the flexible board product before shipment are particularly important. In the conventional technology, for AOI and AVI detection of FPC products, a line scan camera is generally used to capture images of FPC products and analyze the captured images. However, the conventional scheme has the following problems that the detection light source is monotonous, the light source angle is single, missed detection is easy to occur, the detection light source has single wave band (basically white light), the product compatibility of some special manufacturing processes is insufficient, the conventional equipment optical scheme cannot be compatible with the dual requirements of actual detection on directivity and uniformity, and the conventional detection algorithm is easy to have the conditions of missed detection or incapability of detection. Disclosure of Invention The invention provides an FPC defect detection method, device, system and storage medium, which can solve the problems that the detection omission and the insufficient compatibility are easy to occur due to the limitation of the traditional light source, and the detection omission or the detection incapability is easy to occur due to the defect of the traditional algorithm. In order to solve the technical problems, the invention provides an FPC defect detection method which is applied to an optical detection system, wherein the optical detection system comprises an optical camera, an encoder connected with the optical camera, a light source controller connected with the encoder, and a multispectral combined light source which is connected with the light source controller and consists of a high-uniformity coaxial light source and a multi-angle high-directivity line light source, and the optical camera is correspondingly matched with the multispectral combined light source; the method comprises the following steps: the method comprises the steps that a control encoder synchronously sends a high-frequency line trigger signal to an optical camera and a light source controller, so that the light source controller controls a high-uniformity coaxial light source and a multi-angle high-directivity line light source of a multispectral combined light source to work according to the received high-frequency line trigger signal, and the optical camera collects a time-sharing line scanning image of an FPC to be detected; Carrying out algorithm correction positioning on the obtained time-sharing line scanning image to obtain a correction scanning image; And (3) fusing a traditional machine vision detection algorithm and an AI deep learning detection algorithm, comparing and analyzing the correction scanning image with the Gerber image of the product of the FPC to be detected, and acquiring defect information of the FPC to be detected from the correction scanning image. The invention also provides an FPC defect detection device which is applied to an optical detection system, wherein the optical detection system comprises an optical camera, an encoder connected with the optical camera, a light source controller connected with the encoder, and a multispectral combined light source which is connected with the light source controller and consists of a high-uniformity coaxial light source and a multi-angle high-directivity line light source, wherein the optical camera is correspondingly matched with the multispectral combined light source; the device comprises: The image acquisition module is used for controlling the encoder to synchronousl