AI-assisted reliability assessment for gravure offset printing system

Anton Nailevich Gafurov,Inyoung Kim,Thanh Huy Phung,Taik-Min Lee

doi:10.1038/s41598-022-07048-z

Anton Nailevich Gafurov, Inyoung Kim + Show 2 more

Open Access

https://doi.org/10.1038/s41598-022-07048-z

Copy DOI

Abstract

In printed electronics, flawless printing quality is crucial for electronic device fabrication. While printing defects may reduce the performance or even cause a failure in the electronic device, there is a challenge in quality evaluation using conventional computer vision tools for printing defect recognition. This study proposed the computer vision approach based on artificial intelligence (AI) and deep convolutional neural networks. First, the data set with printed line images was collected and labeled. Second, the overall printing quality classification model was trained and evaluated using the Grad-CAM visualization technique. Third and last, the pretrained object detection model YOLOv3 was fine-tuned for local printing defect detection. Before fine-tuning, ground truth bounding boxes were analyzed, and anchor box sizes were chosen using the k-means clustering algorithm. The overall printing quality and local defect detection AI models were integrated with the roll-based gravure offset system. This AI approach is also expected to complement more accurate printing reliability analysis firmly.

Highlights

In printed electronics, flawless printing quality is crucial for electronic device fabrication
The gravure offset is one of the printing techniques used for manufacturing various electronics, such as silver grid transparent electrodes[3], pressure sensors[4], and planar inductors[5], mainly through fine line patterning, because of the following reasons
As gravure offset printing evolved from the so-called pad printing suitable for printing electronics onto nonplanar surfaces, such as electroluminescent displays (ELDs)[8,9] and radio-frequency identification (RFID) antennas[10], it inherited its main feature: the pad-like blanket made of silicone polymer and wrapped around a cylinder

Summary

Introduction

Flawless printing quality is crucial for electronic device fabrication. The ink is transferred from the blanket to the substrate, called the set process This transferring mechanism (Fig. 2) is carried out through applied pressure and is a key factor of printing quality, which may cause the following defects to form: printed line width gain, bulge outs, and bad surface roughness. During the print run, the PDMS blanket becomes saturated by the ink solvent, which causes its absorbing ability to decrease, and the viscosity of the following ink portion is not being tailored anymore This leads the ink to be split in half during the set process, and its residuals are left on the blanket, causing bulges and roughness with local defects in the subsequent prints (Fig. 3). If the line width is within tolerance during consecutive printing but other defects are present, the failure regime of the printing should be detected, which becomes tricky when using the conventional tools[17,18] of computer vision

Methods

Findings

Conclusion