Abstract
In Thin-Film Transistor Liquid-Crystal Display (TFT-LCD) manufacturing, conducting a machine learning based system with multiple data types has become actively desired to solve complicated problems. This paper proposes a multi-modal learning approach: <i>TabVisionNet</i>, which is modeled by utilizing the information from both tabular data and image data. A novel attention mechanism called <i>Sequential Decision Attention</i> was integrated into the multi-modal modeling framework that improves the comprehension of the information from two modalities. This cross-modal attention mechanism can capture the complex relationship between modalities then gain better generalization and faster convergence in the training process. Conducting an experiment, the performance of our novel approach was significantly better than single-modal and other multi-modal learning approaches in our real case scenario.
Highlights
In thin-film transistor liquid-crystal display (TFT-LCD)industry, computer vision techniques have been widely used to help manufacturers monitor abnormalities, identify potential process bottlenecks, and swiftly respond to process problems to reduce yield loss
The architecture of the proposed approach contains three main components: 1) TabNet [69] encoder is used as the tabular encoder to extract features from the structure data, e.g. the defect log or parameters collected from the previous process; 2) Convolutional neural networks (CNNs) [70], [47], [48] is used as the vision encoder to extract the visual features from the defective images; 3) a novel attention mechanism called Sequential Decision Attention is integrated to control the vision encoder to focus on the suitable visual features by the given global context information extracted from both the image and tabular data
Deep learning techniques are gradually being developed in the TFT-LCD industry for solving complex problems in various manufacturing scenarios
Summary
Industry, computer vision techniques have been widely used to help manufacturers monitor abnormalities, identify potential process bottlenecks, and swiftly respond to process problems to reduce yield loss. Deep learning models are being gradually applied in the TFT-LCD industry for detecting and identifying panel defects [4], [5], [6]. Cases where MMML is implemented on industrial problems are very rare, especially in TFT-LCD manufacturing. This sparked an opportunity as tabular and image data are commonly collected during TFT-LCD manufacturing processes, and it would be perfect to construct a learning framework to model tabular and image data simultaneously. A novel deep learning framework is proposed for the repair rate improvement by utilizing a MMML. A novel deep neural network architecture which can be trained with multi-modal data is proposed to improve the repair rate in the real TFT-LCD manufacturing cases.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
