When IC meets text: Towards a rich annotated integrated circuit text dataset
Automated Optical Inspection (AOI) is a process that uses cameras to autonomously scan printed circuit boards for quality control. Text is often printed on chip components, and it is crucial that this text is correctly recognized during AOI, as it contains valuable information. In this paper, we int...
Saved in:
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Published: |
Elsevier
2024
|
Subjects: | |
Online Access: | http://eprints.um.edu.my/44285/ https://doi.org/10.1016/j.patcog.2023.110124 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.um.eprints.44285 |
---|---|
record_format |
eprints |
spelling |
my.um.eprints.442852024-07-01T04:39:56Z http://eprints.um.edu.my/44285/ When IC meets text: Towards a rich annotated integrated circuit text dataset Ng, Chun Chet Lin, Che-Tsung Tan, Zhi Qin Wang, Xinyu Kew, Jie Long Chan, Chee Seng Zach, Christopher QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering Automated Optical Inspection (AOI) is a process that uses cameras to autonomously scan printed circuit boards for quality control. Text is often printed on chip components, and it is crucial that this text is correctly recognized during AOI, as it contains valuable information. In this paper, we introduce ICText, the largest dataset for text detection and recognition on integrated circuits. Uniquely, it includes labels for character quality attributes such as low contrast, blurry, and broken. While loss-reweighting and Curriculum Learning (CL) have been proposed to improve object detector performance by balancing positive and negative samples and gradually training the model from easy to hard samples, these methods have had limited success with one-stage object detectors commonly used in industry. To address this, we propose Attribute-Guided Curriculum Learning (AGCL), which leverages the labeled character quality attributes in ICText. Our extensive experiments demonstrate that AGCL can be applied to different detectors in a plug-and-play fashion to achieve higher Average Precision (AP), significantly outperforming existing methods on ICText without any additional computational overhead during inference. Furthermore, we show that AGCL is also effective on the generic object detection dataset Pascal VOC. Our code and dataset will be publicly available at https://github.com/chunchet-ng/ICText-AGCL. Elsevier 2024-03 Article PeerReviewed Ng, Chun Chet and Lin, Che-Tsung and Tan, Zhi Qin and Wang, Xinyu and Kew, Jie Long and Chan, Chee Seng and Zach, Christopher (2024) When IC meets text: Towards a rich annotated integrated circuit text dataset. Pattern Recognition Letters, 147. ISSN 0167-8655, DOI https://doi.org/10.1016/j.patcog.2023.110124 <https://doi.org/10.1016/j.patcog.2023.110124>. https://doi.org/10.1016/j.patcog.2023.110124 10.1016/j.patcog.2023.110124 |
institution |
Universiti Malaya |
building |
UM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Malaya |
content_source |
UM Research Repository |
url_provider |
http://eprints.um.edu.my/ |
topic |
QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering |
spellingShingle |
QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering Ng, Chun Chet Lin, Che-Tsung Tan, Zhi Qin Wang, Xinyu Kew, Jie Long Chan, Chee Seng Zach, Christopher When IC meets text: Towards a rich annotated integrated circuit text dataset |
description |
Automated Optical Inspection (AOI) is a process that uses cameras to autonomously scan printed circuit boards for quality control. Text is often printed on chip components, and it is crucial that this text is correctly recognized during AOI, as it contains valuable information. In this paper, we introduce ICText, the largest dataset for text detection and recognition on integrated circuits. Uniquely, it includes labels for character quality attributes such as low contrast, blurry, and broken. While loss-reweighting and Curriculum Learning (CL) have been proposed to improve object detector performance by balancing positive and negative samples and gradually training the model from easy to hard samples, these methods have had limited success with one-stage object detectors commonly used in industry. To address this, we propose Attribute-Guided Curriculum Learning (AGCL), which leverages the labeled character quality attributes in ICText. Our extensive experiments demonstrate that AGCL can be applied to different detectors in a plug-and-play fashion to achieve higher Average Precision (AP), significantly outperforming existing methods on ICText without any additional computational overhead during inference. Furthermore, we show that AGCL is also effective on the generic object detection dataset Pascal VOC. Our code and dataset will be publicly available at https://github.com/chunchet-ng/ICText-AGCL. |
format |
Article |
author |
Ng, Chun Chet Lin, Che-Tsung Tan, Zhi Qin Wang, Xinyu Kew, Jie Long Chan, Chee Seng Zach, Christopher |
author_facet |
Ng, Chun Chet Lin, Che-Tsung Tan, Zhi Qin Wang, Xinyu Kew, Jie Long Chan, Chee Seng Zach, Christopher |
author_sort |
Ng, Chun Chet |
title |
When IC meets text: Towards a rich annotated integrated circuit text dataset |
title_short |
When IC meets text: Towards a rich annotated integrated circuit text dataset |
title_full |
When IC meets text: Towards a rich annotated integrated circuit text dataset |
title_fullStr |
When IC meets text: Towards a rich annotated integrated circuit text dataset |
title_full_unstemmed |
When IC meets text: Towards a rich annotated integrated circuit text dataset |
title_sort |
when ic meets text: towards a rich annotated integrated circuit text dataset |
publisher |
Elsevier |
publishDate |
2024 |
url |
http://eprints.um.edu.my/44285/ https://doi.org/10.1016/j.patcog.2023.110124 |
_version_ |
1805881151081938944 |
score |
13.18916 |