File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A defect detection method for topological phononic materials based on few-shot learning

TitleA defect detection method for topological phononic materials based on few-shot learning
Authors
KeywordsDeep learning
Defect detection
Few-shot learning
Structure classification
Topological materials
Issue Date2022
Citation
New Journal of Physics, 2022, v. 24, n. 8, article no. 083012 How to Cite?
AbstractTopological phononic materials have been widely used in many fields, such as topological antennas, asymmetric waveguides, and noise insulation. However, due to the limitations of the manufacturing process, topological protection is vulnerable to some severe defects that may affect the application effect. Therefore, the quality inspection of topological materials is essential to ensure reliable results. Due to the low contrast and irregularity of defects and the similarity of topological phononics, they are difficult to recognize by traditional image processing algorithms, so manual detection is still mainstream at present. But manual detection requires experienced inspectors, which is expensive and time-consuming. In addition, topological materials are expensive to produce, and there is no large publicly available dataset, but deep learning usually relies on large datasets for training. To solve the above problems, we propose an automatic deep learning topology structure defect detection method (ADLTSDM) in this work, which could classify not only the structure of topological materials but also detect the defects of topological phononics based on a small dataset. ADLTSDM exploits the prior knowledge of the topological material structure and achieves an augmentation factor of more than 100 times through the random and fixed interval screenshot algorithm, thus enabling the training of deep neural networks with only two raw data. For defect detection, ADLTSDM has an accuracy of more than 97% and improves detection speed by more than 38% compared with manual detection. For structure classification, ADLTSDM can achieve an accuracy of over 99% and seven times faster speed compared with manual classification. Besides, the detection standard of ADLTSDM is unified, so the accuracy will not be affected by the experience of the inspectors, which has more potential in high-throughput industrial applications.
Persistent Identifierhttp://hdl.handle.net/10722/329874
ISSN
2023 Impact Factor: 2.8
2023 SCImago Journal Rankings: 1.090
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZhang, Beini-
dc.contributor.authorLuo, Xiao-
dc.contributor.authorLyu, Yetao-
dc.contributor.authorWu, Xiaoxiao-
dc.contributor.authorWen, Weijia-
dc.date.accessioned2023-08-09T03:35:58Z-
dc.date.available2023-08-09T03:35:58Z-
dc.date.issued2022-
dc.identifier.citationNew Journal of Physics, 2022, v. 24, n. 8, article no. 083012-
dc.identifier.issn1367-2630-
dc.identifier.urihttp://hdl.handle.net/10722/329874-
dc.description.abstractTopological phononic materials have been widely used in many fields, such as topological antennas, asymmetric waveguides, and noise insulation. However, due to the limitations of the manufacturing process, topological protection is vulnerable to some severe defects that may affect the application effect. Therefore, the quality inspection of topological materials is essential to ensure reliable results. Due to the low contrast and irregularity of defects and the similarity of topological phononics, they are difficult to recognize by traditional image processing algorithms, so manual detection is still mainstream at present. But manual detection requires experienced inspectors, which is expensive and time-consuming. In addition, topological materials are expensive to produce, and there is no large publicly available dataset, but deep learning usually relies on large datasets for training. To solve the above problems, we propose an automatic deep learning topology structure defect detection method (ADLTSDM) in this work, which could classify not only the structure of topological materials but also detect the defects of topological phononics based on a small dataset. ADLTSDM exploits the prior knowledge of the topological material structure and achieves an augmentation factor of more than 100 times through the random and fixed interval screenshot algorithm, thus enabling the training of deep neural networks with only two raw data. For defect detection, ADLTSDM has an accuracy of more than 97% and improves detection speed by more than 38% compared with manual detection. For structure classification, ADLTSDM can achieve an accuracy of over 99% and seven times faster speed compared with manual classification. Besides, the detection standard of ADLTSDM is unified, so the accuracy will not be affected by the experience of the inspectors, which has more potential in high-throughput industrial applications.-
dc.languageeng-
dc.relation.ispartofNew Journal of Physics-
dc.subjectDeep learning-
dc.subjectDefect detection-
dc.subjectFew-shot learning-
dc.subjectStructure classification-
dc.subjectTopological materials-
dc.titleA defect detection method for topological phononic materials based on few-shot learning-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1088/1367-2630/ac8307-
dc.identifier.scopuseid_2-s2.0-85136041259-
dc.identifier.volume24-
dc.identifier.issue8-
dc.identifier.spagearticle no. 083012-
dc.identifier.epagearticle no. 083012-
dc.identifier.isiWOS:000839216400001-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats