File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1364/AO.403366
- Scopus: eid_2-s2.0-85094320062
- WOS: WOS:000614630300005
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Digital holographic imaging and classification of microplastics using deep transfer learning
Title | Digital holographic imaging and classification of microplastics using deep transfer learning |
---|---|
Authors | |
Keywords | Classification accuracy Cross entropy Holographic imaging Imbalanced Data-sets Learning network |
Issue Date | 2021 |
Publisher | Optical Society of America. The Journal's web site is located at http://ao.osa.org/journal/ao/about.cfm |
Citation | Applied Optics, 2021, v. 60 n. 4, p. A38-A47 How to Cite? |
Abstract | We devise an inline digital holographic imaging system equipped with a lightweight deep learning network, termed CompNet, and develop the transfer learning for classification and analysis. It has a compression block consisting of a concatenated rectified linear unit (CReLU) activation to reduce the channels, and a class-balanced cross-entropy loss for training. The method is particularly suitable for small and imbalanced datasets, and we apply it to the detection and classification of microplastics. Our results show good improvements both in feature extraction, and generalization and classification accuracy, effectively overcoming the problem of overfitting. This method could be attractive for future in situ microplastic particle detection and classification applications. |
Persistent Identifier | http://hdl.handle.net/10722/290882 |
ISSN | 2010 Impact Factor: 1.707 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | ZHU, Y | - |
dc.contributor.author | Yeung, CH | - |
dc.contributor.author | Lam, EY | - |
dc.date.accessioned | 2020-11-02T05:48:27Z | - |
dc.date.available | 2020-11-02T05:48:27Z | - |
dc.date.issued | 2021 | - |
dc.identifier.citation | Applied Optics, 2021, v. 60 n. 4, p. A38-A47 | - |
dc.identifier.issn | 0003-6935 | - |
dc.identifier.uri | http://hdl.handle.net/10722/290882 | - |
dc.description.abstract | We devise an inline digital holographic imaging system equipped with a lightweight deep learning network, termed CompNet, and develop the transfer learning for classification and analysis. It has a compression block consisting of a concatenated rectified linear unit (CReLU) activation to reduce the channels, and a class-balanced cross-entropy loss for training. The method is particularly suitable for small and imbalanced datasets, and we apply it to the detection and classification of microplastics. Our results show good improvements both in feature extraction, and generalization and classification accuracy, effectively overcoming the problem of overfitting. This method could be attractive for future in situ microplastic particle detection and classification applications. | - |
dc.language | eng | - |
dc.publisher | Optical Society of America. The Journal's web site is located at http://ao.osa.org/journal/ao/about.cfm | - |
dc.relation.ispartof | Applied Optics | - |
dc.rights | Applied Optics. Copyright © Optical Society of America. | - |
dc.rights | © XXXX [year] Optical Society of America]. One print or electronic copy may be made for personal use only. Systematic reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modifications of the content of this paper are prohibited. | - |
dc.subject | Classification accuracy | - |
dc.subject | Cross entropy | - |
dc.subject | Holographic imaging | - |
dc.subject | Imbalanced Data-sets | - |
dc.subject | Learning network | - |
dc.title | Digital holographic imaging and classification of microplastics using deep transfer learning | - |
dc.type | Article | - |
dc.identifier.email | Yeung, CH: chjyeung@hku.hk | - |
dc.identifier.email | Lam, EY: elam@eee.hku.hk | - |
dc.identifier.authority | Yeung, CH=rp02422 | - |
dc.identifier.authority | Lam, EY=rp00131 | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1364/AO.403366 | - |
dc.identifier.scopus | eid_2-s2.0-85094320062 | - |
dc.identifier.hkuros | 318392 | - |
dc.identifier.volume | 60 | - |
dc.identifier.issue | 4 | - |
dc.identifier.spage | A38 | - |
dc.identifier.epage | A47 | - |
dc.identifier.isi | WOS:000614630300005 | - |
dc.publisher.place | United States | - |
dc.identifier.issnl | 0003-6935 | - |