Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/22248
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Byerly, A | - |
dc.contributor.author | Kalganova, T | - |
dc.contributor.author | Grichnik, AJ | - |
dc.contributor.editor | Czarnowski, I | - |
dc.contributor.editor | Howlett, RJ | - |
dc.contributor.editor | Jain, LC | - |
dc.date.accessioned | 2021-02-10T14:09:17Z | - |
dc.date.available | 2021-02-10T14:09:17Z | - |
dc.date.issued | 2021-07-08 | - |
dc.identifier | 17 | - |
dc.identifier.citation | Byerly, A., Kalganova, T. and Grichnik A.J. (2021) 'On the Importance of Capturing a Sufficient Diversity of Perspective for the Classification of Micro-PCBs', In: Czarnowski I., Howlett R.J. and Jain L.C. (eds.) Intelligent Decision Technologies: Proceedings of the 13th KES-IDT 2021 Conference, KES Virtual Conference Centre, 14-16 June 2021 (Smart Innovation, Systems and Technologies, vol. 238). Singapore: Springer Nature Singapore, pp. 209-219. doi: 10.1007/978-981-16-2765-1_17. | en_US |
dc.identifier.isbn | 978-981-16-2764-4 | - |
dc.identifier.isbn | 978-981-16-2765-1 | - |
dc.identifier.issn | 2190-3018 | - |
dc.identifier.uri | https://bura.brunel.ac.uk/handle/2438/22248 | - |
dc.description | Pre-print of an original work presented at KES-IDT 2021 held virtually. | - |
dc.description.abstract | We present a dataset consisting of high-resolution images of 13 micro-PCBs captured in various rotations and perspectives relative to the camera, with each sample labeled for PCB type, rotation category, and perspective categories. We then present the design and results of experimentation on combinations of rotations and perspectives used during training and the resulting impact on test accuracy. We then show when and how well data augmentation techniques are capable of simulating rotations vs. perspectives not present in the training data. We perform all experiments using CNNs with and without homogeneous vector capsules (HVCs) and investigate and show the capsules' ability to better encode the equivariance of the sub-components of the micro-PCBs. The results of our experiments lead us to conclude that training a neural network equipped with HVCs, capable of modeling equivariance among sub-components, coupled with training on a diversity of perspectives, achieves the greatest classification accuracy on micro-PCB data. | en_US |
dc.format.extent | 209 - 219 | - |
dc.format.medium | Print-Electronic | - |
dc.language.iso | en_US | en_US |
dc.publisher | Springer Nature Singapore Pte Ltd. | en_US |
dc.relation | arXiv:2101.11164 [cs.CV] | - |
dc.relation.uri | https://arxiv.org/abs/2101.11164v1 | - |
dc.rights | © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021. This is a pre-submission manuscript (preprint), author-produced version of a book chapter submitted for publication in Intelligent Decision Technologies following peer review. The final authenticated version is available online at https://doi.org/10.1007/978-981-16-2765-1_17. | - |
dc.subject | printed circuit boards (PCBs) | en_US |
dc.subject | convolutional neural network (CNN) | en_US |
dc.subject | homogeneous vector capsules (HVCs) | en_US |
dc.subject | capsule | en_US |
dc.subject | data augmentation | en_US |
dc.title | On the Importance of Capturing a Sufficient Diversity of Perspective for the Classification of micro-PCBs | en_US |
dc.type | Article | en_US |
dc.identifier.doi | https://doi.org/10.1007/978-981-16-2765-1_17 | - |
dc.relation.isPartOf | Intelligent Decision Technologies | - |
pubs.notes | 12 pages, 6 figures, 8 tables | - |
dc.identifier.eissn | 2190-3026 | - |
Appears in Collections: | Dept of Electronic and Electrical Engineering Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Pre-print.pdf | 7.22 MB | Adobe PDF | View/Open |
Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.