PulseNetOne: Fast unsupervised pruning of convolutional neural networks for remote sensing
| dc.contributor.author | Browne, David | |
| dc.contributor.author | Giering, Michael | |
| dc.contributor.author | Prestwich, Steven D. | |
| dc.contributor.funder | United Technologies Research Center | en |
| dc.contributor.funder | Science Foundation Ireland | en |
| dc.date.accessioned | 2021-02-22T14:39:25Z | |
| dc.date.available | 2021-02-22T14:39:25Z | |
| dc.date.issued | 2020-03-29 | |
| dc.date.updated | 2021-02-22T14:28:33Z | |
| dc.description.abstract | Scene classification is an important aspect of image/video understanding and segmentation. However, remote-sensing scene classification is a challenging image recognition task, partly due to the limited training data, which causes deep-learning Convolutional Neural Networks (CNNs) to overfit. Another difficulty is that images often have very different scales and orientation (viewing angle). Yet another is that the resulting networks may be very large, again making them prone to overfitting and unsuitable for deployment on memory- and energy-limited devices. We propose an efficient deep-learning approach to tackle these problems. We use transfer learning to compensate for the lack of data, and data augmentation to tackle varying scale and orientation. To reduce network size, we use a novel unsupervised learning approach based on k-means clustering, applied to all parts of the network: most network reduction methods use computationally expensive supervised learning methods, and apply only to the convolutional or fully connected layers, but not both. In experiments, we set new standards in classification accuracy on four remote-sensing and two scene-recognition image datasets. | en |
| dc.description.status | Peer reviewed | en |
| dc.description.version | Published Version | en |
| dc.format.mimetype | application/pdf | en |
| dc.identifier.articleid | 1092 | en |
| dc.identifier.citation | Browne, D., Giering, M. and Prestwich, S. (2020) 'PulseNetOne: Fast Unsupervised Pruning of Convolutional Neural Networks for Remote Sensing', Remote Sensing, 12(7), 1092 (23 pp). doi: 10.3390/rs12071092 | en |
| dc.identifier.doi | 10.3390/rs12071092 | en |
| dc.identifier.endpage | 23 | en |
| dc.identifier.issn | 2072-4292 | |
| dc.identifier.issued | 7 | en |
| dc.identifier.journaltitle | Remote Sensing | en |
| dc.identifier.startpage | 1 | en |
| dc.identifier.uri | https://hdl.handle.net/10468/11087 | |
| dc.identifier.volume | 12 | en |
| dc.language.iso | en | en |
| dc.publisher | MDPI | en |
| dc.relation.project | info:eu-repo/grantAgreement/SFI/SFI Research Centres/12/RC/2289/IE/INSIGHT - Irelands Big Data and Analytics Research Centre/ | en |
| dc.relation.uri | https://www.mdpi.com/2072-4292/12/7/1092 | |
| dc.rights | © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). | en |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | en |
| dc.subject | Convolutional Neural Network | en |
| dc.subject | Network compression | en |
| dc.subject | Pre-trained AlexNet | en |
| dc.subject | Pre-trained VGG16 | en |
| dc.subject | Pruning networks | en |
| dc.subject | Remote-sensing image classification | en |
| dc.subject | Transfer learning | en |
| dc.title | PulseNetOne: Fast unsupervised pruning of convolutional neural networks for remote sensing | en |
| dc.type | Article (peer-reviewed) | en |
