Deep Learning for Handwritten Digits Recognition Using MATLAB Toolbox
| dc.contributor.author | Chen, JiaCong | |
| dc.contributor.supervisor | Lu, Wu-Sheng | |
| dc.date.accessioned | 2019-12-10T18:06:07Z | |
| dc.date.available | 2019-12-10T18:06:07Z | |
| dc.date.copyright | 2019 | en_US |
| dc.date.issued | 2019-12-10 | |
| dc.degree.department | Department of Electrical and Computer Engineering | en_US |
| dc.degree.level | Master of Engineering M.Eng. | en_US |
| dc.description.abstract | In this report, we describe several neural network architectures for the classification of handwritten digits. In particular, our attention is focused on the class of convolutional neural networks (CNNs) for performance superiority. By using MATLAB deep learning toolbox, we provide the implementation details necessary for constructing and applying CNNs to a high-quality data set known as MNIST which collects as many as 60,000 handwritten digits for training and 10,000 digits for testing the CNNs. This report also presents several variants of the original LeNet-5 architecture, which has been known for its excellent performance for classifying handwritten digits, for potential performance improvement. Using the deep learning toolbox, extensive simulation studies are conducted for performance evaluation and comparisons between various neural networks as well as two well-known classifiers that are not based on neural-networks. | en_US |
| dc.description.scholarlevel | Graduate | en_US |
| dc.identifier.uri | http://hdl.handle.net/1828/11353 | |
| dc.language.iso | en | en_US |
| dc.rights | Available to the World Wide Web | en_US |
| dc.subject | Deep Learning | en_US |
| dc.subject | Convolutional Neural Network | en_US |
| dc.subject | MATLAB | en_US |
| dc.title | Deep Learning for Handwritten Digits Recognition Using MATLAB Toolbox | en_US |
| dc.type | project | en_US |