Image classification with various deep learning architectures
The goal of the image classification is to correctly predict the subject of an image. For this project, because of the restrictions on resources and time, we worked by using a smaller dataset called Tiny ImageNet, then attempted to train an image classifier using this data. This project implemen...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/76031 |
| _version_ | 1826121117011542016 |
|---|---|
| author | Liu, Hexin |
| author2 | Ponnuthurai N. Suganthan |
| author_facet | Ponnuthurai N. Suganthan Liu, Hexin |
| author_sort | Liu, Hexin |
| collection | NTU |
| description | The goal of the image classification is to correctly predict the subject of an image.
For this project, because of the restrictions on resources and time, we worked by
using a smaller dataset called Tiny ImageNet, then attempted to train an image
classifier using this data.
This project implemented some famous Convolutional Neural Networks with various
useful techniques. The deep learning architectures we implemented in this project
include AlexNet, GoogLeNet, ResNet and DenseNet and their several different
versions, the techniques we applied in this project include dropout, data
augmentation, weight decay and snapshot ensembles and cyclic learning rates.
Consequently, we compared the performance of them in image classification to get
the best one with the highest accuracy. |
| first_indexed | 2024-10-01T05:27:29Z |
| format | Thesis |
| id | ntu-10356/76031 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:27:29Z |
| publishDate | 2018 |
| record_format | dspace |
| spelling | ntu-10356/760312023-07-04T15:56:25Z Image classification with various deep learning architectures Liu, Hexin Ponnuthurai N. Suganthan School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering The goal of the image classification is to correctly predict the subject of an image. For this project, because of the restrictions on resources and time, we worked by using a smaller dataset called Tiny ImageNet, then attempted to train an image classifier using this data. This project implemented some famous Convolutional Neural Networks with various useful techniques. The deep learning architectures we implemented in this project include AlexNet, GoogLeNet, ResNet and DenseNet and their several different versions, the techniques we applied in this project include dropout, data augmentation, weight decay and snapshot ensembles and cyclic learning rates. Consequently, we compared the performance of them in image classification to get the best one with the highest accuracy. Master of Science (Computer Control and Automation) 2018-09-24T02:33:50Z 2018-09-24T02:33:50Z 2018 Thesis http://hdl.handle.net/10356/76031 en 70 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Electrical and electronic engineering Liu, Hexin Image classification with various deep learning architectures |
| title | Image classification with various deep learning architectures |
| title_full | Image classification with various deep learning architectures |
| title_fullStr | Image classification with various deep learning architectures |
| title_full_unstemmed | Image classification with various deep learning architectures |
| title_short | Image classification with various deep learning architectures |
| title_sort | image classification with various deep learning architectures |
| topic | DRNTU::Engineering::Electrical and electronic engineering |
| url | http://hdl.handle.net/10356/76031 |
| work_keys_str_mv | AT liuhexin imageclassificationwithvariousdeeplearningarchitectures |