Fast convolutional neural network for image classification
Convolutional Neural Network(CNN) has proven its excellence in various classification tasks over the recent years. However, the major drawback of deep CNN was its days of computation time to train on large dataset with thousands of classes. The main cause for this slow computation is m...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/72105 |
| _version_ | 1826110077332881408 |
|---|---|
| author | Jeon, Young Seok |
| author2 | Chan Chok You, John |
| author_facet | Chan Chok You, John Jeon, Young Seok |
| author_sort | Jeon, Young Seok |
| collection | NTU |
| description | Convolutional Neural Network(CNN) has proven its excellence in various classification tasks over the recent years. However, the major drawback of deep CNN was its days of computation time to train on large dataset with thousands of classes. The main cause for this slow computation is mostly due to the convolutional layers which performs 2D convolutions with lots of for-loops. This paper thus seeks for faster convolution algorithms such as FFT, Winograd and im2col convolutions to minimize the overall computation time and also introduces CNN written in Matlab for image classification tasks. Performance of the fast CNN algorithm written in Matlab is tested on MNIST and CIFAR-10 dataset with CNN architecture with [3*3] filter Convolutional layer sets(Convolutional, pooling and Sigmoid layers) , Fully Connected(FC) layer and followed by a Softmax layer. |
| first_indexed | 2024-10-01T02:28:35Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/72105 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:28:35Z |
| publishDate | 2017 |
| record_format | dspace |
| spelling | ntu-10356/721052023-07-07T16:10:18Z Fast convolutional neural network for image classification Jeon, Young Seok Chan Chok You, John Yap Kim Hui School of Electrical and Electronic Engineering DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision DRNTU::Science::Mathematics::Applied mathematics::Optimization Convolutional Neural Network(CNN) has proven its excellence in various classification tasks over the recent years. However, the major drawback of deep CNN was its days of computation time to train on large dataset with thousands of classes. The main cause for this slow computation is mostly due to the convolutional layers which performs 2D convolutions with lots of for-loops. This paper thus seeks for faster convolution algorithms such as FFT, Winograd and im2col convolutions to minimize the overall computation time and also introduces CNN written in Matlab for image classification tasks. Performance of the fast CNN algorithm written in Matlab is tested on MNIST and CIFAR-10 dataset with CNN architecture with [3*3] filter Convolutional layer sets(Convolutional, pooling and Sigmoid layers) , Fully Connected(FC) layer and followed by a Softmax layer. Bachelor of Engineering 2017-05-25T09:19:01Z 2017-05-25T09:19:01Z 2017 Final Year Project (FYP) http://hdl.handle.net/10356/72105 en Nanyang Technological University 50 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision DRNTU::Science::Mathematics::Applied mathematics::Optimization Jeon, Young Seok Fast convolutional neural network for image classification |
| title | Fast convolutional neural network for image classification |
| title_full | Fast convolutional neural network for image classification |
| title_fullStr | Fast convolutional neural network for image classification |
| title_full_unstemmed | Fast convolutional neural network for image classification |
| title_short | Fast convolutional neural network for image classification |
| title_sort | fast convolutional neural network for image classification |
| topic | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision DRNTU::Science::Mathematics::Applied mathematics::Optimization |
| url | http://hdl.handle.net/10356/72105 |
| work_keys_str_mv | AT jeonyoungseok fastconvolutionalneuralnetworkforimageclassification |