Scene classification based on convolutional neural network

Convolutional Neural Network(CNN) has been widely used in image recognition and classificaiton. The objectives of this project is to implement the mupltiple CNNs on MIT Indoor67 dataset, to evaluate their performance, and therefore gain first hand experience on transfer learning. First, Place205-VG...

Full description

Saved in:
Bibliographic Details
Main Author: Zou, Bojing
Other Authors: Jiang Xudong
Format: Final Year Project
Language:English
Published: 2018
Subjects:
Online Access:http://hdl.handle.net/10356/75435
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-75435
record_format dspace
spelling sg-ntu-dr.10356-754352023-07-07T16:06:42Z Scene classification based on convolutional neural network Zou, Bojing Jiang Xudong School of Electrical and Electronic Engineering DRNTU::Engineering DRNTU::Engineering::Electrical and electronic engineering Convolutional Neural Network(CNN) has been widely used in image recognition and classificaiton. The objectives of this project is to implement the mupltiple CNNs on MIT Indoor67 dataset, to evaluate their performance, and therefore gain first hand experience on transfer learning. First, Place205-VGG CNN model has been used in order to evaluate its performance. Later, after performance evaluation, a relatively new technique CAM has been implemented on CNN, as a result, a heat map will be generated so that human can visualize and indirectly understand the relative importance of feature information learned by CNN. This manoeuvre enables a deeper understanding of CNN’s learning aspect. Second, pre-trained ResNet-152 has been used for fine- tuning, by freezing some low-level layers and training the final classifier, a better classification accuracy is obtained. Bachelor of Engineering 2018-05-31T05:43:08Z 2018-05-31T05:43:08Z 2018 Final Year Project (FYP) http://hdl.handle.net/10356/75435 en Nanyang Technological University 64 p. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic DRNTU::Engineering
DRNTU::Engineering::Electrical and electronic engineering
spellingShingle DRNTU::Engineering
DRNTU::Engineering::Electrical and electronic engineering
Zou, Bojing
Scene classification based on convolutional neural network
description Convolutional Neural Network(CNN) has been widely used in image recognition and classificaiton. The objectives of this project is to implement the mupltiple CNNs on MIT Indoor67 dataset, to evaluate their performance, and therefore gain first hand experience on transfer learning. First, Place205-VGG CNN model has been used in order to evaluate its performance. Later, after performance evaluation, a relatively new technique CAM has been implemented on CNN, as a result, a heat map will be generated so that human can visualize and indirectly understand the relative importance of feature information learned by CNN. This manoeuvre enables a deeper understanding of CNN’s learning aspect. Second, pre-trained ResNet-152 has been used for fine- tuning, by freezing some low-level layers and training the final classifier, a better classification accuracy is obtained.
author2 Jiang Xudong
author_facet Jiang Xudong
Zou, Bojing
format Final Year Project
author Zou, Bojing
author_sort Zou, Bojing
title Scene classification based on convolutional neural network
title_short Scene classification based on convolutional neural network
title_full Scene classification based on convolutional neural network
title_fullStr Scene classification based on convolutional neural network
title_full_unstemmed Scene classification based on convolutional neural network
title_sort scene classification based on convolutional neural network
publishDate 2018
url http://hdl.handle.net/10356/75435
_version_ 1772825877316370432