Deep learning for image processing and restoration
Image restoration has always been an ill-posed process due to the information loss. Some degradation can easily be simulated using mathematical formula. This simulation helps training data can be achieved at low cost. However, degradation as shadow is impossible to explicitly simulate by computer pr...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/166645 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-166645 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1666452023-07-07T16:59:59Z Deep learning for image processing and restoration Le, Ky Nam Wen Bihan School of Electrical and Electronic Engineering bihan.wen@ntu.edu.sg Engineering::Electrical and electronic engineering Image restoration has always been an ill-posed process due to the information loss. Some degradation can easily be simulated using mathematical formula. This simulation helps training data can be achieved at low cost. However, degradation as shadow is impossible to explicitly simulate by computer program. This makes dataset in the field of shadow removal become limited. This project aims to solve the shadow removal problem with low-cost dataset using deep learning methods. In this project, MaskshadowGAN unpaired shadow removal model is improved by equipping additional process to the original pipeline. Moreover, a method to attack BDRAR shadow detection model is discovered while experimenting a new unsupervised shadow removal pipeline. Finally, an application is developed for the users to interact with the shadow removal model. Bachelor of Engineering (Electrical and Electronic Engineering) 2023-05-09T01:34:34Z 2023-05-09T01:34:34Z 2023 Final Year Project (FYP) Le, K. N. (2023). Deep learning for image processing and restoration. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/166645 https://hdl.handle.net/10356/166645 en A3248-221 application/pdf Nanyang Technological University |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Electrical and electronic engineering |
spellingShingle |
Engineering::Electrical and electronic engineering Le, Ky Nam Deep learning for image processing and restoration |
description |
Image restoration has always been an ill-posed process due to the information loss. Some degradation can easily be simulated using mathematical formula. This simulation helps training data can be achieved at low cost. However, degradation as shadow is impossible to explicitly simulate by computer program. This makes dataset in the field of shadow removal become limited. This project aims to solve the shadow removal problem with low-cost dataset using deep learning methods. In this project, MaskshadowGAN unpaired shadow removal model is improved by equipping additional process to the original pipeline. Moreover, a method to attack BDRAR shadow detection model is discovered while experimenting a new unsupervised shadow removal pipeline. Finally, an application is developed for the users to interact with the shadow removal model. |
author2 |
Wen Bihan |
author_facet |
Wen Bihan Le, Ky Nam |
format |
Final Year Project |
author |
Le, Ky Nam |
author_sort |
Le, Ky Nam |
title |
Deep learning for image processing and restoration |
title_short |
Deep learning for image processing and restoration |
title_full |
Deep learning for image processing and restoration |
title_fullStr |
Deep learning for image processing and restoration |
title_full_unstemmed |
Deep learning for image processing and restoration |
title_sort |
deep learning for image processing and restoration |
publisher |
Nanyang Technological University |
publishDate |
2023 |
url |
https://hdl.handle.net/10356/166645 |
_version_ |
1772825492948254720 |