Structural edge detection of photographic images of rooms with machine learning

In the task to reconstruct 3D models of room architecture from photographic images, identifying the relevant structural edges of the room amidst the noise has been a tremendous challenge. This Final Year Project sets out to determine if machine learning can be a viable alternative to classical edge-...

Full description

Saved in:
Bibliographic Details
Main Author: Yao, Xin Meng
Other Authors: Lee Yong Tsui
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2020
Subjects:
Online Access:https://hdl.handle.net/10356/139122
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-139122
record_format dspace
spelling sg-ntu-dr.10356-1391222023-03-04T20:00:16Z Structural edge detection of photographic images of rooms with machine learning Yao, Xin Meng Lee Yong Tsui School of Mechanical and Aerospace Engineering mytlee@ntu.edu.sg Engineering::Mechanical engineering In the task to reconstruct 3D models of room architecture from photographic images, identifying the relevant structural edges of the room amidst the noise has been a tremendous challenge. This Final Year Project sets out to determine if machine learning can be a viable alternative to classical edge-detection algorithms and, if so, determine the machine learning model that has the best performance. The methodology for this project involves four main parts – generating a labelled dataset, augmenting the labelled data to enlarge the dataset, processing the dataset, and training various models on the dataset. For this project, training is carried out on four different Fully Convolutional Network (FCN) architectures, namely SegNet, U-Net, DenseNet and a pre-trained FCN-RESNET101. For each model, the input is an RGB image and the output is a greyscale image with each pixel indicating its probability of not laying on a relevant edge. The results obtained from the training indicated that the model based on U-Net had the best performance out of the four. Using this finding, further finetuning of the U-Net model’s parameters and hyperparameters are performed to further enhance its performance. Post-processing such as edge-thinning and feature extraction is applied to the output of the final model to obtain the line equation of every predicted edge. The results obtained showed strong promise in discerning structural edges, thus validating the initial hypothesis that machine learning is a viable alternative to classical algorithms. Future works include further enhancing the current model’s accuracy and creating an algorithm to construct a wireframe model from the line equations. Bachelor of Engineering (Mechanical Engineering) 2020-05-15T08:59:21Z 2020-05-15T08:59:21Z 2020 Final Year Project (FYP) https://hdl.handle.net/10356/139122 en 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::Mechanical engineering
spellingShingle Engineering::Mechanical engineering
Yao, Xin Meng
Structural edge detection of photographic images of rooms with machine learning
description In the task to reconstruct 3D models of room architecture from photographic images, identifying the relevant structural edges of the room amidst the noise has been a tremendous challenge. This Final Year Project sets out to determine if machine learning can be a viable alternative to classical edge-detection algorithms and, if so, determine the machine learning model that has the best performance. The methodology for this project involves four main parts – generating a labelled dataset, augmenting the labelled data to enlarge the dataset, processing the dataset, and training various models on the dataset. For this project, training is carried out on four different Fully Convolutional Network (FCN) architectures, namely SegNet, U-Net, DenseNet and a pre-trained FCN-RESNET101. For each model, the input is an RGB image and the output is a greyscale image with each pixel indicating its probability of not laying on a relevant edge. The results obtained from the training indicated that the model based on U-Net had the best performance out of the four. Using this finding, further finetuning of the U-Net model’s parameters and hyperparameters are performed to further enhance its performance. Post-processing such as edge-thinning and feature extraction is applied to the output of the final model to obtain the line equation of every predicted edge. The results obtained showed strong promise in discerning structural edges, thus validating the initial hypothesis that machine learning is a viable alternative to classical algorithms. Future works include further enhancing the current model’s accuracy and creating an algorithm to construct a wireframe model from the line equations.
author2 Lee Yong Tsui
author_facet Lee Yong Tsui
Yao, Xin Meng
format Final Year Project
author Yao, Xin Meng
author_sort Yao, Xin Meng
title Structural edge detection of photographic images of rooms with machine learning
title_short Structural edge detection of photographic images of rooms with machine learning
title_full Structural edge detection of photographic images of rooms with machine learning
title_fullStr Structural edge detection of photographic images of rooms with machine learning
title_full_unstemmed Structural edge detection of photographic images of rooms with machine learning
title_sort structural edge detection of photographic images of rooms with machine learning
publisher Nanyang Technological University
publishDate 2020
url https://hdl.handle.net/10356/139122
_version_ 1759858170777305088