Dai, M., Meyers, G.M. orcid.org/0000-0003-4157-3991, Densley Tingley, D.O. orcid.org/0000-0002-2477-7629 et al. (1 more author) (2019) Initial investigations into using an ensemble of deep neural networks for building façade image semantic segmentation. In: Erbertseder, T., Chrysoulakis, N., Zhang, Y. and Baier, F., (eds.) Proceedings of SPIE. Remote Sensing Technologies and Applications in Urban Environments IV, 09-12 Sep 2019, Strasbourg, France. Society of Photo-optical Instrumentation Engineers ISBN 9781510630178
Abstract
Due to now outdated construction technology, houses which have not been retrofitted since construction typically fail to meet modern energy performance levels. However, identifying at a city scale which houses could benefit the most from retrofit solutions is currently a labour intensive process. In this paper, a system that uses a vehicle mounted camera to capture pictures of residential buildings and then performs semantic segmentation to differentiate components of captured buildings is presented. An ensemble of U-Net semantic segmentation models are trained to identify walls, roofs, chimneys, windows and doors from building façade images and differentiate between window and door instances which are partially visible or obscured. Results show that the ensemble of U-Net models achieved high accuracy in identifying walls, roofs and chimneys, moderate accuracy in identifying windows and low accuracy in identifying doors and instances of windows and doors which were partially visible or obscured. When U-Net models were retrained to identify doors or windows, irrespective of partially visible and obscured instances, a significant rise in door and window identification accuracy was observed. It is believed that a larger training dataset would produce significantly improved results across all classes. The results presented here prove the operational feasibility in the first part of a process to combine this model with high-resolution thermography and GPS for automating building retrofitting evaluations.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Editors: |
|
Copyright, Publisher and Additional Information: | © 2019 SPIE. Reproduced in accordance with the publisher's self-archiving policy. |
Keywords: | deep learning; image segmentation; building retrofit; environmental modelling; U-Net |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Civil and Structural Engineering (Sheffield) |
Funding Information: | Funder Grant number Engineering and Physical Science Research Council EP/R013411/1 |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 28 Apr 2021 11:04 |
Last Modified: | 28 Apr 2021 11:04 |
Status: | Published |
Publisher: | Society of Photo-optical Instrumentation Engineers |
Refereed: | Yes |
Identification Number: | 10.1117/12.2532828 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:173501 |