https://www.mdu.se/

mdu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Deep Learning in Remote Sensing: An Application to Detect Snow and Water in Construction Sites
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems.ORCID iD: 0000-0002-1547-4386
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems.ORCID iD: 0000-0003-3802-4721
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems.ORCID iD: 0000-0002-1212-7637
Mälardalen University.
Show others and affiliations
2021 (English)In: Proceedings - 2021 4th International Conference on Artificial Intelligence for Industries, AI4I 2021, 2021, p. 52-56Conference paper, Published paper (Refereed)
Abstract [en]

It is important for a construction and property development company to know weather conditions in their daily operation. In this paper, a deep learning-based approach is investigated to detect snow and rain conditions in construction sites using drone imagery. A Convolutional Neural Network (CNN) is developed for the feature extraction and performing classification on those features using machine learning (ML) algorithms. Well-known existing deep learning algorithms AlexNet and VGG16 models are also deployed and tested on the dataset. Results show that smaller CNN architecture with three convolutional layers was sufficient at extracting relevant features to the classification task at hand compared to the larger state-of-the-art architectures. The proposed model reached a top accuracy of 97.3% in binary classification and 96.5% while also taking rain conditions into consideration. It was also found that ML algorithms,i.e., support vector machine (SVM), logistic regression and k-nearest neighbors could be used as classifiers using feature maps extracted from CNNs and a top accuracy of 90% was obtained using SVM algorithms.

Place, publisher, year, edition, pages
2021. p. 52-56
Keywords [en]
Support vector machines;Deep learning;Training;Rain;Machine learning algorithms;Snow;Feature extraction;Classification;deep learning;convolutional neural networks
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:mdh:diva-56384DOI: 10.1109/AI4I51902.2021.00021Scopus ID: 2-s2.0-85126069912ISBN: 978-1-6654-3410-2 (electronic)OAI: oai:DiVA.org:mdh-56384DiVA, id: diva2:1610027
Conference
2021 4th International Conference on Artificial Intelligence for Industries (AI4I), 20-22 Sept. 2021
Available from: 2021-11-09 Created: 2021-11-09 Last updated: 2022-03-18Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Rahman, HamidurAhmed, Mobyen UddinBegum, Shahina

Search in DiVA

By author/editor
Rahman, HamidurAhmed, Mobyen UddinBegum, ShahinaFridberg, MatsHoflin, Adam
By organisation
Embedded SystemsMälardalen University
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 37 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf