Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Survival Tree01:19

Survival Tree

171
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
 Building a Survival Tree
Constructing a...
171
Aggregates Classification01:29

Aggregates Classification

398
Aggregate classification is generally based on its size, petrographic characteristics, weight, and source. Size classification ranges from coarse to fine aggregates, defined by the size of the particles. Coarse aggregates are particles that do not pass through ASTM sieve No. 4, and aggregates that pass through the sieve are fine aggregates.
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...
398

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same journal

BenchMetrics Prob: benchmarking of probabilistic error/loss performance evaluation instruments for binary classification problems.

International journal of machine learning and cybernetics·2023
Same journal

ICUnet++: an Inception-CBAM network based on Unet++ for MR spine image segmentation.

International journal of machine learning and cybernetics·2023
Same journal

A three-way decisions approach based on double hierarchy linguistic aggregation operators of strict t-norms and t-conorms.

International journal of machine learning and cybernetics·2023
Same journal

RNON: image inpainting via repair network and optimization network.

International journal of machine learning and cybernetics·2023
Same journal

Optimal interventional policy based on discrete-time fuzzy rules equivalent model utilizing with COVID-19 pandemic data.

International journal of machine learning and cybernetics·2023
Same journal

SecureFed: federated learning empowered medical imaging technique to analyze lung abnormalities in chest X-rays.

International journal of machine learning and cybernetics·2023

Related Experiment Video

Updated: Sep 26, 2025

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.4K

Tuning of data augmentation hyperparameters in deep learning to building construction image classification with small

André Luiz C Ottoni1,2, Raphael M de Amorim2, Marcela S Novo3

  • 1Technologic and Exact Center, Federal University of Recôncavo da Bahia, Cruz das Almas, Brazil.

International Journal of Machine Learning and Cybernetics
|April 18, 2022
PubMed
Summary

This study introduces a rigorous method for tuning Data Augmentation hyperparameters in Deep Learning for building construction image classification. Optimized parameters significantly improve Convolutional Neural Networks performance on small datasets, aiding vegetation recognition.

Keywords:
Building construction image classificationConvolutional neural networksData augmentationDeep learningHyperparameter tuning

More Related Videos

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

905

Related Experiment Videos

Last Updated: Sep 26, 2025

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.4K
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

905

Area of Science:

  • Computer Vision
  • Machine Learning
  • Building Science

Background:

  • Deep Learning (DL) methods are crucial for building construction image classification.
  • A key challenge is the adoption of Convolutional Neural Networks (CNNs) with limited datasets.
  • Vegetation recognition in facades and roofs analysis requires robust image classification techniques.

Purpose of the Study:

  • To propose a rigorous methodology for tuning Data Augmentation hyperparameters in DL for building construction image classification.
  • To specifically address vegetation recognition in facades and roofs structure analysis.
  • To enhance CNN performance on small datasets within this domain.

Main Methods:

  • A systematic approach to tune Data Augmentation hyperparameters was developed.
  • Logistic Regression models were employed to analyze CNN performance across 128 image transformation combinations.
  • Experiments utilized three established DL architectures implemented with the Keras library.

Main Results:

  • The optimal configuration (Height Shift Range=0.2, Width Shift Range=0.2, Zoom Range=0.2) achieved high accuracy in the first case study.
  • The proposed hyperparameter tuning method yielded the best test results for the second case study as well.
  • The methodology demonstrated effectiveness in improving CNN performance for specific building construction image classification tasks.

Conclusions:

  • The proposed rigorous methodology for Data Augmentation hyperparameter tuning is effective for DL in building construction image classification.
  • Optimized hyperparameters significantly enhance CNN performance, particularly for vegetation recognition tasks on small datasets.
  • This approach provides a valuable strategy for improving the reliability and accuracy of AI in analyzing building structures.