A master thesis from Aalborg University
Enhancing Convolutional Neural Network Classification Performance through Dataset Enrichment: A Comparative Study
Author(s)
Term
4. term
Education
Publication year
2024
Submitted on
2024-06-07
Pages
21 pages
Abstract
This research is an exploration to compare the effect of enriching a small dataset through different techniques. The techniques considered are duplicating images, adding augmented variations of the original images or generating synthetic images by means of an independent Deep Convolutional Generative Neural Network (DCGAN). Different classifiers were trained on different datasets, enriched through the different techniques. The classifiers trained on enriched datasets, including DCGAN generated datasets, shows promising results and were able to match or exceed the performance of the models trained on the baseline dataset. This suggests that artificially generating additional training data can enhance the classification process. However, further research is required to fully understand the limitations and potentials of these methods in different contexts and with various dataset characteristics.
Keywords
ai ; gan ; cnn ; computer vision ; archaeology
Documents
Colophon: This page is part of the AAU Student Projects portal, which is run by Aalborg University. Here, you can find and download publicly available bachelor's theses and master's projects from across the university dating from 2008 onwards. Student projects from before 2008 are available in printed form at Aalborg University Library.
If you have any questions about AAU Student Projects or the research registration, dissemination and analysis at Aalborg University, please feel free to contact the VBN team. You can also find more information in the AAU Student Projects FAQs.