Affordable Access

Access to the full text

Transfer Learning Methods as a New Approach in Computer Vision Tasks with Small Datasets

Authors
  • Brodzicki, Andrzej1
  • Piekarski, Michal1, 2
  • Kucharski, Dariusz1
  • Jaworek-Korjakowska, Joanna1
  • Gorgon, Marek1
  • 1 AGH University of Science and Technology, Poland , (Poland)
  • 2 Jagiellonian University, Poland , (Poland)
Type
Published Article
Journal
Foundations of Computing and Decision Sciences
Publisher
De Gruyter Open
Publication Date
Sep 01, 2020
Volume
45
Issue
3
Pages
179–193
Identifiers
DOI: 10.2478/fcds-2020-0010
Source
De Gruyter
Keywords
License
Green

Abstract

Deep learning methods, used in machine vision challenges, often face the problem of the amount and quality of data. To address this issue, we investigate the transfer learning method. In this study, we briefly describe the idea and introduce two main strategies of transfer learning. We also present the widely-used neural network models, that in recent years performed best in ImageNet classification challenges. Furthermore, we shortly describe three different experiments from computer vision field, that confirm the developed algorithms ability to classify images with overall accuracy 87.2-95%. Achieved numbers are state-of-the-art results in melanoma thickness prediction, anomaly detection and Clostridium di cile cytotoxicity classification problems.

Report this publication

Statistics

Seen <100 times