Publication detail

Family Coat of Arms and Armorial Achievement Classification

ŠŮSTEK, M. VÍDEŇSKÝ, F. ZBOŘIL, F. ZBOŘIL, F.

Original Title

Family Coat of Arms and Armorial Achievement Classification

Type

conference paper

Language

English

Original Abstract

This paper presents an approach to classification of family coats of arms and armorial achievement. It is difficult to obtain images with coats of arms because not many of them are publicly available. To the best of our knowledge, there is no dataset. Therefore, we artificially extend our dataset using Neural Style Transfer technique and simple image transformations. We describe our dataset and the division into training and test sets that respects the lack of data examples. We discuss results obtained with both small convolutional neural network (convnet) trained from scratch and modified architectures of various convents pretrained on Imagenet dataset. This paper further focuses on the VGG architecture which produces the best accuracy. We show accuracy progress during training, per-class accuracy and a normalized confusion matrix for VGG16 architecture. We reach top-1 accuracy of nearly 60% and top-5 accuracy of 80%. To the best of our knowledge, this is the first family coats of arms classification work, so we cannot compare our results with others.

Keywords

coats of arms, image classification, convolutional neural network, artificial intelligence, machine learning

Authors

ŠŮSTEK, M.; VÍDEŇSKÝ, F.; ZBOŘIL, F.; ZBOŘIL, F.

Released

17. 4. 2019

Publisher

Springer International Publishing

Location

Los Alamitos

ISBN

2194-5357

Periodical

Advances in Intelligent Systems and Computing

Year of study

941

Number

2

State

Swiss Confederation

Pages from

577

Pages to

586

Pages count

9

URL

BibTex

@inproceedings{BUT156844,
  author="Martin {Šůstek} and František {Vídeňský} and František {Zbořil} and František {Zbořil}",
  title="Family Coat of Arms and Armorial Achievement Classification",
  booktitle="Intelligent Systems Design and Applications",
  year="2019",
  series="Advances in Intelligent Systems and Computing",
  journal="Advances in Intelligent Systems and Computing",
  volume="941",
  number="2",
  pages="577--586",
  publisher="Springer International Publishing",
  address="Los Alamitos",
  doi="10.1007/978-3-030-16660-1\{_}56",
  issn="2194-5357",
  url="https://www.fit.vut.cz/research/publication/11848/"
}