Improvement of Tensor Representation Label in Image Recognition: Evaluation on Selection, Complexity and Size

Shinji Niihara, Shinji Niihara, Minoru Mori

2024

Abstract

One-hot vectors representing correct/incorrect answer classes as {1/0} are usually used as labels for classification problems in Deep Neural Networks. On the other hand, a method using a tensor consisting of speech spectrograms of class names as labels has been proposed and reported to improve resistance to Adversarial Examples. However, effective representations for tensor-based labels have not been sufficiently studied. In this paper, we evaluate the effects of selections of image, complexity, and tensor size expansion on the tensor representation labels. Evaluation experiments using several databases and DNN models show that higher accuracies and tolerances can be achieved by improving tensor representations.

Paper Citation

in Harvard Style

Niihara S. and Mori M. (2024). Improvement of Tensor Representation Label in Image Recognition: Evaluation on Selection, Complexity and Size. In Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-684-2, SciTePress, pages 232-239. DOI: 10.5220/0012313700003654

in Bibtex Style

@conference{icpram24,
author={Shinji Niihara and Minoru Mori},
title={Improvement of Tensor Representation Label in Image Recognition: Evaluation on Selection, Complexity and Size},
booktitle={Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2024},
pages={232-239},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012313700003654},
isbn={978-989-758-684-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Improvement of Tensor Representation Label in Image Recognition: Evaluation on Selection, Complexity and Size
SN - 978-989-758-684-2
AU - Niihara S.
AU - Mori M.
PY - 2024
SP - 232
EP - 239
DO - 10.5220/0012313700003654
PB - SciTePress