HNAS: Hyper Neural Architecture Search for Image Segmentation

Yassir Houreh, Mahsa Mahdinejad, Mahsa Mahdinejad, Enrique Naredo, Enrique Naredo, Douglas Dias, Douglas Dias, Douglas Dias, Conor Ryan, Conor Ryan

2021

Abstract

Deep learning is a well suited approach to successfully address image processing and there are several Neural Networks architectures proposed on this research field, one interesting example is the U-net architecture and and its variants. This work proposes to automatically find the best architecture combination from a set of the current most relevant U-net architectures by using a genetic algorithm (GA) applied to solve the Retinal Blood Vessel Segmentation (RVS), which it is relevant to diagnose and cure blindness in diabetes patients. Interestingly, the experimental results show that avoiding human-bias in the design, GA finds novel combinations of U-net architectures, which at first sight seems to be complex but it turns out to be smaller, reaching competitive performance than the manually designed architectures and reducing considerably the computational effort to evolve them.

Paper Citation

in Harvard Style

Houreh Y., Mahdinejad M., Naredo E., Dias D. and Ryan C. (2021). HNAS: Hyper Neural Architecture Search for Image Segmentation.In Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-484-8, pages 246-256. DOI: 10.5220/0010260902460256

in Bibtex Style

@conference{icaart21,
author={Yassir Houreh and Mahsa Mahdinejad and Enrique Naredo and Douglas Dias and Conor Ryan},
title={HNAS: Hyper Neural Architecture Search for Image Segmentation},
booktitle={Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2021},
pages={246-256},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010260902460256},
isbn={978-989-758-484-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - HNAS: Hyper Neural Architecture Search for Image Segmentation
SN - 978-989-758-484-8
AU - Houreh Y.
AU - Mahdinejad M.
AU - Naredo E.
AU - Dias D.
AU - Ryan C.
PY - 2021
SP - 246
EP - 256
DO - 10.5220/0010260902460256