Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation

Abstract Accurately segmenting foods from optical images is a challenging task, yet becoming possible with the help of recent advances in Deep Learning based solutions. Automated identification of food items opens up possibilities of useful applications like nutrition intake monitoring. Given large...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Mia S. N. Siemon, A. S. M. Shihavuddin, Gitte Ravn-Haren
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/d8ba52c9b7714b63971be8b054b8cae2
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d8ba52c9b7714b63971be8b054b8cae2
record_format	dspace
spelling	oai:doaj.org-article:d8ba52c9b7714b63971be8b054b8cae22021-12-02T14:12:46ZSequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation10.1038/s41598-020-79677-12045-2322https://doaj.org/article/d8ba52c9b7714b63971be8b054b8cae22021-01-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-79677-1https://doaj.org/toc/2045-2322Abstract Accurately segmenting foods from optical images is a challenging task, yet becoming possible with the help of recent advances in Deep Learning based solutions. Automated identification of food items opens up possibilities of useful applications like nutrition intake monitoring. Given large variations in food choices, Deep Learning based solutions still struggle to generate human level accuracy. In this work, we propose a novel Sequential Transfer Learning method using Hierarchical Clustering. This novel approach simulates a step by step problem solving framework based on clustering of similar types of foods. The proposed approach provides up to 6% gain in accuracy compared to traditional network training and generated a robust model performing better in challenging unseen cases. This approach is also tested for segmenting foods in Danish school children meals for dietary intake monitoring as an application.Mia S. N. SiemonA. S. M. ShihavuddinGitte Ravn-HarenNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-14 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Mia S. N. Siemon A. S. M. Shihavuddin Gitte Ravn-Haren Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
description	Abstract Accurately segmenting foods from optical images is a challenging task, yet becoming possible with the help of recent advances in Deep Learning based solutions. Automated identification of food items opens up possibilities of useful applications like nutrition intake monitoring. Given large variations in food choices, Deep Learning based solutions still struggle to generate human level accuracy. In this work, we propose a novel Sequential Transfer Learning method using Hierarchical Clustering. This novel approach simulates a step by step problem solving framework based on clustering of similar types of foods. The proposed approach provides up to 6% gain in accuracy compared to traditional network training and generated a robust model performing better in challenging unseen cases. This approach is also tested for segmenting foods in Danish school children meals for dietary intake monitoring as an application.
format	article
author	Mia S. N. Siemon A. S. M. Shihavuddin Gitte Ravn-Haren
author_facet	Mia S. N. Siemon A. S. M. Shihavuddin Gitte Ravn-Haren
author_sort	Mia S. N. Siemon
title	Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
title_short	Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
title_full	Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
title_fullStr	Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
title_full_unstemmed	Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
title_sort	sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/d8ba52c9b7714b63971be8b054b8cae2
work_keys_str_mv	AT miasnsiemon sequentialtransferlearningbasedonhierarchicalclusteringforimprovedperformanceindeeplearningbasedfoodsegmentation AT asmshihavuddin sequentialtransferlearningbasedonhierarchicalclusteringforimprovedperformanceindeeplearningbasedfoodsegmentation AT gitteravnharen sequentialtransferlearningbasedonhierarchicalclusteringforimprovedperformanceindeeplearningbasedfoodsegmentation
_version_	1718391752693383168

Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation

Ejemplares similares