Weakly supervised underwater fish segmentation using affinity LCFCN

Abstract Estimating fish body measurements like length, width, and mass has received considerable research due to its potential in boosting productivity in marine and aquaculture applications. Some methods are based on manual collection of these measurements using tools like a ruler which is time co...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Issam H. Laradji, Alzayat Saleh, Pau Rodriguez, Derek Nowrouzezahrai, Mostafa Rahimi Azghadi, David Vazquez
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/9de071b756e24d92ac88cac60a6b6108
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9de071b756e24d92ac88cac60a6b6108
record_format	dspace
spelling	oai:doaj.org-article:9de071b756e24d92ac88cac60a6b61082021-12-02T15:28:47ZWeakly supervised underwater fish segmentation using affinity LCFCN10.1038/s41598-021-96610-22045-2322https://doaj.org/article/9de071b756e24d92ac88cac60a6b61082021-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-96610-2https://doaj.org/toc/2045-2322Abstract Estimating fish body measurements like length, width, and mass has received considerable research due to its potential in boosting productivity in marine and aquaculture applications. Some methods are based on manual collection of these measurements using tools like a ruler which is time consuming and labour intensive. Others rely on fully-supervised segmentation models to automatically acquire these measurements but require collecting per-pixel labels which are also time consuming. It can take up to 2 minutes per fish to acquire accurate segmentation labels. To address this problem, we propose a segmentation model that can efficiently train on images labeled with point-level supervision, where each fish is annotated with a single click. This labeling scheme takes an average of only 1 second per fish. Our model uses a fully convolutional neural network with one branch that outputs per-pixel scores and another that outputs an affinity matrix. These two outputs are aggregated using a random walk to get the final, refined per-pixel output. The whole model is trained end-to-end using the localization-based counting fully convolutional neural network (LCFCN) loss and thus we call our method Affinity-LCFCN (A-LCFCN). We conduct experiments on the DeepFish dataset, which contains several fish habitats from north-eastern Australia. The results show that A-LCFCN outperforms a fully-supervised segmentation model when the annotation budget is fixed. They also show that A-LCFCN achieves better segmentation results than LCFCN and a standard baseline.Issam H. LaradjiAlzayat SalehPau RodriguezDerek NowrouzezahraiMostafa Rahimi AzghadiDavid VazquezNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-10 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Issam H. Laradji Alzayat Saleh Pau Rodriguez Derek Nowrouzezahrai Mostafa Rahimi Azghadi David Vazquez Weakly supervised underwater fish segmentation using affinity LCFCN
description	Abstract Estimating fish body measurements like length, width, and mass has received considerable research due to its potential in boosting productivity in marine and aquaculture applications. Some methods are based on manual collection of these measurements using tools like a ruler which is time consuming and labour intensive. Others rely on fully-supervised segmentation models to automatically acquire these measurements but require collecting per-pixel labels which are also time consuming. It can take up to 2 minutes per fish to acquire accurate segmentation labels. To address this problem, we propose a segmentation model that can efficiently train on images labeled with point-level supervision, where each fish is annotated with a single click. This labeling scheme takes an average of only 1 second per fish. Our model uses a fully convolutional neural network with one branch that outputs per-pixel scores and another that outputs an affinity matrix. These two outputs are aggregated using a random walk to get the final, refined per-pixel output. The whole model is trained end-to-end using the localization-based counting fully convolutional neural network (LCFCN) loss and thus we call our method Affinity-LCFCN (A-LCFCN). We conduct experiments on the DeepFish dataset, which contains several fish habitats from north-eastern Australia. The results show that A-LCFCN outperforms a fully-supervised segmentation model when the annotation budget is fixed. They also show that A-LCFCN achieves better segmentation results than LCFCN and a standard baseline.
format	article
author	Issam H. Laradji Alzayat Saleh Pau Rodriguez Derek Nowrouzezahrai Mostafa Rahimi Azghadi David Vazquez
author_facet	Issam H. Laradji Alzayat Saleh Pau Rodriguez Derek Nowrouzezahrai Mostafa Rahimi Azghadi David Vazquez
author_sort	Issam H. Laradji
title	Weakly supervised underwater fish segmentation using affinity LCFCN
title_short	Weakly supervised underwater fish segmentation using affinity LCFCN
title_full	Weakly supervised underwater fish segmentation using affinity LCFCN
title_fullStr	Weakly supervised underwater fish segmentation using affinity LCFCN
title_full_unstemmed	Weakly supervised underwater fish segmentation using affinity LCFCN
title_sort	weakly supervised underwater fish segmentation using affinity lcfcn
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/9de071b756e24d92ac88cac60a6b6108
work_keys_str_mv	AT issamhlaradji weaklysupervisedunderwaterfishsegmentationusingaffinitylcfcn AT alzayatsaleh weaklysupervisedunderwaterfishsegmentationusingaffinitylcfcn AT paurodriguez weaklysupervisedunderwaterfishsegmentationusingaffinitylcfcn AT dereknowrouzezahrai weaklysupervisedunderwaterfishsegmentationusingaffinitylcfcn AT mostafarahimiazghadi weaklysupervisedunderwaterfishsegmentationusingaffinitylcfcn AT davidvazquez weaklysupervisedunderwaterfishsegmentationusingaffinitylcfcn
_version_	1718387220106182656

Weakly supervised underwater fish segmentation using affinity LCFCN

Ejemplares similares