Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques

Underwater video surveys play a significant role in marine benthic research. Usually, surveys are filmed in transects, which are stitched into 2D mosaic maps for further analysis. Due to the massive amount of video data and time-consuming analysis, the need for automatic image segmentation and quant...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Kazimieras Buškus, Evaldas Vaičiukynas, Antanas Verikas, Saulė Medelytė, Andrius Šiaulys, Aleksej Šaškov
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	underwater imagery Ophiuroidea deep learning semantic segmentation Chemical technology TP1-1185
Acceso en línea:	https://doaj.org/article/35ef8684b6664060b4d7e550b3d47c8a
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:35ef8684b6664060b4d7e550b3d47c8a
record_format	dspace
spelling	oai:doaj.org-article:35ef8684b6664060b4d7e550b3d47c8a2021-11-25T18:57:44ZAutomated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques10.3390/s212275981424-8220https://doaj.org/article/35ef8684b6664060b4d7e550b3d47c8a2021-11-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/22/7598https://doaj.org/toc/1424-8220Underwater video surveys play a significant role in marine benthic research. Usually, surveys are filmed in transects, which are stitched into 2D mosaic maps for further analysis. Due to the massive amount of video data and time-consuming analysis, the need for automatic image segmentation and quantitative evaluation arises. This paper investigates such techniques on annotated mosaic maps containing hundreds of instances of brittle stars. By harnessing a deep convolutional neural network with pre-trained weights and post-processing results with a common blob detection technique, we investigate the effectiveness and potential of such segment-and-count approach by assessing the segmentation and counting success. Discs could be recommended instead of full shape masks for brittle stars due to faster annotation among marker variants tested. Underwater image enhancement techniques could not improve segmentation results noticeably, but some might be useful for augmentation purposes.Kazimieras BuškusEvaldas VaičiukynasAntanas VerikasSaulė MedelytėAndrius ŠiaulysAleksej ŠaškovMDPI AGarticleunderwater imageryOphiuroideadeep learningsemantic segmentationChemical technologyTP1-1185ENSensors, Vol 21, Iss 7598, p 7598 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	underwater imagery Ophiuroidea deep learning semantic segmentation Chemical technology TP1-1185
spellingShingle	underwater imagery Ophiuroidea deep learning semantic segmentation Chemical technology TP1-1185 Kazimieras Buškus Evaldas Vaičiukynas Antanas Verikas Saulė Medelytė Andrius Šiaulys Aleksej Šaškov Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques
description	Underwater video surveys play a significant role in marine benthic research. Usually, surveys are filmed in transects, which are stitched into 2D mosaic maps for further analysis. Due to the massive amount of video data and time-consuming analysis, the need for automatic image segmentation and quantitative evaluation arises. This paper investigates such techniques on annotated mosaic maps containing hundreds of instances of brittle stars. By harnessing a deep convolutional neural network with pre-trained weights and post-processing results with a common blob detection technique, we investigate the effectiveness and potential of such segment-and-count approach by assessing the segmentation and counting success. Discs could be recommended instead of full shape masks for brittle stars due to faster annotation among marker variants tested. Underwater image enhancement techniques could not improve segmentation results noticeably, but some might be useful for augmentation purposes.
format	article
author	Kazimieras Buškus Evaldas Vaičiukynas Antanas Verikas Saulė Medelytė Andrius Šiaulys Aleksej Šaškov
author_facet	Kazimieras Buškus Evaldas Vaičiukynas Antanas Verikas Saulė Medelytė Andrius Šiaulys Aleksej Šaškov
author_sort	Kazimieras Buškus
title	Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques
title_short	Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques
title_full	Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques
title_fullStr	Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques
title_full_unstemmed	Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques
title_sort	automated quantification of brittle stars in seabed imagery using computer vision techniques
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/35ef8684b6664060b4d7e550b3d47c8a
work_keys_str_mv	AT kazimierasbuskus automatedquantificationofbrittlestarsinseabedimageryusingcomputervisiontechniques AT evaldasvaiciukynas automatedquantificationofbrittlestarsinseabedimageryusingcomputervisiontechniques AT antanasverikas automatedquantificationofbrittlestarsinseabedimageryusingcomputervisiontechniques AT saulemedelyte automatedquantificationofbrittlestarsinseabedimageryusingcomputervisiontechniques AT andriussiaulys automatedquantificationofbrittlestarsinseabedimageryusingcomputervisiontechniques AT aleksejsaskov automatedquantificationofbrittlestarsinseabedimageryusingcomputervisiontechniques
_version_	1718410455501766656

Automated Quantification of Brittle Stars in Seabed Imagery Using Computer Vision Techniques

Ejemplares similares