Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning

Miguel Chicchon; Hector Bedon

doi:10.1007/978-981-16-4016-2_29

Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning

Miguel Chicchon, Hector Bedon

Research output: Chapter in Book/Report › Conference contribution › peer-review

1 Scopus citations

Abstract

The semantic segmentation approach is essential in automated scene analysis, but its application in underwater environments is still limited. Datasets generally have insufficient labeled data, unbalanced data classes, and different lighting conditions, making it difficult to obtain optimal results. Currently, deep convolutional neural networks allow very good results in machine vision tasks, and one of the network architectures with good performance in semantic segmentation is DeepLabv3 +. This paper evaluates the performance of DeepLabv3 + and transfer learning based on pre-trained backend networks in ImageNet to study underwater scenes. The experimentation is carried out on a dataset available on the Internet with labels of eight classes. Experimental results show that DeepLabv3 + and transfer learning are effective for semantic segmentation of multiple underwater scene objects with insufficient tagged data and unbalanced classes.

Original language	American English
Title of host publication	Smart Trends in Computing and Communications - Proceedings of SmartCom 2021
Editors	Yu-Dong Zhang, Tomonobu Senjyu, Chakchai So-In, Amit Joshi
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	301-309
Number of pages	9
ISBN (Print)	9789811640155
DOIs	https://doi.org/10.1007/978-981-16-4016-2_29
State	Indexed - 2022
Externally published	Yes
Event	5th International Conference on Smart Trends in Computing and Communications, SmartCom 2021 - Virtual, Online Duration: 15 Apr 2021 → 16 Apr 2021

Publication series

Name	Lecture Notes in Networks and Systems
Volume	286
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	5th International Conference on Smart Trends in Computing and Communications, SmartCom 2021
City	Virtual, Online
Period	15/04/21 → 16/04/21

Bibliographical note

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Keywords

Computer vision
Convolutional neural network
Deep learning
DeepLabv3+
Semantic segmentation
Transfer learning
Underwater images

Access to Document

10.1007/978-981-16-4016-2_29

Cite this

Chicchon, M., & Bedon, H. (2022). Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning. In Y.-D. Zhang, T. Senjyu, C. So-In, & A. Joshi (Eds.), Smart Trends in Computing and Communications - Proceedings of SmartCom 2021 (pp. 301-309). (Lecture Notes in Networks and Systems; Vol. 286). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-16-4016-2_29

Chicchon, Miguel ; Bedon, Hector. / Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning. Smart Trends in Computing and Communications - Proceedings of SmartCom 2021. editor / Yu-Dong Zhang ; Tomonobu Senjyu ; Chakchai So-In ; Amit Joshi. Springer Science and Business Media Deutschland GmbH, 2022. pp. 301-309 (Lecture Notes in Networks and Systems).

@inproceedings{7d83fa46eccb4312b60bd6bf71861313,

title = "Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning",

abstract = "The semantic segmentation approach is essential in automated scene analysis, but its application in underwater environments is still limited. Datasets generally have insufficient labeled data, unbalanced data classes, and different lighting conditions, making it difficult to obtain optimal results. Currently, deep convolutional neural networks allow very good results in machine vision tasks, and one of the network architectures with good performance in semantic segmentation is DeepLabv3 +. This paper evaluates the performance of DeepLabv3 + and transfer learning based on pre-trained backend networks in ImageNet to study underwater scenes. The experimentation is carried out on a dataset available on the Internet with labels of eight classes. Experimental results show that DeepLabv3 + and transfer learning are effective for semantic segmentation of multiple underwater scene objects with insufficient tagged data and unbalanced classes.",

keywords = "Computer vision, Convolutional neural network, Deep learning, DeepLabv3+, Semantic segmentation, Transfer learning, Underwater images",

author = "Miguel Chicchon and Hector Bedon",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 5th International Conference on Smart Trends in Computing and Communications, SmartCom 2021 ; Conference date: 15-04-2021 Through 16-04-2021",

year = "2022",

doi = "10.1007/978-981-16-4016-2_29",

language = "American English",

isbn = "9789811640155",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "301--309",

editor = "Yu-Dong Zhang and Tomonobu Senjyu and Chakchai So-In and Amit Joshi",

booktitle = "Smart Trends in Computing and Communications - Proceedings of SmartCom 2021",

address = "Germany",

}

Chicchon, M & Bedon, H 2022, Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning. in Y-D Zhang, T Senjyu, C So-In & A Joshi (eds), Smart Trends in Computing and Communications - Proceedings of SmartCom 2021. Lecture Notes in Networks and Systems, vol. 286, Springer Science and Business Media Deutschland GmbH, pp. 301-309, 5th International Conference on Smart Trends in Computing and Communications, SmartCom 2021, Virtual, Online, 15/04/21. https://doi.org/10.1007/978-981-16-4016-2_29

Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning. / Chicchon, Miguel; Bedon, Hector.
Smart Trends in Computing and Communications - Proceedings of SmartCom 2021. ed. / Yu-Dong Zhang; Tomonobu Senjyu; Chakchai So-In; Amit Joshi. Springer Science and Business Media Deutschland GmbH, 2022. p. 301-309 (Lecture Notes in Networks and Systems; Vol. 286).

Research output: Chapter in Book/Report › Conference contribution › peer-review

TY - GEN

T1 - Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning

AU - Chicchon, Miguel

AU - Bedon, Hector

PY - 2022

Y1 - 2022

N2 - The semantic segmentation approach is essential in automated scene analysis, but its application in underwater environments is still limited. Datasets generally have insufficient labeled data, unbalanced data classes, and different lighting conditions, making it difficult to obtain optimal results. Currently, deep convolutional neural networks allow very good results in machine vision tasks, and one of the network architectures with good performance in semantic segmentation is DeepLabv3 +. This paper evaluates the performance of DeepLabv3 + and transfer learning based on pre-trained backend networks in ImageNet to study underwater scenes. The experimentation is carried out on a dataset available on the Internet with labels of eight classes. Experimental results show that DeepLabv3 + and transfer learning are effective for semantic segmentation of multiple underwater scene objects with insufficient tagged data and unbalanced classes.

AB - The semantic segmentation approach is essential in automated scene analysis, but its application in underwater environments is still limited. Datasets generally have insufficient labeled data, unbalanced data classes, and different lighting conditions, making it difficult to obtain optimal results. Currently, deep convolutional neural networks allow very good results in machine vision tasks, and one of the network architectures with good performance in semantic segmentation is DeepLabv3 +. This paper evaluates the performance of DeepLabv3 + and transfer learning based on pre-trained backend networks in ImageNet to study underwater scenes. The experimentation is carried out on a dataset available on the Internet with labels of eight classes. Experimental results show that DeepLabv3 + and transfer learning are effective for semantic segmentation of multiple underwater scene objects with insufficient tagged data and unbalanced classes.

KW - Computer vision

KW - Convolutional neural network

KW - Deep learning

KW - DeepLabv3+

KW - Semantic segmentation

KW - Transfer learning

KW - Underwater images

UR - http://www.scopus.com/inward/record.url?scp=85118997690&partnerID=8YFLogxK

U2 - 10.1007/978-981-16-4016-2_29

DO - 10.1007/978-981-16-4016-2_29

M3 - Conference contribution

AN - SCOPUS:85118997690

SN - 9789811640155

T3 - Lecture Notes in Networks and Systems

SP - 301

EP - 309

BT - Smart Trends in Computing and Communications - Proceedings of SmartCom 2021

A2 - Zhang, Yu-Dong

A2 - Senjyu, Tomonobu

A2 - So-In, Chakchai

A2 - Joshi, Amit

PB - Springer Science and Business Media Deutschland GmbH

T2 - 5th International Conference on Smart Trends in Computing and Communications, SmartCom 2021

Y2 - 15 April 2021 through 16 April 2021

ER -

Chicchon M, Bedon H. Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning. In Zhang YD, Senjyu T, So-In C, Joshi A, editors, Smart Trends in Computing and Communications - Proceedings of SmartCom 2021. Springer Science and Business Media Deutschland GmbH. 2022. p. 301-309. (Lecture Notes in Networks and Systems). doi: 10.1007/978-981-16-4016-2_29

Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this