Proceedings of the International Conference on Computer, Information Technology and Intelligent Computing (CITIC 2022)

Underwater Image Semantic Segmentation with Weighted Average Ensemble

Authors
Muhammad Hidayat Jauhari1, *, Noramiza Hashim1
1Multimedia University, Cyberjaya, Malaysia
*Corresponding author. Email: muhidayat4800@gmail.com
Corresponding Author
Muhammad Hidayat Jauhari
Available Online 27 December 2022.
DOI
10.2991/978-94-6463-094-7_43How to use a DOI?
Keywords
Underwater Image; Weighted Average Ensemble; Deep Learning; Neural Network; Semantic Segmentation
Abstract

Underwater image segmentation is a method that could help with underwater exploration because it is useful and impactful in the understanding and study of the marine environment. However, it is a difficult and challenging task compared to regular image segmentation due to the nature of the images themselves, which are of lesser quality, as well as the limited availability of publicly accessible datasets. In this work, several deep learning-based approaches were implemented and a solution for underwater image segmentation was proposed. The proposed method was developed using a smaller dataset with low resolution images. The proposed method consisted of several image segmentation deep learning models such as U-Net, LinkNet, and Feature Pyramid Network (FPN) with different encoders specifically Inception-V3 and ResNet34. The weighted average ensemble method was used to combine the results of each models mentioned to identify the optimized combination. The proposed method was then compared with the individual models to provide comparison and benchmark on the ensemble approach to the single model approach. The proposed method achieved an accuracy of 62.48% where it outperformed all individual models. As a result, aggregating expected results from numerous models gives better performance when compared to individual model predictions.

Copyright
© 2022 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

Volume Title
Proceedings of the International Conference on Computer, Information Technology and Intelligent Computing (CITIC 2022)
Series
Atlantis Highlights in Computer Sciences
Publication Date
27 December 2022
ISBN
978-94-6463-094-7
ISSN
2589-4900
DOI
10.2991/978-94-6463-094-7_43How to use a DOI?
Copyright
© 2022 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

TY  - CONF
AU  - Muhammad Hidayat Jauhari
AU  - Noramiza Hashim
PY  - 2022
DA  - 2022/12/27
TI  - Underwater Image Semantic Segmentation with Weighted Average Ensemble
BT  - Proceedings of the International Conference on Computer, Information Technology and Intelligent Computing (CITIC 2022)
PB  - Atlantis Press
SP  - 533
EP  - 543
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-094-7_43
DO  - 10.2991/978-94-6463-094-7_43
ID  - Jauhari2022
ER  -