|
Research Article: Artificial intelligence models for identifying several fish species based on otolith morphology index analysis from nearshore areas of Vietnam
|
Q.T. Vu *1   , T.D. Pham2 , V.Q. Nguyen3 |
1- Joint Vietnam-Russia Tropical Science and Technology Research Center , vuquyetthanh@gmail.com
2- Joint Vietnam-Russia Tropical Science and Technology Research Center
3- Vietnam Academy of Science and Technology, Hanoi, Vietnam |
|
|
Abstract: (429 Views) |
| Fish species can be identified based on the analysis of morphological indices including basic dimension parameters and shape index. Several pattern recognition methods have been proposed to classify fish species through the morphological characteristics of otolith outlines. Machine learning methods have been applied in various fields, particularly in the differentiation of object shapes. Applying machine learning models to identify species based on basic dimension parameters and shape index of otoliths is highly promising. The purpose of this study is to apply machine learning models to classify marine fish species, aiming to determine which machine learning model and indices are suitable for otolith shape classification. A total of 720 samples of left otoliths (sagittae) from 12 fish species, with 60 individuals per species, were used to develop and evaluate the identification model using Python language. For the first time, a comparative evaluation of six machine learning models and three deep learning models was conducted to distinguish 12 fish species in the nearshore areas of northern and central Vietnam. The results of this study have identified machine learning and deep learning models based on high-performing basic dimension parameter (BDP) and/or shape index ShI indices for species identification. This lays the groundwork for developing software for automatic species or population identification based on otolith morphological analysis. |
|
| Keywords: BDP, ShI, East Vietnam Sea, Deep Learning, Machine learing, Otolith |
|
|
Full-Text [PDF 1540 kb]
(389 Downloads)
|
Type of Study: Orginal research papers |
Subject:
Ecology
ePublished: 2025/03/17
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
.png)
.png)
.png)
.png)
.png)
