Parallel desires: unifying local and semantic feature representations in marine species images for classification

IF 1.6 4区 地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS
Dhana Lakshmi Manikandan, Sakthivel Murugan Santhanam
{"title":"Parallel desires: unifying local and semantic feature representations in marine species images for classification","authors":"Dhana Lakshmi Manikandan, Sakthivel Murugan Santhanam","doi":"10.1007/s11001-024-09551-6","DOIUrl":null,"url":null,"abstract":"<p>Accurate identification of marine species is essential for ecological monitoring, habitat assessment, biodiversity conservation, and sustainable resource management. To address the challenges associated with diverse and complex marine environments, the paper proposes a integrated model that combines the strengths of a Vision Transformer (ViT) and Transfer Learning (TL). The paper introduces a novel methodology for the classification of marine species images by integrating the capabilities of a Amended Dual Attention oN Self-locale and External (ADANSE) Vision Transformer and a DenseNet-169 Transfer Learning model. The ADANSE-ViT, serving as the foundational architecture, excels in capturing long-range dependencies and intricate patterns in large-scale images, forming a robust basis for subsequent classification tasks. On Fine-tuning further, it customizes the model for marine species images. Additionally, we utilize transfer learning with the DenseNet-169 architecture, pre-trained on a comprehensive dataset, to extract relevant features and enhance classification effectiveness specifically for marine species. This synergistic combination enables a comprehensive analysis of both local and semantic features in species images, leading to accurate classification results. Experimental evaluations conducted on self-collected and benchmark datasets showcase the efficacy of our approach, surpassing existing fish classifiers and TL variants in terms of classification accuracy. Our integrated model achieves an impressive accuracy of 96.21% for the self-collected dataset and 95.09% for the benchmarked dataset.</p>","PeriodicalId":49882,"journal":{"name":"Marine Geophysical Research","volume":null,"pages":null},"PeriodicalIF":1.6000,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Marine Geophysical Research","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s11001-024-09551-6","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"GEOCHEMISTRY & GEOPHYSICS","Score":null,"Total":0}
引用次数: 0

Abstract

Accurate identification of marine species is essential for ecological monitoring, habitat assessment, biodiversity conservation, and sustainable resource management. To address the challenges associated with diverse and complex marine environments, the paper proposes a integrated model that combines the strengths of a Vision Transformer (ViT) and Transfer Learning (TL). The paper introduces a novel methodology for the classification of marine species images by integrating the capabilities of a Amended Dual Attention oN Self-locale and External (ADANSE) Vision Transformer and a DenseNet-169 Transfer Learning model. The ADANSE-ViT, serving as the foundational architecture, excels in capturing long-range dependencies and intricate patterns in large-scale images, forming a robust basis for subsequent classification tasks. On Fine-tuning further, it customizes the model for marine species images. Additionally, we utilize transfer learning with the DenseNet-169 architecture, pre-trained on a comprehensive dataset, to extract relevant features and enhance classification effectiveness specifically for marine species. This synergistic combination enables a comprehensive analysis of both local and semantic features in species images, leading to accurate classification results. Experimental evaluations conducted on self-collected and benchmark datasets showcase the efficacy of our approach, surpassing existing fish classifiers and TL variants in terms of classification accuracy. Our integrated model achieves an impressive accuracy of 96.21% for the self-collected dataset and 95.09% for the benchmarked dataset.

Abstract Image

并行欲望:统一海洋物种图像中的局部和语义特征表征以进行分类
准确识别海洋物种对于生态监测、栖息地评估、生物多样性保护和可持续资源管理至关重要。为了应对复杂多样的海洋环境所带来的挑战,本文提出了一种结合视觉转换器(ViT)和迁移学习(TL)优势的综合模型。本文介绍了一种新颖的海洋物种图像分类方法,它整合了 "自我定位与外部双重关注(ADANSE)视觉转换器 "和 "DenseNet-169 转移学习模型 "的功能。作为基础架构的 ADANSE-ViT 擅长捕捉大规模图像中的长距离依赖关系和复杂模式,为后续分类任务奠定了坚实的基础。在进一步微调时,它为海洋物种图像定制了模型。此外,我们还利用在综合数据集上预先训练过的 DenseNet-169 体系结构进行迁移学习,以提取相关特征并提高专门针对海洋物种的分类效果。这种协同组合能够全面分析物种图像中的局部特征和语义特征,从而获得准确的分类结果。在自采数据集和基准数据集上进行的实验评估表明,我们的方法非常有效,在分类准确率方面超过了现有的鱼类分类器和 TL 变体。我们的集成模型在自收集数据集和基准数据集上分别达到了 96.21% 和 95.09% 的准确率,令人印象深刻。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Marine Geophysical Research
Marine Geophysical Research 地学-地球化学与地球物理
CiteScore
2.80
自引率
14.30%
发文量
41
审稿时长
>12 weeks
期刊介绍: Well-established international journal presenting marine geophysical experiments on the geology of continental margins, deep ocean basins and the global mid-ocean ridge system. The journal publishes the state-of-the-art in marine geophysical research including innovative geophysical data analysis, new deep sea floor imaging techniques and tools for measuring rock and sediment properties. Marine Geophysical Research reaches a large and growing community of readers worldwide. Rooted on early international interests in researching the global mid-ocean ridge system, its focus has expanded to include studies of continental margin tectonics, sediment deposition processes and resulting geohazards as well as their structure and stratigraphic record. The editors of MGR predict a rising rate of advances and development in this sphere in coming years, reflecting the diversity and complexity of marine geological processes.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信