On Vision Transformers for Classification Tasks in Side-Scan Sonar Imagery

BW Sheffield, Jeffrey Ellen, Ben Whitmore
{"title":"On Vision Transformers for Classification Tasks in Side-Scan Sonar Imagery","authors":"BW Sheffield, Jeffrey Ellen, Ben Whitmore","doi":"arxiv-2409.12026","DOIUrl":null,"url":null,"abstract":"Side-scan sonar (SSS) imagery presents unique challenges in the\nclassification of man-made objects on the seafloor due to the complex and\nvaried underwater environments. Historically, experts have manually interpreted\nSSS images, relying on conventional machine learning techniques with\nhand-crafted features. While Convolutional Neural Networks (CNNs) significantly\nadvanced automated classification in this domain, they often fall short when\ndealing with diverse seafloor textures, such as rocky or ripple sand bottoms,\nwhere false positive rates may increase. Recently, Vision Transformers (ViTs)\nhave shown potential in addressing these limitations by utilizing a\nself-attention mechanism to capture global information in image patches,\noffering more flexibility in processing spatial hierarchies. This paper\nrigorously compares the performance of ViT models alongside commonly used CNN\narchitectures, such as ResNet and ConvNext, for binary classification tasks in\nSSS imagery. The dataset encompasses diverse geographical seafloor types and is\nbalanced between the presence and absence of man-made objects. ViT-based models\nexhibit superior classification performance across f1-score, precision, recall,\nand accuracy metrics, although at the cost of greater computational resources.\nCNNs, with their inductive biases, demonstrate better computational efficiency,\nmaking them suitable for deployment in resource-constrained environments like\nunderwater vehicles. Future research directions include exploring\nself-supervised learning for ViTs and multi-modal fusion to further enhance\nperformance in challenging underwater environments.","PeriodicalId":501130,"journal":{"name":"arXiv - CS - Computer Vision and Pattern Recognition","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Computer Vision and Pattern Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.12026","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Side-scan sonar (SSS) imagery presents unique challenges in the classification of man-made objects on the seafloor due to the complex and varied underwater environments. Historically, experts have manually interpreted SSS images, relying on conventional machine learning techniques with hand-crafted features. While Convolutional Neural Networks (CNNs) significantly advanced automated classification in this domain, they often fall short when dealing with diverse seafloor textures, such as rocky or ripple sand bottoms, where false positive rates may increase. Recently, Vision Transformers (ViTs) have shown potential in addressing these limitations by utilizing a self-attention mechanism to capture global information in image patches, offering more flexibility in processing spatial hierarchies. This paper rigorously compares the performance of ViT models alongside commonly used CNN architectures, such as ResNet and ConvNext, for binary classification tasks in SSS imagery. The dataset encompasses diverse geographical seafloor types and is balanced between the presence and absence of man-made objects. ViT-based models exhibit superior classification performance across f1-score, precision, recall, and accuracy metrics, although at the cost of greater computational resources. CNNs, with their inductive biases, demonstrate better computational efficiency, making them suitable for deployment in resource-constrained environments like underwater vehicles. Future research directions include exploring self-supervised learning for ViTs and multi-modal fusion to further enhance performance in challenging underwater environments.
关于侧扫声纳图像分类任务中的视觉变换器
由于水下环境复杂多变,侧扫声纳(SSS)图像为海底人造物体的分类带来了独特的挑战。一直以来,专家们都是依靠传统的机器学习技术和人工创建的特征来人工解读 SSS 图像。虽然卷积神经网络(CNN)在这一领域大大推进了自动分类的发展,但在处理岩石或波纹沙底等多种海底纹理时,它们往往会出现不足,因为在这些海底纹理中,假阳性率可能会增加。最近,视觉变换器(ViTs)利用自身关注机制捕捉图像斑块中的全局信息,在处理空间层次方面提供了更大的灵活性,从而显示出解决这些局限性的潜力。本文主要比较了 ViT 模型与常用 CNN 体系结构(如 ResNet 和 ConvNext)在 SSS 图像二元分类任务中的性能。数据集涵盖了不同的海底地理类型,并在存在和不存在人造物体之间进行了平衡。基于 ViT 的模型在 f1 分数、精确度、召回率和准确度指标上都表现出卓越的分类性能,但代价是需要耗费更多的计算资源。未来的研究方向包括探索 ViT 的自我监督学习和多模态融合,以进一步提高在具有挑战性的水下环境中的性能。
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
自引率
0.00%
发文量
0
×
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