Acoustic classification in multifrequency echosounder data using deep convolutional neural networks

Brautaset Olav; Waldeland Anders Ueland; Johnsen Espen; Malde Ketil; Eikvil Line; Salberg Arnt-Borre; Handegard Nils Olav

首页> 外文期刊>ICES Journal of Marine Science >Acoustic classification in multifrequency echosounder data using deep convolutional neural networks

【24h】

Acoustic classification in multifrequency echosounder data using deep convolutional neural networks

机译：使用深卷积神经网络的多频性echosounder数据中的声学分类

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

y Acoustic target classification is the process of assigning observed acoustic backscattering intensity to an acoustic category. A deep learning strategy for acoustic target classification using a convolutional network is developed, consisting of an encoder and a decoder, which allow the network to use pixel information and more abstract features. The network can learn features directly from data, and the learned feature space may include both frequency response and school morphology. We tested the method on multifrequency data collected between 2007 and 2018 during the Norwegian sandeel survey. The network was able to distinguish between sandeel schools, schools of other species, and background pixels (including seabed) in new survey data with an F1 score of 0.87 when tested against manually labelled schools. The network separated schools of sandeel and schools of other species with an F1 score of 0.94. A traditional school classification algorithm obtained substantially lower F1 scores (0.77 and 0.82) when tested against the manually labelled schools. To train the network, it was necessary to develop sampling and preprocessing strategies to account for unbalanced classes, inaccurate annotations, and biases in the training data. This is a step towards a method to be applied across a range of acoustic trawl surveys.

机译：y声学目标分类是将观察到的声学反向散射强度分配给声学类别的过程。开发了一种使用卷积网络的声学目标分类的深度学习策略，由编码器和解码器组成，该解码器允许网络使用像素信息和更抽象的特征。网络可以直接从数据学习特征，并且学习的特征空间可能包括频率响应和学校形态。我们在挪威桑德尔调查期间测试了2007年至2018年期间收集的多频数据的方法。该网络能够区分沙线学校，其他物种学校以及新的调查数据中的背景像素（包括海底），而在手动标记的学校测试时，F1得分为0.87。网络分开的桑德利和其他物种学校的学校，F1得分为0.94。当对手动标记的学校进行测试时，传统的学校分类算法基本上降低F1分数（0.77和0.82）。要培训网络，有必要开发采样和预处理策略，以考虑培训数据中的不平衡类，不准确的注释和偏见。这是朝向一系列声学拖网调查施加的方法的步骤。

著录项

来源
《ICES Journal of Marine Science》 |2020年第4期|1391-1400|共10页
作者
Brautaset Olav; Waldeland Anders Ueland; Johnsen Espen; Malde Ketil; Eikvil Line; Salberg Arnt-Borre; Handegard Nils Olav;
展开▼
作者单位

Norwegian Comp Ctr POB 114 Blindern N-0314 Oslo Norway;

Norwegian Comp Ctr POB 114 Blindern N-0314 Oslo Norway;

Inst Marine Res Nordnesgaten 50 N-5005 Bergen Norway;

Inst Marine Res Nordnesgaten 50 N-5005 Bergen Norway;

Norwegian Comp Ctr POB 114 Blindern N-0314 Oslo Norway;

Norwegian Comp Ctr POB 114 Blindern N-0314 Oslo Norway;

Inst Marine Res Nordnesgaten 50 N-5005 Bergen Norway;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词
acoustic classification; big data; deep learning; machine learning; sandeel;

机译：声学分类;大数据;深度学习;机器学习;沙缘;

相似文献

外文文献
中文文献
专利

1. Application of Deep Convolutional Neural Networks in Attention-Deficit/Hyperactivity Disorder Classification: Data Augmentation and Convolutional Neural Network Transfer Learning [J] . Zhu Li, Chang Weike Journal of Medical Imaging and Health Informatics . 2019,第8期

机译：深度卷积神经网络在注意力缺陷/多动障碍分类中的应用：数据增强与卷积神经网络转移学习
2. Classification of Restlessness Level by Deep Learning of Visual Geometry Group Convolution Neural Network with Acoustic Speech and Visual Face Sensor Data for Smart Care Applications [J] . Ing-Jr Ding, Nai-Wei Zheng Sensors and materials . 2020,第7期

机译：通过声学语音和视觉脸部传感器数据对视觉几何集团卷积神经网络深度学习的躁动水平分类
3. Deep convolutional neural networks for multi-scale time-series classification and application to tokamak disruption prediction using raw, high temporal resolution diagnostic data [J] . Churchill R. M., Tobias B., Zhu Y. Physics of plasmas . 2020,第6期

机译：用于多尺度时间序列分类的深度卷积神经网络与原始，高时分辨率诊断数据的托卡马克中断预测
4. Classification of Finger Tapping Tasks using Convolutional Neural Network Based on Augmented Data with Deep Convolutional Generative Adversarial Network [C] . Seong-Woo Woo, Min-Kyoung Kang, Keum-Shik Hong IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics . 2020

机译：基于深度卷积生成对抗网络的增强数据卷积神经网络的手指敲击任务分类
5. Deep Convolutional Neural Networks for the Classification of the EMBER Malware Dataset [D] . Nallamothu, Anudeep 2018

机译：深度卷积神经网络用于EMBER恶意软件数据集的分类
6. Deep Learning with Convolutional Neural Networks Applied to Electromyography Data: A Resource for the Classification of Movements for Prosthetic Hands [O] . Manfredo Atzori, Matteo Cognolato, Henning Müller 2016

机译：卷积神经网络深度学习应用于肌电图数据：假手运动分类的资源
7. Acoustic classification in multifrequency echosounder data using deep convolutional neural networks [O] . Olav Brautaset, Anders Ueland Waldeland, Espen Johnsen, 2020

机译：使用深卷积神经网络的多频性echosounder数据中的声学分类

Acoustic classification in multifrequency echosounder data using deep convolutional neural networks

摘要

著录项

相似文献

相关主题

期刊订阅