Fish Feeding Intensity Assessment in Aquaculture: A New Audio Dataset AFFIA3K and a Deep Learning Algorithm

Meng Cui; Xubo Liu; Jinzheng Zhao; Jianyuan Sun; Guoping Lian; Tao Chen; Mark D. Plumbley; Daoliang Li; Wenwu Wang

doi:10.1109/MLSP55214.2022.9943405

Back

Fish Feeding Intensity Assessment in Aquaculture: A New Audio Dataset AFFIA3K and a Deep Learning Algorithm

Conference proceeding

Open access

Fish Feeding Intensity Assessment in Aquaculture: A New Audio Dataset AFFIA3K and a Deep Learning Algorithm

Meng Cui, Xubo Liu, Jinzheng Zhao, Jianyuan Sun, Guoping Lian, Tao Chen, Mark D. Plumbley, Daoliang Li and Wenwu Wang

2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing (MLSP 2022), pp.1-6

2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing (MLSP 2022) (Xi’an, China, 22/08/2022–25/08/2022)

02/2023

DOI: https://doi.org/10.1109/MLSP55214.2022.9943405

Abstract

Aquaculture

Audio classification

Classification algorithms

Deep learning

Feature extraction

Fish

Fish feeding intensity assessment

High frequency

Support vector machines

Acoustics

Fish feeding intensity assessment (FFIA) aims to evaluate the change of fish appetite during the feeding process, which is potentially useful in industrial aquaculture. Previous methods are mainly based on computer vision techniques. However, these methods are limited by water refraction and uneven illumination. In this paper, we introduce a new approach for FFIA using audio. We create a new audio dataset for FFIA, namely AFFIA3K, which contains 3000 labelled audio clips of different fish feeding intensity (None, Weak, Medium, Strong). We present a deep learning framework for FFIA, where the audio signal is first transformed into acoustic features, i.e. mel spectrogram, then a convolutional neural network (CNN)-based model is used to classify the fish feeding intensity. Experimental results show that our approach achieves an mAP of 0.74 on the test set of AFFIA3K, and considerably outperforms baseline systems. This indicates the potential of our proposed approach in aquaculture applications.

Files and links (2)

pdf

MLSP_FFIA_2022 (3)2.96 MBDownload View

Author's Accepted Manuscript Open Access

url

https://2022.ieeemlsp.org/View

Event WebsiteConference website

Metrics

297 File views/ downloads

278 Record Views

12 Times Cited - Web of Science

Details

Title: Fish Feeding Intensity Assessment in Aquaculture: A New Audio Dataset AFFIA3K and a Deep Learning Algorithm
Creators: Meng Cui - University of Surrey, School of Computer Science and Electronic Engineering
Xubo Liu - University of Surrey, School of Computer Science and Electronic Engineering
Jinzheng Zhao - University of Surrey, School of Computer Science and Electronic Engineering
Jianyuan Sun - University of Surrey, School of Computer Science and Electronic Engineering
Guoping Lian - University of Surrey, School of Chemistry and Chemical Engineering
Tao Chen - University of Surrey, School of Chemistry and Chemical Engineering
Mark D. Plumbley - University of Surrey, School of Computer Science and Electronic Engineering
Daoliang Li - China Agricultural University
Wenwu Wang - University of Surrey, School of Computer Science and Electronic Engineering
Publication Details: 2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing (MLSP 2022), pp.1-6
Conference: 2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing (MLSP 2022) (Xi’an, China, 22/08/2022–25/08/2022)
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Number of pages: 6
First online publication date: 17/11/2022
Publication Date: 02/2023
Date accepted for publication: 01/06/2022
Grants: AI for Sound, EP/T019751/1, Engineering and Physical Sciences Research Council (United Kingdom, Swindon) - EPSRC
Advancing digital precision aquaculture in China (ADPAC), BB/S020896/1, Biotechnology and Biological Sciences Research Council (United Kingdom, Swindon) - BBSRC
Grant note: This work was supported by the UK-China collaborative research program “Advancing digital precision aquaculture in China (ADPAC)” [UK BBSRC grant BB/S020896/1], and UK EPSRC Grant EP/T019751/1 “AI for Sound”, and a PhD scholarship from the University of Surrey, and a Research Scholarship from the China Scholarship Council. Ethics approval for this study was obtained from the Welfare and Ethical Committee of China Agricultural University (Ref: AW30901202-5-1).
Identifiers: 99666866502346; WOS:000893200300028
Copyright: © 2022 by the Institute of Electrical and Electronics Engineers, Inc. All Rights Reserved. For the purpose of open access, the authors have applied a creative commons attribution (CC BY) licence to any author accepted manuscript version arising
Academic Unit: School of Chemistry and Chemical Engineering; School of Computer Science and Electronic Engineering
Language: English
Resource Type: Conference proceeding

Fish Feeding Intensity Assessment in Aquaculture: A New Audio Dataset AFFIA3K and a Deep Learning Algorithm

Abstract

Files and links (2)

Metrics

Details

Usage Policy