{"title":"Review on Quantitative Methods of Fish School Behaviors","authors":"Yaoguang Wei, Lin Ji, Dong An","doi":"10.1111/raq.70023","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>In aquaculture, the quantitative analysis of fish school behavior refers to the systematic application of mathematical and statistical tools for the precise measurement and description of fish school behavior characteristics through metrics, statistics, and modeling. Compared to studies on individual behavior, the analysis of fish school behavior is crucial for managing fish health and enhancing aquaculture efficiency. Quantitative analysis deepens our understanding of fish school structure and interaction patterns, facilitating the development of more rational and efficient feeding strategies. Traditional manual detection methods are time-consuming, labor-intensive, and have limited accuracy, resulting in inadequate quantitative analysis of fish schools and difficulties in parametrically assessing their behavior and physiological states, which pose challenges to accurate evaluations. However, in recent years, with the emergence of new technologies and quantification indicators, the assessment of fish school behavior has become more accurate and objective. This review summarizes three key technologies for quantitatively analyzing fish school behavior: computer vision, acoustics, and sensors. It outlines three types of quantitative indicators: behavior, biomass estimation, and environment. Furthermore, it provides insights into the response of fish school behavior to four key factors: environmental stress, feeding, disease, and reproduction. The study indicates that comprehensive behavior recognition information often requires selecting suitable technologies or integrating multiple technologies based on the specific needs and conditions of the aquaculture site. Therefore, future research in multimodal data fusion will likely contribute to further advancements in the field of aquaculture.</p>\n </div>","PeriodicalId":227,"journal":{"name":"Reviews in Aquaculture","volume":"17 3","pages":""},"PeriodicalIF":8.8000,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Reviews in Aquaculture","FirstCategoryId":"97","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/raq.70023","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"FISHERIES","Score":null,"Total":0}
引用次数: 0
Abstract
In aquaculture, the quantitative analysis of fish school behavior refers to the systematic application of mathematical and statistical tools for the precise measurement and description of fish school behavior characteristics through metrics, statistics, and modeling. Compared to studies on individual behavior, the analysis of fish school behavior is crucial for managing fish health and enhancing aquaculture efficiency. Quantitative analysis deepens our understanding of fish school structure and interaction patterns, facilitating the development of more rational and efficient feeding strategies. Traditional manual detection methods are time-consuming, labor-intensive, and have limited accuracy, resulting in inadequate quantitative analysis of fish schools and difficulties in parametrically assessing their behavior and physiological states, which pose challenges to accurate evaluations. However, in recent years, with the emergence of new technologies and quantification indicators, the assessment of fish school behavior has become more accurate and objective. This review summarizes three key technologies for quantitatively analyzing fish school behavior: computer vision, acoustics, and sensors. It outlines three types of quantitative indicators: behavior, biomass estimation, and environment. Furthermore, it provides insights into the response of fish school behavior to four key factors: environmental stress, feeding, disease, and reproduction. The study indicates that comprehensive behavior recognition information often requires selecting suitable technologies or integrating multiple technologies based on the specific needs and conditions of the aquaculture site. Therefore, future research in multimodal data fusion will likely contribute to further advancements in the field of aquaculture.
期刊介绍:
Reviews in Aquaculture is a journal that aims to provide a platform for reviews on various aspects of aquaculture science, techniques, policies, and planning. The journal publishes fully peer-reviewed review articles on topics including global, regional, and national production and market trends in aquaculture, advancements in aquaculture practices and technology, interactions between aquaculture and the environment, indigenous and alien species in aquaculture, genetics and its relation to aquaculture, as well as aquaculture product quality and traceability. The journal is indexed and abstracted in several databases including AgBiotech News & Information (CABI), AgBiotechNet, Agricultural Engineering Abstracts, Environment Index (EBSCO Publishing), SCOPUS (Elsevier), and Web of Science (Clarivate Analytics) among others.