Duanrui Wang , Meng Wu , Xingyue Zhu , Qiwei Qin , Shaowen Wang , Haibin Ye , Kaiyuan Guo , Chi Wu , Yi Shi
{"title":"Real-time detection and identification of fish skin health in the underwater environment based on improved YOLOv10 model","authors":"Duanrui Wang , Meng Wu , Xingyue Zhu , Qiwei Qin , Shaowen Wang , Haibin Ye , Kaiyuan Guo , Chi Wu , Yi Shi","doi":"10.1016/j.aqrep.2025.102723","DOIUrl":null,"url":null,"abstract":"<div><div>In densely populated aquaculture net cages, real-time detection and identification of fish skin diseases can effectively prevent large-scale outbreaks, thereby reducing fish mortality rates and economic losses. This study proposes an identification model, DCW-YOLO, based on deep learning-driven object recognition technology. By capturing images of fish in a seabed environment, the model can detect fish showing signs of skin diseases. Currently, there is limited research on automatic fish disease recognition specifically for Miichthys miiuy. To address this gap, we introduce, for the first time, a novel dataset for Miichthys miiuy and train, validate, and test the model on this dataset. DCW-YOLO substitutes the CIoU loss function in YOLOv10 with the NWD loss function, thereby improving the model’s ability to detect densely packed targets. The C2f-D-LKA layer is employed in place of the C2f convolutional layer, improving the model’s capacity to capture irregularly shaped and sized objects while effectively reducing computational overhead and parameter load. Additionally, the DySample upsampling structure, which utilizes point sampling, is introduced to increase image resolution without adding significant computational cost. Underwater experimental results show that the mAP50 and precision of DCW-YOLO reach 96.87 % and 95.46 %, respectively, representing improvements of 4.61 % and 3.24 % over the original YOLOv10 model. When deployed in aquaculture settings, this model provides rapid, low-cost real-time disease detection, helping farmers identify diseases early and mitigate potential losses.</div></div>","PeriodicalId":8103,"journal":{"name":"Aquaculture Reports","volume":"42 ","pages":"Article 102723"},"PeriodicalIF":3.2000,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Aquaculture Reports","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2352513425001097","RegionNum":2,"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 densely populated aquaculture net cages, real-time detection and identification of fish skin diseases can effectively prevent large-scale outbreaks, thereby reducing fish mortality rates and economic losses. This study proposes an identification model, DCW-YOLO, based on deep learning-driven object recognition technology. By capturing images of fish in a seabed environment, the model can detect fish showing signs of skin diseases. Currently, there is limited research on automatic fish disease recognition specifically for Miichthys miiuy. To address this gap, we introduce, for the first time, a novel dataset for Miichthys miiuy and train, validate, and test the model on this dataset. DCW-YOLO substitutes the CIoU loss function in YOLOv10 with the NWD loss function, thereby improving the model’s ability to detect densely packed targets. The C2f-D-LKA layer is employed in place of the C2f convolutional layer, improving the model’s capacity to capture irregularly shaped and sized objects while effectively reducing computational overhead and parameter load. Additionally, the DySample upsampling structure, which utilizes point sampling, is introduced to increase image resolution without adding significant computational cost. Underwater experimental results show that the mAP50 and precision of DCW-YOLO reach 96.87 % and 95.46 %, respectively, representing improvements of 4.61 % and 3.24 % over the original YOLOv10 model. When deployed in aquaculture settings, this model provides rapid, low-cost real-time disease detection, helping farmers identify diseases early and mitigate potential losses.
Aquaculture ReportsAgricultural and Biological Sciences-Animal Science and Zoology
CiteScore
5.90
自引率
8.10%
发文量
469
审稿时长
77 days
期刊介绍:
Aquaculture Reports will publish original research papers and reviews documenting outstanding science with a regional context and focus, answering the need for high quality information on novel species, systems and regions in emerging areas of aquaculture research and development, such as integrated multi-trophic aquaculture, urban aquaculture, ornamental, unfed aquaculture, offshore aquaculture and others. Papers having industry research as priority and encompassing product development research or current industry practice are encouraged.