您好，欢迎访问北京市农林科学院 机构知识库！
北京市农林科学院机构知识库

全部

  • 全部
  • 标题
  • 学者
  • 机构
  • 关键词

DCA-MVIT: Fused DSGated convolution and CA attention for fish feeding behavior recognition in recirculating aquaculture systems

文献类型： 外文期刊

作者： Hu, Weichen 1 ; Yang, Xinting 1 ; Ma, Pingchuan 1 ; Fu, Tingting 1 ; Zhou, Chao 1 ;

作者机构： 1.Natl Engn Res Ctr Informat Technol Agr, Beijing 100097, Peoples R China

2.Shanghai Ocean Univ, Coll Informat, Shanghai 201306, Peoples R China

3.Natl Engn Lab Agriprod Qual Traceabil, Beijing 100097, Peoples R China

4.Beijing Acad Agr & Forestry Sci, Res Ctr Informat Technol, Beijing 100097, Peoples R China

关键词： Recirculating aquaculture system; Fish feeding behavior; Feature extraction; DCA-MVIT

期刊名称：AQUACULTURE （ 影响因子：3.9； 五年影响因子：4.4 ）

ISSN： 0044-8486

年卷期： 2025 年 598 卷

页码：

收录情况： SCI

摘要： In recirculating aquaculture systems, real-time and accurate recognition of fish feeding behavior from video is the foundation for the development of intelligent feeding equipment and system. However, complex features and abnormal postures have brought great challenges to achieve the above objectives. Therefore, this paper proposes a fish feeding behavior recognition model, DCA-MVIT (DSGated convolution and Coordinate Attention for Multiscale Vision Transformers version 2), with enhanced feature extraction ability. Firstly, data enhancement such as horizontal flipping, random noise and random filtering are introduced to solve the dataset equilibration issues. Then, the Coordinate Attention enhances the feature capture ability of the feeding domain. Finally, the proposed DSGated convolution further improves the feature capture ability and reduces model parameters. The experimental results showed that the Precision and Top-1 Accuracy of DCA-MVIT in fish feeding behavior recognition reached 96.62 % and 83.13 %. These are 1.86 % and 5.52 % higher than the original model, respectively. The proposed model provides important technical support for developing scientific feeding strategies and intelligent feeding systems.

  • 相关文献
作者其他论文 更多>>
置顶
购物车
反馈

© 京ICP备09089781号-29 主办单位：北京市农林科学院

学术资源： 中国农业科学院 农业专业知识服务系统 农科联盟资源共建共享平台 中国农业科技文献与信息服务平台 中国农业期刊集成服务平台