徐建瑜¹，刘  鹰²，崔绍荣³，苗香雯³
（1 宁波大学信息科学与工程学院，宁波 315211；2 中国科学院海洋研究所，青岛 266071；
3 浙江大学农业生物环境工程研究所，杭州 310029）

摘要：从行为参数的量化方法、应激状态下的行为变化两方面介绍了计算机视觉技术在鱼类应激状态监测方面的研究进展，并提出了进一步研究的方向是实际应用过程中量化鱼行为的新方法研究，以及根据鱼在不同条件下行为参数的变化特点，结合统计方法和人工智能算法建立自动判断应激的方法。

关键词：水产养殖；图像处理；计算机视觉；应激；行为参数；监测

Research progress in the application of computer-vision
to monitor the stress state of fish
XU Jian-yu¹, LIU Ying², MIAO Xiang-wen³, CUI Shao-rong³
( 1 Faculty of Information Science and Technology, Ningbo University, Ningbo 315211, China;
2 Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China;
3 Institute of Agricultural Bio-Environmental Engineering, Zhejiang University, Hangzhou 310029, China )

Abstract: The recent research progresses in the applications of computer vision in aquaculture to acquire aquatic’s stress state information of fish were reviewed in this paper, mainly on the methods of quantifying fish behaviors and the behavioral responses of fish under stress conditions. New methods using image processing to quantify fish behavioral parameters under culture conditions and behavioral alterations under a variety of stress deserve further research. Approach to predicting stress state applying statistics and artificial intelligence technique are also the future research directions.

Key words: aquaculture; image processing; computer vision; stress; behavioral parameter; monitoring

基金项目： 国家“863”计划项目（2006AA100305）；宁波大学人才基金项目（2006020）

作者简介：徐建瑜（1973—），女，副教授，博士，主要从事水产养殖工程及环境调控、数字图像处理等方面的研究。Email: xujianyu@nbu.edu.cn