摘要： 集约化水产养殖中，实时监测鱼群摄食状态是制定科学投喂策略的重要依据之一。为实现鱼群摄食和非摄食状态的时序探测，解决现有方法在探测摄食状态切换时误差较高的问题，本文提出了基于改进边界匹配网络(Boundary Matching Network，BMN)的鱼群状态检测网络BMN-Fish。该网络在BMN的基础上，引入高效深度模块，模块包含高效通道注意力模块（Efficient Channel Attention Module Net，ECANet）和基础残差模块（Base Residual Module，BRM）。其中，高效通道注意力模块可扩大时序维度特征的感受野，提高了模型提取局部信息的能力；基础残差模块可关注特征图中感兴趣的区域，增强算法的全局感知能力。结果显示，相比于原始的鱼群时序摄食动作检测网络BMN，BMN-Fish的AUC和AR@100指标达93.32%和95.28%，分别提高2.17%和1.95%。研究表明，该方法可以高效检测鱼群时序摄食动作，为集约化水产养殖制定智能投饲策略提供参考。

关键词: 改进边界匹配网络, 鱼群摄食, 摄食状态检测, 摄食行为检测, 注意力机制, 水产养殖

Abstract: In intensive aquaculture, real-time monitoring of fish feeding status is one of the important bases for developing scientific feeding strategies. In order to achieve temporal detection of fish feeding states , and to solve the problem of high error in detecting feeding state switching by existing methods, this paper proposes a fish state detection network BMN-Fish based on an improved BMN (Boundary Matching Network), which introduces an efficient depth module and consists of an Efficient Channel Attention Module Net (ECANet) and a Base Residual Module (BRM). Among them, the Efficient Channel Attention Module can expand the perceptual field of the temporal dimensional features, so as to improve the ability of the model to extract local information; the Base Residual Module enables the algorithm to focus on the regions of interest in the feature map, which can strengthen the global perception ability of the algorithm. The experimental results showed that the AUC and AR@100 of BMN-Fish were 93.32% and 95.28%, which were 2.17% and 1.95% higher than BMN, respectively. BMN-Fish could effectively detect the sequential feeding movements of fish, and provide important information for the formulation of intelligent feeding strategies for intensive aquaculture.

Key words:  improved boundary matching network, fish feeding, feeding state detection, feeding behavior detection, attention mechanism, aquaculture