基于关键点检测的鱼类游动轨迹提取  

Fish Trajectory Extraction Based on Landmark Detection

作　　者：石小涛[1] 马欣 黄志勇[1,3] 胡晓[1,2] 威力斯 SHI Xiao-tao;MA Xin;HUANG Zhi-yong;HU Xiao;WEI Li-si(Hubei International Science and Technology Cooperation Base of Fish Passage,China Three Gorges University,Yichang 443002,China;College of Hydraulic and Environmental Engineering,China Three Gorges University,Yichang 443002,China;College of Computer and Information Technology,China Three Gorges University,Yichang 443002,China)

机构地区：[1]三峡大学湖北省鱼类过坝技术国际科技合作基地,湖北宜昌443002 [2]三峡大学水利与环境学院,湖北宜昌443002 [3]三峡大学计算机与信息学院,湖北宜昌443002

出　　处：《长江科学院院报》2024年第3期30-36,53,共8页Journal of Changjiang River Scientific Research Institute

基　　金：国家自然科学基金项目(52179070);国家优秀青年科学基金项目(51922065)。

摘　　要：针对现有的鱼类游动轨迹提取方法不能兼顾轨迹提取效率和准确率的局限性,提出了一种基于鱼类关键点识别与定位的鱼类游动轨迹提取方法。该方法在RetinaFace算法的基础上,通过改进网络结构和损失函数、优化锚框的尺寸设计、编解码鱼类关键点(头部点和形心点)、为鱼类目标的关键点添加额外的标注并制作成鱼类关键点数据集等改进策略,构建了基于关键点识别的鱼类轨迹提取模型。研究结果表明,本研究方法对鱼体关键点识别的精度很高,准确率、召回率、平均精度均值3项精度评价指标分别为97.12%、95.72%、96.42%;所提取的轨迹坐标平均相对偏差为MRE x(0.065%,0.092%)、MRE y(0.112%,0.011%),与鱼类的实际游动轨迹基本吻合;鱼类目标关键点的识别速度可达32帧/s,能够满足实时提取鱼类轨迹的需求。The existing fish trajectory extraction methods fail to balance efficiency and accuracy.This study introduces a fish trajectory extraction approach based on fish landmark recognition and location utilizing the RetinaFace algorithm.The method entails constructing a fish trajectory extraction model through enhanced network structure and loss function for landmark detection,optimizing anchor size design,and encoding and decoding fish landmarks(specifically,the head point and centroid point).Additionally,it involves supplementing landmarks of fish targets with extra labels and generating a fish key point dataset.The findings demonstrate that the proposed research method achieves high accuracy in identifying fish landmarks,with precision evaluation indices including an accuracy rate of 97.12%,a recall rate of 95.72%,and a mean average precision of 96.42%.Moreover,the average relative deviation of the extracted trajectory coordinates is MRE x(0.065%,0.092%)and MRE y(0.112%,0.011%),aligning closely with the actual swimming trajectory of fish.The recognition rate for landmarks of fish targets reaches 32 frames per second,which meets the real-time extraction requirements for fish trajectory recognition.

关 键 词：鱼类 鱼道监测 鱼类关键点检测 鱼类游动轨迹提取 RetinaFace模型 

分 类 号：TV213.4[水利工程—水文学及水资源]