长江科学院院报2024，Vol.41Issue(3)：30-36,53,8.DOI:10.11988/ckyyb.20221436
基于关键点检测的鱼类游动轨迹提取
Fish Trajectory Extraction Based on Landmark Detection
摘要
Abstract
The existing fish trajectory extraction methods fail to balance efficiency and accuracy.This study intro-duces 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 offish targets with extra labels and generating a fish key point dataset.The findings demonstrate that the proposed research meth-od 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 MREx(0.065％,0.092％)and MREy(0.112％,0.011％),alig-ning 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Key words
fish/fishway monitoring/detection of fish landmark/fish trajectory extraction/RetinaFace model分类
建筑与水利引用本文复制引用
石小涛,马欣,黄志勇,胡晓,威力斯..基于关键点检测的鱼类游动轨迹提取[J].长江科学院院报,2024,41(3):30-36,53,8.基金项目
国家自然科学基金项目(52179070) （52179070）
国家优秀青年科学基金项目(51922065) （51922065）
