海洋渔业2026，Vol.48Issue(4)：433-443,11.
人工智能驱动的水产品智慧加工:数据、算法与应用
AI-driven intelligent processing of aquatic products:data,algorithms,and applications
摘要
Abstract
Current aquatic product processing in China faces practical challenges such as high labor costs,labor shortages,and low detection efficiency.The rapid development of artificial intelligence provides a new technological pathway for the transformation of the aquatic product processing industry.Based on recent studies in related fields,this review systematically summarizes artificial intelligence-driven intelligent processing technologies for aquatic products,focusing on three key aspects:data acquisition,algorithm modeling,and application deployment.In terms of data acquisition,computer vision,spectroscopic techniques,hyperspectral imaging,and multimodal data fusion are introduced.In terms of algorithm models,commonly used deep learning algorithms in aquatic product processing,including convolutional neural network(CNN),YOLO(you only look once)object detection algorithms,and Transformers,are systematically reviewed.Model evaluation methods and the value of explainable artificial intelligence in improving model credibility are also discussed.In terms of application deployment,recent research cases involving typical scenarios such as intelligent screening,intelligent cutting,and intelligent detection are reviewed,demonstrating the practical effectiveness of artificial intelligence in improving processing efficiency.Finally,future development trends in explainable models,lightweight algorithms,and multi-technology integration for intelligent aquatic product processing are discussed,with the aim of providing a reference for promoting the intelligent upgrading of the aquatic product processing industry.关键词
水产品/智慧加工/深度学习/人工智能/应用部署Key words
aquatic products/intelligent processing/deep learning/artificial intelligence/application deployment分类
农业科技引用本文复制引用
郑尧,郑汉丰,郭全友,方辉..人工智能驱动的水产品智慧加工:数据、算法与应用[J].海洋渔业,2026,48(4):433-443,11.基金项目
国家重点研发计划"海洋农业与淡水渔业科技创新"专项(2024YFD2400601) （2024YFD2400601）
