深度学习在果蔬品质检测中的应用综述  

A review of deep learning application in fruit and vegetable quality inspection

作　　者：吴智丽 杜志龙[1,2,3] 何亚凯 赵丹[1,2] 霍达 万丽娜[1,2] 杨延辰 毕峰华 WU Zhili;DU Zhilong;HE Yakai;ZHAO Dan;HUO Da;WAN Lina;YANG Yanchen;BI Fenghua(Chinese Academy of Agricultural Mechanization Sciences Group Co.,Ltd.,Beijing 100083,China;China National Packaging and Food Machinery Corporation,Beijing 100083,China;State Key Laboratory of Agricultural Equipment Technology,Beijing 100083,China)

机构地区：[1]中国农业机械化科学研究院集团有限公司,北京100083 [2]中国包装和食品机械有限公司,北京100083 [3]农业装备技术全国重点实验室,北京100083

出　　处：《农业工程》2025年第1期120-128,共9页AGRICULTURAL ENGINEERING

基　　金：国家重点研发计划项目(2022YFD2002205-01);农业农村部农产品产地初加工重点实验室开放课题项目(KLAPPP2022-01)。

摘　　要：随着全球人口的增长和对果蔬质量的高需求,果蔬品质检测的效率和准确性成为食品行业关注的重点。传统的果蔬品质检测方法存在劳动强度大、效率低和准确性难以保证等问题。详细介绍了深度学习在果蔬品质检测视觉任务中的应用。深度学习可以通过学习和模拟人的视觉识别过程,实现高效、准确地检测果蔬品质。具体应用包括果蔬色度分析、形状分析、纹理分析和尺寸分析,以及病害识别和成熟度识别等。然而,深度学习在果蔬品质检测中仍面临数据收集、模型训练的复杂性和实时性等问题。未来的研究可以致力于提高检测精度、处理速度,降低成本,并探索深度学习与物联网、大数据等技术的结合,实现更智能、更自动化的果蔬检测。深度学习在果蔬检测中的应用具有广阔的发展前景,将为果蔬生产和消费者带来重要影响。With increase of global population and high demand for fruit and vegetable quality,efficiency and accuracy of fruit and ve-getable quality inspection have become focal points in food industry.Traditional methods of fruit and vegetable quality inspection face challenges such as high labor intensity,low efficiency,and difficulty in ensuring accuracy.Application of deep learning in computer vision tasks of fruit and vegetable quality inspection was described in detail.Deep learning could achieve efficient and accurate inspec-tion by learning and simulating human visual recognition process to detect quality of fruits and vegetables.Specific applications include chromatic analysis,shape analysis,texture analysis,size analysis,disease identification,and ripeness recognition.However,deep learning in fruit and vegetable quality inspection still faces challenges in data collection,model training complexity,and achieving real-time performance.Future research could focus on improving inspection accuracy,processing speed,and reducing costs,exploring in-tegration of deep learning with internet of things,big data and other technologies,to achieve more intelligent and automated fruit and ve-getable inspection system.Application of deep learning in fruit and vegetable inspection has a broad development prospect and will bring a significant impact on fruit and vegetable production and consumers.

关 键 词：果蔬 品质检测 深度学习 计算机视觉 模式识别 

分 类 号：S24[农业科学—农业电气化与自动化]