机器学习在苹果智慧生产中的应用  

Application of Machine Learning Algorithms in Smart Production of Apple

作　　者：荆珮 李民赞[1] 程涛 孙妮娜[2] 张焕春[2] 夏秀波[2] 杨剑超[2] JING Pei;LI Minzan;CHENG Tao;SUN Ni′na;ZHANG Huanchun;XIA Xiubo;YANG Jianchao(Key Laboratory on Modern Precision Agriculture System Integration Research,Ministry of Educa-tion,China Agricultural University,Beijing 100083,China;Institute of Smart Agriculture,Yantai Academy of Agricultural Sciences,Yantai 265500,China)

机构地区：[1]中国农业大学现代精细农业系统集成研究教育部重点实验室,北京100083 [2]烟台市农业科学研究院智慧农业研究所,烟台265500

出　　处：《吉林农业大学学报》2021年第2期138-145,共8页Journal of Jilin Agricultural University

基　　金：烟台市校地融合发展项目(2020XDRHXMPT35);国家自然科学基金项目(31971785)。

摘　　要：苹果是我国重要的园艺作物,机器学习和计算机视觉融合促进了苹果检测与识别技术的发展,为苹果智慧生产提供了新的支撑。文章以苹果智慧生产中苹果果实识别、苹果品质和营养的无损检测、苹果品质分级技术为重点,介绍了机器学习在苹果智慧生产领域中的应用进展。包括基于支持向量机的苹果果实识别方法,基于深度学习的在树水果实时识别方法,光谱技术结合机器学习的苹果品质检测,电子鼻结合机器学习的苹果品质检测,以及机器学习在苹果分级中的应用。最后分析指出机器学习在苹果智慧生产应用中的存在问题,展望了未来的发展方向。Apple is an important horticultural crop in China. The integration of machine learning and computer vision promotes the development of apple detection and recognition technology, and provides a new support for apple smart production. In this paper, focusing on apple fruit recognition, non-destructive detection of apple quality and nutrition, and apple quality grading technology, the progress of machine learning in the field of apple smart production is introduced, including apple fruit recognition based on SVM algorithm, real-time recognition of tree fruit based on deep learning, apple quality detection based on spectral technology combined with machine learning, apple quality detection based on electronic nose combined with machine learning, and application of machine learning in apple grading. Finally, the existing problems of machine learning in the application of apple smart production are pointed out, and the future development direction is prospected.

关 键 词：机器学习 苹果 计算机视觉 智慧农业 

分 类 号：S661.1[农业科学—果树学] TP181[农业科学—园艺学]