深度学习在基于叶片的油茶品种识别中的研究

尹显明; 棘玉; 张日清; 莫登奎; 彭邵锋; 韦维

doi:10.12302/j.issn.1000-2006.202112037

尹显明, 棘玉, 张日清, 莫登奎, 彭邵锋, 韦维

南京林业大学学报（自然科学版） ›› 2023, Vol. 47 ›› Issue (3) : 29-36.

PDF(2921 KB)

国家林草科技领军期刊
中国精品科技期刊
中国高校百佳科技期刊
江苏省新闻出版政府奖期刊奖
RCCSE林学权威期刊（A+）
CSCD核心期刊
Scopus数据库收录期刊
中文核心期刊
SCD核心期刊

作者加群：102861116

微信公众号：南京林业大学学报

高级检索

PDF(2921 KB)

南京林业大学学报（自然科学版） ›› 2023, Vol. 47 ›› Issue (3) : 29-36. DOI: 10.12302/j.issn.1000-2006.202112037

专题报道：第三届中国林草计算机应用大会论文精选（执行主编李凤日）

深度学习在基于叶片的油茶品种识别中的研究

尹显明 ¹ ,
棘玉 ¹ ,
张日清 ¹^,^* ,
莫登奎 ¹ ,
彭邵锋 ² ,
韦维 ³

作者信息 +

Research on recognition of Camellia oleifera leaf varieties based on deep learning

YIN Xianming ¹ ,
JI Yu ¹ ,
ZHANG Riqing ¹^,^* ,
MO Dengkui ¹ ,
PENG Shaofeng ² ,
WEI Wei ³

Author information +

文章历史 +

摘要

【目的】 采用深度学习方法开展基于叶片的油茶品种识别研究,开发油茶品系图像识别技术,为油茶品种鉴别提供科学依据。【方法】选择自然光照条件下生长的11个油茶品种叶片作为研究对象,采集完整、无明显病虫害的叶片,以白色硬纸板为背景,利用智能手机对叶片的正、背面进行图像采集,通过可用性筛选去除无效图像,构建图像数量为2 791张的油茶叶片品种数据集,采用深度学习网络(GoogLeNet、ResNet)对11个油茶品种的叶片图像进行识别研究。【结果】GoogLeNet和ResNet网络均能满足基于叶片的油茶品种识别要求,总体识别准确率、召回率的调和平均值(F₁)分别达94.0%和80.7%;其中GoogLeNet网络识别效果更好,平均准确率、召回率、多分类模型指标宏观F₁(Macro F₁)和微观F₁(Micro F₁)分别为94.1%、94.0%、94.0%和96.9%,其对油茶品种编号1和编号8的识别召回率高达100%。【结论】深度学习网络(GoogLeNet、ResNet)能够实现基于叶片的油茶品种识别,可为基于其他作物的品种识别提供参考。

Abstract

【Objective】Deep learning methods are used to carry out research on Camellia oleifera based variety recognition on leaves, this study developed C. oleifera strain image recognition technology to provide scientific basis for C. oleifera variety identification.【Method】Eleven leaves of C. oleifera varieties grown under natural lighting conditions and free from pests and diseases were collected for a study. Images of the front and back of the leaves with a white cardboard background were captured using a smartphone. Invalid images were removed by usability screening, and a dataset of camellia leaf varieties with 2 791 images was constructed. Deep learning networks (GoogLeNet and ResNet) were used to identify and study the leaf images of 11 C. oleifera varieties.【Result】Both GoogLeNet and ResNet networks can meet the requirements of C. oleifera variety recognition based on leaves, with overall F₁ scores of 94.0% and 80.7%. Among them, the GoogLeNet network was more effective in recognition, with average accuracy, recall, Macro F₁ and Micro F₁ value of 94.1%, 94.0%, 94.0% and 96.9%, respectively, and its recognition recall for two varieties, NO. 1 and 8, reached 100%.【Conclution】Deep learning networks (GoogLeNet and ResNet) can achieve C. oleifera variety recognition based on leaves, which can provide a reference for rapid leaf-based C. oleifera variety recognition.

导出引用

尹显明, 棘玉, 张日清, 等. 深度学习在基于叶片的油茶品种识别中的研究[J]. 南京林业大学学报（自然科学版）. 2023, 47(3): 29-36 https://doi.org/10.12302/j.issn.1000-2006.202112037

YIN Xianming, JI Yu, ZHANG Riqing, et al. Research on recognition of Camellia oleifera leaf varieties based on deep learning[J]. JOURNAL OF NANJING FORESTRY UNIVERSITY. 2023, 47(3): 29-36 https://doi.org/10.12302/j.issn.1000-2006.202112037

中图分类号： TP391.41

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	王金凤, 谭新建, 吴喜昌, 等. 我国油茶产业发展现状与对策建议[J]. 世界林业研究, 2020, 33(6): 80-85. WANG J F, TAN X J, WU X C, et al. Development status and suggestions of Camellia industry in China[J]. World For Res, 2020, 33(6): 80-85. DOI: 10.13348/j.cnki.sjlyyj.2020.0103.y. 本文引用 [2]

[2]	李志钢, 马力, 陈永忠, 等. 我国油茶籽的综合利用现状概述[J]. 绿色科技, 2018(6): 191-194. LI Z G, MA L, CHEN Y Z, et al. Study on the comprehensive utilization of oil-tea Camellia seed[J]. J Green Sci Technol, 2018(6): 191-194. DOI: 10.16663/j.cnki.lskj.2018.06.068. 本文引用 [1]

[3]	南玉龙, 张慧春, 郑加强, 等. 深度学习在林业中的应用[J]. 世界林业研究, 2021, 34(5): 87-90. NAN Y L, ZHANG H C, ZHENG J Q, et al. Application of deep learning to forestry[J]. World For Res, 2021, 34(5): 87-90. DOI: 10.13348/j.cnki.sjlyyj.2021.0020.y. 本文引用 [1]

[4]	阳灵燕, 张红燕, 陈玉峰, 等. 机器学习在农作物品种识别中的应用研究进展[J]. 中国农学通报, 2020, 36(30): 158-164. YANG L Y, ZHANG H Y, CHEN Y F, et al. The application of machine learning in crop variety recognition: a review[J]. Chin Agric Sci Bull, 2020, 36(30): 158-164. DOI: 10.11924/j.issn.1000-6850.casb20191100813. 本文引用 [1]

[5]	周惠汝, 吴波明. 深度学习在作物病害图像识别方面应用的研究进展[J]. 中国农业科技导报, 2021, 23(5): 61-68. ZHOU H R, WU B M. Advances in research on deep learning for crop disease image recognition[J]. J Agric Sci Technol, 2021, 23(5): 61-68. DOI: 10.13304/j.nykjdb.2019.1041. 本文引用 [1]

[6]

严恩萍, 棘玉, 尹显明, 等. 基于无人机影像自动检测冠层果的油茶快速估产方法[J]. 农业工程学报, 2021, 37(16): 39-46.

YAN

E P

, JI

, YIN

X M

, et al. Rapid estimation of Camellia oleifera yield based on automatic detection of canopy fruits using UAV images[J]. Trans Chin Soc Agric Eng, 2021, 37(16): 39-46. DOI: 10.11975/j.issn.1002-6819.2021.16.006.

本文引用 [1]

[7]	余凯, 贾磊, 陈雨强, 等. 深度学习的昨天、今天和明天[J]. 计算机研究与发展, 2013, 50(9): 1799-1804. YU K, JIA L, CHEN Y Q, et al. Deep learning: yesterday, today, and tomorrow[J]. J Comput Res Dev, 2013, 50(9): 1799-1804. 本文引用 [1]

[8]	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Commun ACM, 2017, 60(6): 84-90. DOI: 10.1145/3065386. 本文引用 [1]

[9]	ZEILER M D, FERGUS R. Visualizing andunderstanding convolutional networks[C]// European Conference on Computer Vision. Cham: Springer, 2014: 818-833. DOI: 10.1007/978-3-319-10590-1_53. 本文引用 [1]

[10]	SIMONYAN K, ZISSERMAN A. Verydeep convolutional networks for large-scale image recognition[C]// International Conference on Learning Representations (ICLR). 2015: 1-14. arXiv preprint arXiv:14091556, 2014. 本文引用 [1]

[11]	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016:770-778. DOI: 10.1109/CVPR.2016.90. 本文引用 [2]

[12]	SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions[C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 7-12, 2015, Boston, MA, USA. IEEE, 2015: 1-9. DOI: 10.1109/CVPR.2015.7298594. 本文引用 [2]

[13]	韩斌, 曾松伟. 基于多特征融合和卷积神经网络的植物叶片识别[J]. 计算机科学, 2021, 48(S1): 113-117. HAN B, ZENG S W. Plant leaf image recognition based on multi-feature integration and convolutional neural network[J]. Comput Sci, 2021, 48(S1): 113-117. DOI: 10.11896/jsjkx.201100119. 本文引用 [1]

[14]	原忠虎, 王维, 苏宝玲. 基于改进VGGNet模型的外来入侵植物叶片识别方法[J]. 计算机与现代化, 2021(9): 7-11. YUAN Z H, WANG W, SU B L. Leaf recognition method of invasive alien plants based on improved VGGNet model[J]. Computer and Modernization, 2021(9): 7-11. 本文引用 [1]

[15]	王建霞, 张成, 闫双双. 基于卷积神经网络的宠物猫品种分类研究[J]. 河北工业科技, 2020, 37(6): 407-412. WANG J X, ZHANG C, YAN S S. Research on pet cat breed classification based on convolutional neural network[J]. Hebei J Ind Sci Technol, 2020, 37(6): 407-412. DOI: 10.7535/hbgykj.2020yx06004. 本文引用 [1]

[16]

石洪康, 田涯涯, 杨创, 等. 基于卷积神经网络的家蚕幼虫品种智能识别研究[J]. 西南大学学报(自然科学版), 2020, 42(12): 34-45.

SHI

H K

, TIAN

Y Y

, YANG

, et al. Research on intelligent recognition of silkworm larvae races based on convolutional neural networks[J]. J Southwest Univ (Nat Sci Ed), 2020, 42(12): 34-45. DOI: 10.13718/j.cnki.xdzk.2020.12.004.

本文引用 [2]

[17]	HOWARD A G, ZHU M L, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[EB/OL]. 2017: arXiv: 1704.04861. https://arxiv.org/abs/1704.04861. https://arxiv.org/abs/1704.04861 本文引用 [1]

[18]	杨静亚, 李景霞, 王振宇, 等. 基于卷积神经网络的花朵品种的识别[J]. 黑龙江大学工程学报, 2019, 10(4): 90-96. YANG J Y, LI J X, WANG Z Y, et al. Identification of flower varieties based on convolution neural network[J]. J Eng of Heilongjiang Univ, 2019, 10(4): 90-96. DOI: 10.13524/j.2095-008x.2019.04.060. 本文引用 [1]

[19]	游嘉伟, 王斌, 曾瑞. 用于大豆品种识别的叶片深度特征学习方法[J]. 计算机系统应用, 2021, 30(10): 118-127. YOU J W, WANG B, ZENG R. Learning method of leaf deep features for soybean cultivar recognition[J]. Computer Systems and Applications, 2021, 30(10): 118-127. 本文引用 [1]

[20]	薄琪苇. 基于卷积神经网络的植物叶片识别研究[D]. 杭州: 浙江农林大学, 2018. BO Q W. Study on plant leaf recognition based on convolutional neural network[D]. Hangzhou: Zhejiang A&F University, 2018. 本文引用 [1]

[21]	孙洋. 基于Android平台的植物叶片识别系统[D]. 保定: 河北大学, 2017. SUN Y. Plant leaf recognition system based on android platform[D]. Baoding: Hebei University, 2017. 本文引用 [1]

[22]	ABADI M, BARHAM P, CHEN J, et al. TensorFlow: a system for large-scale machine learning[C]// USENIX Association. Proceedings of the 12th USENIX conference on Operating Systems Design and Implementation. Savannah, GA, USA, 2016: 265-283. DOI: 10.5555/3026877.3026899. 本文引用 [1]

[23]	XIN J, ZHANG Y X, TANG Y, et al. Brain differences between men and women: evidence from deep learning[J]. Front Neurosci, 2019, 13: 185. DOI: 10.3389/fnins.2019.00185. 本文引用 [1]

[24]	IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]// Proceedings of the 32nd International Conference on International Conference on Machine Learning-Volume 37. JMLR.org. Lille, France, 2015: 448-456. 本文引用 [1]

[25]	MANNING C D, RAGHAVAN P, SCHUTZE H. Introduction to information retrieval[M]. Cambridge: Cambridge University Press, 2008. DOI: 10.1017/CBO9780511809071. 本文引用 [1]

[26]	SOKOLOVA M, LAPALME G. A systematic analysis of performance measures for classification tasks[J]. Inf Process Manag, 2009, 45(4): 427-437. DOI: 10.1016/j.ipm.2009.03.002. 本文引用 [1]

[27]	朱晓龙. 基于深度卷积神经网络的树木生境叶片识别方法研究[D]. 哈尔滨: 东北林业大学, 2020. ZHU X L. Study of tree leaf recognition in habitat based on deep convolutional neural networks[D]. Harbin: Northeast Forestry University, 2020. 本文引用 [1]