[1]游梓童,吴福明,赵 淼,等.融合高阶信息增强模块的复杂背景植物叶片图像分类[J].南京师范大学学报(工程技术版),2022,22(03):045-52.[doi:10.3969/j.issn.1672-1292.2022.03.007]
　You Zitong,Wu Fuming,Zhao Miao,et al.Classification of Complex Background Plant Leaf Images Combined with High-Level Information Enhancement Module[J].Journal of Nanjing Normal University(Engineering and Technology),2022,22(03):045-52.[doi:10.3969/j.issn.1672-1292.2022.03.007]
点击复制

融合高阶信息增强模块的复杂背景植物叶片图像分类
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
22卷
期数:
2022年03期
页码:
045-52
栏目:
计算机科学与技术
出版日期:
2022-09-15

文章信息/Info

Title:
Classification of Complex Background Plant Leaf Images Combined with High-Level Information Enhancement Module
文章编号:
1672-1292(2022)03-0045-08
作者:
游梓童; 吴福明; 赵 淼; 业 宁
(南京林业大学信息科学技术学院,江苏 南京 210037)
Author(s):
You Zitong; Wu Fuming; Zhao Miao; Ye Ning
(College of Information Science and Technology,Nanjing Forestry University,Nanjing 210037,China)
关键词:
植物叶片; 叶片分类识别; 特征提取; CNN; 深度学习
Keywords:
plant leaves; classification and identification of leaves; feature extraction; CNN; deep learning
分类号:
TP391.41
DOI:
10.3969/j.issn.1672-1292.2022.03.007
文献标志码:
A
摘要:
植物叶片对植物种类分辨与认知具有重大研究作用. 提出了一种充分提取植物叶片特征信息的高阶信息增强模块,使用包含高阶信息增强模块的卷积神经网络模型对植物叶片图像进行多感受野特征提取. 以复杂背景下的植物叶片图像为研究对象,从中国植物图像库中获取样本来源不同的植物叶片图像构成含有9种叶片的PLD_amp数据集,采用添加高斯噪声、数据增广技术平滑和增扩数据集,增强数据的可操作性. 与现有传统卷积网络相比,所提出的包含高阶信息增强模块的CNN模型最佳分类准确率可达88.7%,具有较高可行性与高分类准确率.
Abstract:
Plant leaves play an essential role in the study of plant species discrimination and cognition. This paper proposes a high-level information enhancement module that fully extracts the feature information of plant leaves,and uses the convolutional neural network containing this module to extract the features of plant leaf images from multiple receptive fields. The experiment takes the plant leaf images in complex background as the research object,and plant leaf images from different sample sources are obtained from Plant Photo Bank of China(PPBC). These images constitute the PLD_amp data set containing nine kinds of leaves. The techniques of adding Gaussian noise and data augmentation are used to smooth and expand the data set,thereby enhancing the data set's operability. The CNN model's best classification accuracy with a high-order information enhancement module proposed in this paper for plant leaf image classification in complex backgrounds reaches 88.7%. Compared with the existing traditional convolutional network,it has higher feasibility and classification accuracy,and it provides a new idea for plant leaf image recognition under complex background.

参考文献/References:

[1]AZLAH M A F,CHUA L S,RAHMAD F R,et al. Review on techniques for plant leaf classification and recognition[J]. Computers,2019,8(4):77.
[2]李俞利. 植物叶片分类方法研究[D]. 兰州:西北师范大学,2021.
[3]田诗晨,徐玉丹,李瑀馨. 基于Bagging-PNN算法的树叶分类方法优化[J]. 自动化与仪表,2020,35(8):52-55,61.
[4]孙丽萍,陈泓钢,岳琪,等. 依据特征融合和深度学习的树木叶片分类方法[J]. 东北林业大学学报,2020,48(6):51-55,65.
[5]HUA Y M,GUO J H, ZHAO H. Deel belief networks and deep learning[C]//Proceedings of 2015 International Conference on Intelligent Computing and Internet of Things. Harbin,China:IEEE,2015.
[6]ALESSANDRINI M,RIVERA R C F,FALASCHETTI L,et al. A grapevine leaves dataset for early detection and classification of Esca disease in vineyards through machine learning[J]. Data in Brief,2021,35:106809.
[7]SONG Y,WANG X Z,XIE H L,et al. Quality Evaluation of Keemun black tea by fusing data obtained from near-infrared reflectance spectroscopy and computer vision sensors[J]. Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy,2021,252:119522.
[8]NANDHINI S,ASHOKKUMAR K. Improved crossover based monarch butterfly optimization for tomato leaf disease classification using convolutional neural network[J]. Multimedia Tools and Applications,2021,80:18583-18610.
[9]AKTER R,HOSEN M I. CNN-based Leaf Image Classification for Bangladeshi Medicinal Plant Recognition[C]//Proceedings of 2020 Emerging Technology in Computing,Communication and Electronics(ETCCE). Bangladesh:IEEE,2021.
[10]EL-MAGED L M A,DARWISH A,HASSANIEN A E. Artificial Intelligence-Based Plant Diseases Classification[M]//HASSANIEN A E,DARWISH A. Machine Learning and Big Data Analytics Paradigms:Analysis,Applications and Challenges. Switzerland:Springer Cham,2021:45-61.
[11]KHEIRKHAH F M,ASGHARI H. Plant leaf classification using gist texture features[J]. IET Computer Vision,2019,13(4):369-375.
[12]张帅,淮永建. 基于分层卷积深度学习系统的植物叶片识别研究[J]. 北京林业大学学报,2016,38(9):108-115.
[13]ZOUMPOURLIS G,DOUMANOGLOU A,VRETOS N,et al. Non-linear convolution filters for CNN-based learning[C]//Proceedings of 2017 International Conference on Computer Vision. Venice,Italy:IEEE,2017.
[14]XIN Y A,LZ B,SY A,et al. A multi-scale convolutional neural network with context for joint segmentation of optic disc and cup[J]. Artificial Intelligence in Medicine,2021,113:102035.
[15]LI L,ZHAO K Y,SUN R Z,et al. Research for an adaptive classifier based on dynamic graph learning[J]. Neural Processing Letters,2021[2021-08-31]. http://link.springer.com/article/10.1007/S11063-021-10452-7.
[16]YANG K L,ZHONG W Z,LI F G. Leaf segmentation and classification with a complicated background using deep learning[J]. Agronomy,2020,10(11):1721.
[17]MOHTASHAMIAN M,KARIMIAN M,MOOLA F,et al. Automated plant species identification using leaf shape-based classification techniques:a case study on iranian maples[J]. Iranian Journal of Science and Technology:Transactions of Electrical Engineering,2021,45:1051-1061.
[18]KADIR A,NUGROHO L E,SUSANTO A,et al. Leaf classification using shape,color,and texture features[J]. International Journal of Computer Trends and Technology,2011,1(3):225-230.
[19]陆健健,江开忠. 融合数据增广技术与机器学习算法的个人信用评分研究[J]. 软件导刊,2020,19(8):40-43.
[20]高强,张凤荔,王瑞锦,等. 轨迹大数据:数据处理关键技术研究综述[J]. 软件学报,2017,28(4):959-992.
[21]LV L Y,SONG X G,SUN W. Modify leave-one-out cross validation by moving validation samples around random normal distributions:move-one-away cross validation[J]. Applied Sciences,2020,10(7):2448.

备注/Memo

备注/Memo:
收稿日期:2021-08-31.
基金项目:国家重点研发计划项目(2016YFD0600101)、江苏省大学生创新实践项目(202110298039).
通讯作者:业宁,博士,教授,研究方向:数据挖掘、机器学习. E-mail:yening@njfu.edu.cn
更新日期/Last Update: 2022-09-15