[1]代国威,陈稼瑜,樊景超.融合ResNet与支持向量机的葡萄园冠层图像叶片覆盖度分类[J].江苏农业学报,2023,(08):1713-1721.[doi:doi:10.3969/j.issn.1000-4440.2023.08.011]
　DAI Guo-wei,CHEN Jia-yu,FAN Jing-chao.Leaf coverage classification of vineyard canopy images based on ResNet and support vector machines[J].,2023,(08):1713-1721.[doi:doi:10.3969/j.issn.1000-4440.2023.08.011]
点击复制

融合ResNet与支持向量机的葡萄园冠层图像叶片覆盖度分类()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2023年08期
页码:
1713-1721
栏目:
农业信息工程
出版日期:
2023-12-31

文章信息/Info

Title:
Leaf coverage classification of vineyard canopy images based on ResNet and support vector machines
作者:
代国威1; 2; 陈稼瑜3; 樊景超1; 2
(1.中国农业科学院农业信息研究所,北京100081;2.国家农业科学数据中心,北京100081;3.浙江农林大学暨阳学院,浙江诸暨311800)
Author(s):
DAI Guo-wei1; 2; CHEN Jia-yu3; FAN Jing-chao1; 2
(1.Agricultural Information Institute, Chinese Academy of Agricultural Sciences, Beijing 100081, China;2.National Agriculture Science Data Center, Beijing 100081, China;3.Jiyang College, Zhejiang A&F University, Zhuji 311800, China)
关键词:
残差网络; 支持向量机; 近邻成分分析; 葡萄园冠层; 叶片覆盖度; 分类
Keywords:
residual network; support vector machine; neighborhood component analysis; canopy of vineyard; leaf coverage; classification
分类号:
TP391.4;S24
DOI:
doi:10.3969/j.issn.1000-4440.2023.08.011
文献标志码:
A
摘要:
在视觉感知的基础上,实现作物智能喷洒作业管理是智慧农业重要的组成部分。针对葡萄园智能喷洒作业的需要,本研究构建了一种融合残差网络(ResNet)和支持向量机模型的葡萄园冠层图像叶片覆盖度分类方法。在对葡萄园冠层图像数据集进行数据增强的基础上,利用不同卷积层数的ResNet模型(ResNet-18、ResNet-34和ResNet-50)提取图像特征向量,结合近邻成分分析(NCA)算法及不同分类模型(Cubic SVM、RBF SVM、Linear SVM、DT、BT、Bayes、KNN、RF),筛选出最优葡萄园冠层图像叶片覆盖度分类方法。结果表明:残差网络模型卷积层数的增加,有利于提高模型的分类精度；葡萄园冠层图像叶片覆盖度适宜的分类方法是利用ResNet-18、ResNet-34和ResNet-50各提取1 000个特征向量,进一步利用NCA算法筛选出1 000个权重值较大的特征向量,并利用Cubic SVM模型进行分类。该方法较好实现了模型训练时间和分类精度的平衡,既能大幅减少冗余的特征向量,缩短训练时间,还可以保证模型的分类精度。该方法下模型的分类准确率、精确率、召回率分别达98.32%、97.41%、98.73%。本研究建立的葡萄园冠层图像叶片覆盖度分类方法为智慧化的果园管理提供了有效的技术支持。
Abstract:
On the basis of visual perception, the realization of crop intelligent spraying operation management is an important part of intelligent agriculture. Aiming at the needs of intelligent spraying operations in vineyards, this study constructed a vineyard canopy image leaf coverage classification method that combines residual network (ResNet) and support vector machine model. Based on the data enhancement of the existing vineyard canopy image data set, the ResNet models with different convolution layers (ResNet-18, ResNet-34 and ResNet-50) were used to extract image feature vectors. Combined with the nearest neighbor component analysis (NCA) algorithm and different classification models (Cubic SVM, RBF SVM, Linear SVM, DT, BT, Bayes, KNN, RF), the optimal vineyard canopy image leaf coverage classification method was screened. The results showed that the increase of the number of convolution layers of the residual network model could improve the classification accuracy of the model. The suitable classification method of leaf coverage in vineyard canopy image was to extract 1 000 feature vectors by using ResNet-18, ResNet-34 and ResNet-50 respectively, and further use NCA algorithm to screen out 1 000 feature vectors with larger weight values, and use Cubic SVM model for classification. This method could realize a good balance between model training time and classification accuracy. It could not only greatly reduce redundant feature vectors, shorten training time, but also ensure the classification accuracy of the model. The classification accuracy, precision and recall rate of the model under this method were 98.32%, 97.41% and 98.73% respectively. The leaf coverage classification method of vineyard canopy image established in this study provides effective technical support for intelligent orchard management.

参考文献/References:

[1]ZHU M, LIU Z, ZENG Y, et al. Nordihydroguaiaretic acid reduces postharvest berry abscission in grapes[J]. Postharvest Biology and Technology,2022,183:111748.
[2]BALTAZAR A R, SANTOS F N, MOREIRA A P, et al. Smarter robotic sprayer system for precision agriculture[J]. Electronics,2021,10(17):2061.
[3]胡玲艳,周婷,刘艳,等. 基于轻量级网络自适应特征提取的番茄病害识别[J]. 江苏农业学报,2022,38(3):696-705.
[4]MALLIKARJUNA S B, SHIVAKUMARA P, KHARE V, et al. Multi-gradient-direction based deep learning model for arecanut disease identification[J]. CAAI Transactions on Intelligence Technology,2022,7(2):156-166.
[5]赵恒谦,杨屹峰,刘泽龙,等. 农作物叶片病害迁移学习分步识别方法[J]. 测绘通报,2021(7):34-38.
[6]YANG J, ZHANG T, FANG C, et al. A defencing algorithm based on deep learning improves the detection accuracy of caged chickens [J]. Computers and Electronics in Agriculture,2023,204:107501.
[7]李科岑,王晓强,林浩,等. 深度学习中的单阶段小目标检测方法综述[J]. 计算机科学与探索,2022,16(1):41-58.
[8]翟先一,魏鸿磊,韩美奇,等. 基于改进YOLO卷积神经网络的水下海参检测[J]. 江苏农业学报,2023,39(7):1543-1553.
[9]DAI G, FAN J, DEWI C. ITF-WPI: Image and text based cross-modal feature fusion model for wolfberry pest recognition[J]. Computers and Electronics in Agriculture,2023,212:108129.
[10]SHE J, ZHAN W, HONG S, et al. A method for automatic real-time detection and counting of fruit fly pests in orchards by trap bottles via convolutional neural network with attention mechanism added[J]. Ecological Informatics,2022,70:101690.
[11]CHEN X, CHENG Z, WANG S, et al. Atrial fibrillation detection based on multi-feature extraction and convolutional neural network for processing ECG signals[J]. Computer Methods and Programs in Biomedicine,2021,202:106009.
[12]TSALERA E, PAPADAKIS A, SAMARAKOU M, et al. Feature extraction with handcrafted methods and convolutional neural networks for facial emotion recognition[J]. Applied Sciences,2022,12(17):8455.
[13]HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]. IEEE:Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016:770-778.
[14]李颀,杨军. 基于多分辨率特征融合的葡萄尺寸检测[J].江苏农业学报,2022,38(2):394-402.
[15]SAEED A, ABDEL-AZIZ A A, MOSSAD A, et al. Smart detection of tomato leaf diseases using transfer learning-based convolutional neural networks[J]. Agriculture,2023,13(1): 139.
[16]KOK Z H, MOHAMED S A R, ALFATNI M S M, et al. Support vector machine in precision agriculture: a review[J]. Computers and Electronics in Agriculture,2021,191:106546.
[17]WANG G, ZHENG H, LI X. ResNeXt-SVM: a novel strawberry appearance quality identification method based on ResNeXt network and support vector machine[J]. Journal of Food Measurement and Characterization,2023,17(5):4345-4356.
[18]BAL F, KAYAALP F. A novel deep learning-based hybrid method for the determination of productivity of agricultural products: Apple case study[J]. IEEE Access,2023,11:7808-7821.
[19]LIU B, TAN C, LI S, et al. A data augmentation method based on generative adversarial networks for grape leaf disease identification[J]. IEEE Access,2020,8:102188-102198.
[20]AL-WESABI F N, ALBRAIKAN A A, HILAL A M, et al. Artificial intelligence enabled apple leaf disease classification for precision agriculture[J]. Cmc-Computers Materials & Continua,2022,70(3):6223-6238.
[21]李颀,陈哲豪. 基于改进单次多目标检测器的果面缺陷冬枣实时检测[J]. 江苏农业学报,2022,38(1):119-128.
[22]LI L, QIAO J, YAO J, et al. Automatic freezing-tolerant rapeseed material recognition using UAV images and deep learning[J]. Plant Methods,2022,18:5.
[23]KARTHIKEYAN M, RAJA D. Deep transfer learning enabled DenseNet model for content based image retrieval in agricultural plant disease images[J]. Multimedia Tools and Applications, 2023,82:36067-36090.
[24]LI K, WANG J, JALIL H, et al. A fast and lightweight detection algorithm for passion fruit pests based on improved YOLOv5[J]. Computers and Electronics in Agriculture,2023,204:107534.
[25]SINGH P, SINGH P, FAROOQ U, et al. CottonLeafNet: cotton plant leaf disease detection using deep neural networks[J]. Multimedia Tools and Applications,2023,82:37151-37176.
[26]QIAN S, DU J, ZHOU J, et al. An effective pest detection method with automatic data augmentation strategy in the agricultural field[J]. Signal, Image and Video Processing,2023,17(2):563-571.
[27]NAN Y, ZHANG H, ZENG Y, et al. Faster and accurate green pepper detection using NSGA-II-based pruned YOLOv5l in the field environment[J]. Computers and Electronics in Agriculture,2023,205:107563.
[28]KHALIFA N E, LOEY M, MIRJALILI S. A comprehensive survey of recent trends in deep learning for digital images augmentation[J]. Artificial Intelligence Review, 2022,55:2352-2377.
[29]BUSLAEV A, IGLOVIKOV V I, KHVEDCHENYA E, et al. Albumentations: fast and flexible image augmentations[J]. Information,2020,11(2):125.
[30]WENG Z, MENG F, LIU S, et al. Cattle face recognition based on a Two-Branch convolutional neural network[J]. Computers and Electronics in Agriculture,2022,196:106871.
[31]DAI G, HU L, FAN J, et al. A deep learning-based object detection scheme by improving yolov5 for sprouted potatoes datasets[J]. IEEE Access,2022,10:85416-85428.
[32]储鑫,李祥,罗斌,等. 基于改进YOLOv4算法的番茄叶部病害识别方法[J]. 江苏农业学报,2023,39(5):1199-1208.
[33]DAI G, HU L, FAN J. Da-actnn-yolov5: hybrid yolo v5 model with data augmentation and activation of compression mechanism for potato disease identification[J]. Computational Intelligence and Neuroscience,2022,2022:1-16.
[34]阮子行,黄勇,王梦,等. 基于改进卷积神经网络的番茄品质分级方法[J]. 江苏农业学报,2023,39(4):1005-1014.
[35]KOC M, SUT S K, SERHATLIOGLU I, et al. Automatic prostate cancer detection model based on ensemble VGGNet feature generation and NCA feature selection using magnetic resonance images[J]. Multimedia Tools and Applications,2022,81(5):7125-7144.
[36]TAO T, WEI X. A hybrid CNN-SVM classifier for weed recognition in winter rape field[J]. Plant Methods,2022,18(1):29.
[37]ZENONE T, VITALE L, FAMULARI D, et al. Application of machine learning techniques to simulate the evaporative fraction and its relationship with environmental variables in corn crops[J]. Ecological Processes,2022,11(1):758-771.

相似文献/References:

[1]单捷,邱琳,孙玲,等.基于Radarsat-2的水稻种植面积提取[J].江苏农业学报,2017,(03):561.[doi:doi:10.3969/j.issn.1000-4440.2017.03.012]
　SHAN Jie,QIU Lin,SUN Ling,et al.Paddy rice planting area extraction based on Radarsat-2 data[J].,2017,(08):561.[doi:doi:10.3969/j.issn.1000-4440.2017.03.012]
[2]苗荣慧,黄锋华,杨华,等.基于空谱一体化的农田高光谱图像分类[J].江苏农业学报,2018,(04):818.[doi:doi:10.3969/j.issn.1000-4440.2018.04.015]
　MIAO Rong-hui,HUANG Feng-hua,YANG hua,et al.Farmland classification of hyperspectral image based on spatial-spectral integration method[J].,2018,(08):818.[doi:doi:10.3969/j.issn.1000-4440.2018.04.015]
[3]孙玉婷,杨红云,王映龙,等.基于支持向量机的水稻叶面积测定[J].江苏农业学报,2018,(05):1027.[doi:doi:10.3969/j.issn.1000-4440.2018.05.009]
　SUN Yu-ting,YANG Hong-yun,WANG Ying-long,et al.Determination of rice leaf area based on support vector machine[J].,2018,(08):1027.[doi:doi:10.3969/j.issn.1000-4440.2018.05.009]
[4]许伟栋,赵忠盖.基于卷积神经网络和支持向量机算法的马铃薯表面缺陷检测[J].江苏农业学报,2018,(06):1378.[doi:doi:10.3969/j.issn.1000-4440.2018.06.025]
　XU Wei-dong,ZHAO Zhong-gai.Potato surface defects detection based on convolution neural networks and support vector machine algorithm[J].,2018,(08):1378.[doi:doi:10.3969/j.issn.1000-4440.2018.06.025]
[5]王新忠,卢青,张晓东,等.基于高光谱图像的黄瓜种子活力无损检测[J].江苏农业学报,2019,(05):1197.[doi:doi:10.3969/j.issn.1000-4440.2019.05.028]
　WANG Xin-zhong,LU Qing,ZHANG Xiao-dong,et al.Non-destructive detection of cucumber seeds vigor based on hyperspectral imaging[J].,2019,(08):1197.[doi:doi:10.3969/j.issn.1000-4440.2019.05.028]
[6]孙晓明,陈小龙,余向阳,等.基于近红外光谱分析技术的水蜜桃产地溯源[J].江苏农业学报,2020,(02):507.[doi:doi:10.3969/j.issn.1000-4440.2020.02.035]
　SUN Xiao-ming,CHEN Xiao-long,YU Xiang-yang,et al.Traceability of honey peach origin using near infrared spectroscopy analysis techniques[J].,2020,(08):507.[doi:doi:10.3969/j.issn.1000-4440.2020.02.035]
[7]江远东,李新国,杨涵.基于连续小波变换的表层土壤有机碳含量的高光谱估算[J].江苏农业学报,2023,(01):118.[doi:doi:10.3969/j.issn.1000-4440.2023.01.014]
　JIANG Yuan-dong,LI Xin-guo,YANG Han.Hyperspectral estimation of organic carbon content in surface soils based on continuous wavelet transform[J].,2023,(08):118.[doi:doi:10.3969/j.issn.1000-4440.2023.01.014]
[8]杨浩,张通,阳苇丽,等.基于图像处理的雪茄烟叶晾制期间含水率预测模型比较[J].江苏农业学报,2023,(09):1891.[doi:doi:10.3969/j.issn.1000-4440.2023.09.011]
　YANG Hao,ZHANG Tong,YANG Wei-li,et al.Comparison of prediction models for moisture content of cigar tobacco leaves during drying period based on image processing[J].,2023,(08):1891.[doi:doi:10.3969/j.issn.1000-4440.2023.09.011]
[9]于天祥,樊红.基于Sentinel-2多时相遥感影像的冬小麦种植面积监测[J].江苏农业学报,2024,(09):1653.[doi:doi:10.3969/j.issn.1000-4440.2024.09.009]
　YU Tianxiang,FAN Hong.Remote sensing monitoring of winter wheat planting area based on multi-temporal Sentinel-2 imagery[J].,2024,(08):1653.[doi:doi:10.3969/j.issn.1000-4440.2024.09.009]

备注/Memo

备注/Memo:
收稿日期:2022-10-12基金项目:国家重点研发计划项目(2021YFF0704200);中国农业科学院院级基本科研业务费项目(Y2022LM20);中国农业科学院科技创新工程项目(CAAS-ASTIP-2016-AII)作者简介:代国威(1997-),男,四川德阳人,硕士研究生,主要从事人工智能及农业信息化研究。(Tel)15623225909；(E-mail)dgwstyle@foxmail.com通讯作者:樊景超,(E-mail)fanjingchao@caas.cn
更新日期/Last Update: 2024-01-15