基于改进的YOLOv5苹果叶部病害识别研究

doi:10.12190/j.issn.2096-1197.2024.01.14

摘要/Abstract

摘要： 【目的】提出一种基于改进YOLOv5模型的病害目标检测算法,实现对苹果叶部病害的自动识别,解决YOLOv5检测模型存在的漏检和误检问题。【方法】基于卷积神经网络改进的YOLOv5模型,采用加权双向特征金字塔网络（BiFPN）特征融合方法,有效改善PANet对多尺度特征融合的不良影响,并加入CBAM模块,使网络能更准确地定位和识别苹果叶部病害,建立一种苹果叶部病害检测的算法模型;使用ATCSP模块和自上而下的特征融合方法来增强模型对多尺度疾病的检测效果,并将该模型与SSD、YOLOv4、YOLOv6和YOLOv7模型进行对比。【结果】改进的YOLOv5检测算法模型显著提高了苹果叶部病害检测的精度,对比原始算法,精度（P）提升了5.1%,达到90.8%;平均精度均值（mAP）提高了1.2%,达到93.4%;模型大小减少21.4 MB。改进后的YOLOV5算法精度比SSD、YOLOv4、YOLOv6和YOLOv7模型分别高11.3、4.4、4.2、3.6个百分点。【结论】提出了一种基于卷积神经网络改进的YOLOv5苹果叶部病害检测模型,改进后的YOLOv5模型检测速度快、准确率高,且模型较小,能够实现对苹果叶部病害的自动识别。

关键词: YOLOv5, 苹果, 叶部病害, 识别, 卷积神经网络

Abstract: 【Objective】Propose a disease target detection algorithm based on improved YOLOv5 model,to achieve automatic recognition of apple leaf diseases and solve the problems of miss and false detection in the YOLOv5 detection model.【Methods】Based on the YOLOv5 model improved by convolutional neural network,weighted bidirectional feature pyramid network（BiFPN）feature fusion method was used to effectively improve the adverse effect of PANet on multi-scale feature fusion. The CBAM module was added to enable the network to more accurately locate and identify apple leaf diseases and establishing an algorithm model for detecting apple leaf diseases. The ATCSP module and top-down feature fusion method were used to enhance the detection performance of the model for multi-scale diseases. The model was compared with SSD,YOLOv4,YOLOv6,and YOLOv7 models.【Results】The improved YOLOv5 detection algorithm model significantly improved the accuracy of apple leaf disease detection. Compared with the original algorithm,accuracy（P） increased by 5.1%,reaching 90.8%;average precision mean（mAP）increased by 1.2%,reaching 93.4%;the model size reduced by 21.4 MB. The accuracy of improved YOLOV5 algorithm was 11.3,4.4,4.2,and 3.6 percentage points higher than SSD,YOLOv4,YOLOv6,and YOLOv7 models,respectively.【Conclusion】A convolutional neural network-based improved YOLOv5 apple leaf disease detection model was proposed. The improved YOLOv5 model had fast detection speed,high detection accuracy,and small size,which can achieve automatic recognition of apple leaf diseases.

Key words: YOLOv5, Apple, Leaf diseases, Identification, Convolutional neural network

中图分类号:

TP391.41

赵兴, 邬欢欢. 基于改进的YOLOv5苹果叶部病害识别研究[J]. 北方农业学报, 2024, 52(1): 125-134.

ZHAO Xing, WU Huanhuan. Research on apple leaf disease detection based on improved YOLOv5[J]. Journal of Northern Agriculture, 2024, 52(1): 125-134.

参考文献

[1] SRINIVASAN K,DASTOOR P H,RADHAKRISHNAIAH P,et al.FDAS:A knowledge-based framework for analysis of defects in woven textile structures[J].Journal of the Textile Institute,1992,83(3):431-448.
[2] 曾晏林,贺壹婷,蔺瑶,等.基于BCE-YOLOv5的苹果叶部病害检测方法[J].江苏农业科学,2023,51(15):155-163.
[3] 周绍发,肖小玲,刘忠意,等.改进的基于YOLOv5s苹果树叶病害检测[J].江苏农业科学,2023,51(13):212-220.
[4] 刘洋,冯全,王书志.基于轻量级CNN的植物病害识别方法及移动端应用[J].农业工程学报,2019,35(17):194-204.
[5] 张俊宁,毕泽洋,闫英,等.基于注意力机制与改进YOLO的温室番茄快速识别[J].农业机械学报,2023,54(5):236-243.
[6] 张小花,李浩林,李善军,等.基于EfficientDet-D1的草莓快速检测及分类[J].华中农业大学学报,2022,41(6):262-269.
[7] 芮修业. 复杂背景下叶片病斑分割与识别策略的研究[D].天津:天津理工大学,2019.
[8] HAN K,WANG Y H,TIAN Q.GhostNet:More features from cheap operations[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle:IEEE,2020:1580-1589.
[9] PRASHAR K,TALEAR R,KANT C.Robust automatic cotton crop disease recognition(ACDR)method using the hybrid feature descriptor with SVM[C]//2016 International Conference on Computing on sustainable Global Development.New Delhi:INDIACom,2017:1-3.
[10] 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.Las Vegas:IEEE,2016:770-778.
[11] HOWARD A G,ZHU M L,CHEN B,et al.MobileNets:Efficient convolutional neural networks for mobile vision applications[EB/OL].(2017-04-17).https://arxiv.org/abs/1704.04861.
[12] KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90.
[13] LIN T Y,DOLLAR P,GIRSHICK R,et al.Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition.Honolulu:IEEE,2017:2117-2125.
[14] LIU B,DING Z F,ZHANG Y,et al.Kiwifruit leaf disease identification using improved deep convolutional neural networks[C]//2020 IEEE Annual Computers,Software,and Applications Conference .Madrid:IEEE,2020:1267-1272.
[15] LIU S,QI L,QIN H F,et al.Path aggregation network for instance segmentation[C]//2018 IEEE Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:8759-8768.
[16] LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single Shot MultiBox Detector[M].Berlin:Springer International Publishing,2016.
[17] REDMON J,FARHADI A.YOLOv3:An incremental improvement[EB/OL](2018-04-08).https://arxiv.org/abs/1804.02767.
[18] HE K M,ZHANG X Y,REN S Q,et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1904-1916.
[19] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE,2016:779-788.
[20] REDMON J,FARHADI A.YOLO9000:Better,faster,stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE,2017:7263-7271.
[21] USHADEVI G,GOKULNATH B V.A survey on plant disease prediction using machine learning and deep learning techniques[J].Inteligencia Artificial,2020,23(65):136-154.
[22] REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:Towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.
[23] SUN H N,XU H W,LIU B,et al.MEAN-SSD:A novel real-time detector for apple leaf diseases using improved light-weight convolutional neural networks[J].Computers and Electronics in Agriculture,2021,189:106379.
[24] SETHY P K,BARPANDA N K,RATH A K,et al.Image processing techniques for diagnosing rice plant disease:A survey[J].Procedia Computer Science,2020,167:516-530.
[25] SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[EB/OL].(2015-04-10).https://arxiv.org/abs/1409.1556.
[26] SETHY P K,BRAPANDA N K,RATH A K.Detection and identification of rice leaf diseases using multiclass SVM and particle swarm optimization technique[J].International Journal of Innovative Technology and Exploring Engineering,2019,8(S2):108-120.
[27] SUN Y Y,JIANG Z H,ZHANG L P,et al.SLIC_SVM based leaf diseases saliency map extraction of tea plant[J].Computers and Electronics in Agriculture,2019,157:102-109.
[28] WOO S,PARK J,LEE J Y,et al.CBAM:Convolutional Block Attention Module[M].Berlin:Springer International Publishing,2018:3-19.
[29] TURKOGLU M,YANIKOGLU B,HANBAY D.PlantDiseaseNet:Convolutional neural network ensemble for plant disease and pest detection[J].Signal,Image and Video Processing,2022,16(2):301-309.
[30] WANG C Y,MARK LIAO H Y,WU Y H,et al.CSPNet:A new backbone that can enhance learning capability of CNN[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops .Seattle:IEEE,2020:390-391.
[31] WANG Y,WANG Y,ZHAO J B.MGA-YOLO:A lightweight one-stage network for apple leaf disease detection[J].Frontiers in Plant Science,2022,13:927424.
[32] SZEGEDY C,LIU W,JIA Y Q,et al.Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE,2015:1-9.
[33] Qi C,GAO J,Pearson S.Tea chrysanthemum detection under unstructured environments using the TC-YOLO model[J].Expert Systems with Applications,2022,193:116473.
[34] XIE X Y,MA Y A,LIU B,et al.A deep-learning-based real-time detector for grape leaf diseases using improved convolutional neural networks[J].Frontiers in Plant Science,2020,11:751.
[35] ZHU R L,ZOU H Y,LI Z Y,et al.Apple-Net:A model based on improved YOLOv5 to detect the apple leaf diseases[J].Plants,2022,12(1):169.
[36] Sharma N,Gupta S,Reshan M S A.EfficientNetB0 cum FPN based semantic segmentation of gastrointestinal tract organs in MRI scans[J].Diagnostics,2023,13(14):2399.
[37] 毕新杰. 机场净空区基于YOLO的无人机识别算法研究[D].天津:中国民航大学,2022.