[1]潘赟,陈松楠,马俊.基于多尺度特征融合Transformer的苹果叶片病害识别模型[J].江苏农业科学,2025,53(15):212-219.
 Pan Yun,et al.Apple leaf disease recognition model based on multiscale feature fusion Transformer[J].Jiangsu Agricultural Sciences,2025,53(15):212-219.
点击复制

基于多尺度特征融合Transformer的苹果叶片病害识别模型()

《江苏农业科学》[ISSN:1002-1302/CN:32-1214/S]

卷:
第53卷
期数:
2025年第15期
页码:
212-219
栏目:
农业工程与信息技术
出版日期:
2025-08-05

文章信息/Info

Title:
Apple leaf disease recognition model based on multiscale feature fusion Transformer
作者:
潘赟1陈松楠2马俊1
1.信阳农林学院信息工程学院,河南信阳 464000; 2.武汉轻工大学数学与计算机学院,湖北武汉 430048
Author(s):
Pan Yunet al
关键词:
苹果叶片病害识别多尺度特征融合Transformer空洞卷积残差连接无锚框机制
Keywords:
-
分类号:
S126;TP391.41
DOI:
-
文献标志码:
A
摘要:
为了应对苹果叶片病害信息稀疏、特征表达能力不足导致现有识别算法出现识别精度低、漏检严重等问题,提出了一种基于多尺度特征融合Transformer的苹果叶片病害识别模型。首先,设计1个多尺度特征融合Transformer模型,该模型通过引入多尺度特征融合网络,有效提取和融合不同尺度的特征信息,增强对病害细节的捕捉能力;其次,利用空洞卷积和残差连接构建特征增强模块,进一步提高模型对病害特征的表达能力;最后,引入无锚框机制,减少病害区域的漏检问题,提高识别精度。在公开的苹果叶片病害数据集上的试验结果表明,与现有主流病害识别算法相比,本研究提出的模型在整体识别精度、召回率和F1分数上均有显著提升。具体而言,本研究提出模型的识别精度达93.2%、召回率为92.5%,F1分数为92.8%,均优于ResNet34、ResNet101、VGG16、VGG19、MobileNet v2和Swin Transformer等主流算法。此外,模型在保持较高识别准确率的同时,平均识别时间仅为0.12 s/张,展现出其较快的识别速度。试验结果验证了该模型在苹果叶片病害识别任务中的有效性与鲁棒性。
Abstract:
-

参考文献/References:

[1]Chen Y P,Pan J C,Wu Q F. Apple leaf disease identification via improved CycleGAN and convolutional neural network[J]. Soft Computing,2023,27(14):9773-9786.
[2]张风伟,朱成杰,朱洪波. 基于改进MobileNet v3的苹果叶片病害识别方法及移动端应用[J]. 江苏农业科学,2024,52(7):205-213.
[3]李云红,张蕾涛,李丽敏,等. 基于CycleGAN-IA方法和 M-ConvNext 网络的苹果叶片病害图像识别[J]. 农业机械学报,2024,55(4):204-212.
[4]Gao Y X,Cao Z Z,Cai W W,et al. Apple leaf disease identification in complex background based on BAM-net[J]. Agronomy,2023,13(5):1240.
[5]Chen C,Yu X,Gong Q. Apple leaf diseases identification based on improved residual network[J]. Journal of Henan Agricultural Sciences,2023,52(4):152-161.
[6]吴刚正,蔡成岗,朱瑞瑜. 基于注意力机制和残差网络的苹果叶片病害分类[J]. 江苏农业科学,2023,51(18):177-185.
[7]孙长兰,林海峰. 一种基于集成学习的苹果叶片病害检测方法[J]. 江苏农业科学,2022,50(20):41-47.
[8]李豫晋,沈陆明,何少芳,等. 基于改进MobileNet v3的苹果叶片病害识别研究[J]. 江苏农业科学,2024,52(12):224-231.
[9]郑金松,谷海红,蒋庆刚,等. 基于局部特征和视觉词袋模型的大豆叶片病害识别[J]. 中国农机化学报,2024,45(8):204-209.
[10]Zheng J Y,Li K Y,Wu W B,et al. RepDI:a light-weight CPU network for apple leaf disease identification[J]. Computers and Electronics in Agriculture,2023,212:108122.
[11]郑超杰,李少波,蒲睿强,等. 基于轻量化卷积神经网络的番茄叶片病害识别[J]. 江苏农业科学,2024,52(11):225-231.
[12]郑果,姜玉松. 基于多任务学习农作物叶片病害诊断方法[J]. 中国农业科技导报,2024,26(1):89-98.
[13]李云红,张蕾涛,谢蓉蓉,等. 基于AT-DenseNet网络的番茄叶片病害识别方法[J]. 江苏农业科学,2023,51(21):209-217.
[14]王娜,陈勇,崔艳荣,等. 基于改进轻量化YOLO v5n的番茄叶片病害识别方法[J]. 江苏农业科学,2024,52(8):192-199.
[15]李仁杰,宋涛,高婕,等. 基于改进YOLOv5的自然环境下番茄患病叶片检测模型[J]. 江苏农业学报,2024,40(6):1028-1037.
[16]Liu S,BAI H,LI F,et al. An apple leaf disease identification model for safeguarding apple food safety[J]. Food Science and Technology,2023,43:e104.
[17]Zhang Y K,Zhou G X,Chen A B,et al. A precise apple leaf diseases detection using BCTNet under unconstrained environments[J]. Computers and Electronics in Agriculture,2023,212:108132.
[18]徐艳蕾,孔朔琳,陈清源,等. 基于Transformer的强泛化苹果叶片病害识别模型[J]. 农业工程学报,2022,38(16):198-206.
[19]黄炜,王娟娟,殷学丽. 基于特征分离的小样本苹果病害叶片检测[J]. 江苏农业科学,2023,51(23):195-202.
[20]王瑞鹏,陈锋军,朱学岩,等. 采用改进的EfficientNet识别苹果叶片病害[J]. 农业工程学报,2023,39(18):201-210.
[21]Tang L,Yi J Z,Li X Y. Improved multi-scale inverse bottleneck residual network based on triplet parallel attention for apple leaf disease identification[J]. Journal of Integrative Agriculture,2024,23(3):901-922.
[22]Thapa R,Zhang K,Snavely N,et al. Plant pathology 2021-FGVC8 [N/OL]. (2021-03-16) [2024-08-01]. https://kaggle.com/competitions/plant-pathology-2021-fgvc8.
[23]张会敏,谢泽奇. 基于多尺度注意力卷积神经网络的苹果叶部病害识别方法[J]. 江苏农业科学,2023,51(16):154-161.
[24]许成果,薛月菊,郑婵,等. 基于自注意力机制与无锚点的仔猪姿态识别[J]. 农业工程学报,2022,38(14):166-173.

相似文献/References:

[1]徐凯宏,米雅婷,谷志新.基于GA-BP神经网络的温室番茄病害诊断[J].江苏农业科学,2016,44(04):387.
 Xu Kaihong,et al.Diagnosis of tomato disease in greenhouse based on GA-BP network[J].Jiangsu Agricultural Sciences,2016,44(15):387.
[2]张会敏,张云龙,张善文,等.基于区分矩阵的属性约简算法的作物病害识别方法[J].江苏农业科学,2015,43(01):387.
 Zhang Huimin,et al.A crop disease recognition method based on attribute reduction of discernibility matrix[J].Jiangsu Agricultural Sciences,2015,43(15):387.
[3]张云龙,袁浩,张晴晴,等.基于颜色特征和差直方图的苹果叶部病害识别方法[J].江苏农业科学,2017,45(14):171.
 Zhang Yunlong,et al.Apple leaf disease recognition based on color characteristics and differential histogram[J].Jiangsu Agricultural Sciences,2017,45(15):171.
[4]刁智华,刁春迎,魏玉泉,等.机器人系统中小麦病害识别与施药算法研究[J].江苏农业科学,2017,45(17):192.
 Diao Zhihua,et al.Study on wheat disease identification and spraying algorithm in robot system[J].Jiangsu Agricultural Sciences,2017,45(15):192.
[5]裴浩杰,冯海宽,李长春,等.基于多元线性回归和随机森林的苹果叶绿素含量高光谱估测方法比较[J].江苏农业科学,2018,46(17):224.
 Pei Haojie,et al.Comparison of hyperspectral estimation methods for chlorophyll contents of apple based on multiple linear regression and random forest[J].Jiangsu Agricultural Sciences,2018,46(15):224.
[6]刁智华,袁万宾,刁春迎,等.病害特征在作物病害识别中的应用研究综述[J].江苏农业科学,2019,47(05):71.
 Diao Zhihua,et al.Application of disease characteristics in crop disease identification:a review[J].Jiangsu Agricultural Sciences,2019,47(15):71.
[7]林彬彬,邱新法,何永健,等.茶树病害智能诊断识别算法研究[J].江苏农业科学,2019,47(06):85.
 Lin Binbin,et al.Study on intelligent diagnosis and recognition algorithm for tea diseases[J].Jiangsu Agricultural Sciences,2019,47(15):85.
[8]何前,郭峰林,方皓正,等.基于改进LeNet-5模型的玉米病害识别[J].江苏农业科学,2022,50(20):35.
 He Qian,et al.Study on maize disease recognition based on improved LeNet-5 model[J].Jiangsu Agricultural Sciences,2022,50(15):35.
[9]孙长兰,林海峰.一种基于集成学习的苹果叶片病害检测方法[J].江苏农业科学,2022,50(20):41.
 Sun Changlan,et al.An apple tree leaf disease detection method based on ensemble learning[J].Jiangsu Agricultural Sciences,2022,50(15):41.
[10]王慧,李康顺,蔡铁,等.基于约束性多目标优化算法的柑橘黄龙病识别算法[J].江苏农业科学,2023,51(6):159.
 Wang Hui,et al.Citrus Huanglongbing recognition algorithm based on constrained multi-objective optimization algorithm[J].Jiangsu Agricultural Sciences,2023,51(15):159.
[11]李豫晋,沈陆明,何少芳,等.基于改进MobileNet v3的苹果叶片病害识别研究[J].江苏农业科学,2024,52(12):224.
 Li Yujin,et al.Identification of apple leaf diseases based on improved MobileNet v3[J].Jiangsu Agricultural Sciences,2024,52(15):224.

备注/Memo

备注/Memo:
收稿日期:2024-08-01
基金项目:河南省科技攻关项目(编号:242102210059)。
作者简介:潘赟(1982—),女,河南信阳人,硕士,副教授,研究方向为网络安全和人工智能。E-mail:fly1230801@163.com。
通信作者:陈松楠,博士,讲师,研究方向为计算机视觉。 E-mail:chensongnan1988@163.com。
更新日期/Last Update: 2025-08-05