CHENG Zhenzhen,ZHANG Bixiang,CHENG Yifan,et al.Grape Leaf Disease Recognition Method Based on SE-MobileNetV2[J].Northern Horticulture,2025,(5):131-140.[doi:10.11937/bfyy.20243329]
基于SE-MobileNetV2的葡萄叶部病害识别方法
- Title:
- Grape Leaf Disease Recognition Method Based on SE-MobileNetV2
- 文章编号:
- 1001-0009(2025)05-0131-10
- Keywords:
- grape; leaf disease; attention mechanism; image classification
- 分类号:
- TP 391.9;S 431.9
- 文献标志码:
- A
- 摘要:
- 以PlantVillage公开数据集的4种葡萄叶部病害为试材,提出了一种基于改进MobileNetV2模型的轻量化识别方法,该方法以轻量级MobileNetV2模型为基础,通过在模型瓶颈层中引入SE注意力机制,增强模型对关键特征的关注能力,从而进一步优化识别性能和减少模型参数数量,以期为实现病害的高精度诊断,同时有效降低计算资源需求提供参考依据。结果表明:改进后的模型在测试集上的识别准确率达97.5%,较原始MobileNetV2提升4.5%;与ResNet50、ResNet34和ShuffleNetV2模型相比,平均准确率分别提高10.2、18.7、28.2个百分点,且模型大小仅为20.7 MB,实现了模型运行成本和精确度的平衡,为葡萄叶部病害识别问题提供了解决方案。
- Abstract:
- Taking four grape leaf diseases in the open data set of Plant Village as the test materials,a lightweight recognition method was proposed based on an improved MobileNetV2 model,aiming to achieve high-accuracy disease diagnosis while effectively reducing computational resource requirements.The method utilizes the lightweight MobileNetV2 framework and integrates the SE attention mechanism into the bottleneck layers,enhancing the model′s ability to focus on key features and further optimizing recognition performance while reducing the number of model parameters,in order to provide reference for the realization of high precision disease diagnosis,while effectively reducing the need for computing resourees.The results showed that the improved model achieved a recognition accuracy of 97.5% on the test set,which was 4.5% higher than the original MobileNetV2.Compared to ResNet50,ResNet34,and ShuffleNetV2,the proposed method improved the average accuracy by 10.2,18.7,and 28.2 percentage points,respectively,with a model size of only 20.7 MB.This study successfully balances computational cost and recognition accuracy,providing an effective solution for the challenge of grape leaf disease recognition.
参考文献/References:
[1]王宇,杨畅,杨晓,等.5省市主栽葡萄的果实病害及防治调查[J].植物保护,2021,47(3):242-246.[2]李敏,闫成功,马帅,等.我国葡萄植保机械化技术发展现状与对策[J].中国农机化学报,2024,45(1):76-82.[3]蔺瑶,曾晏林,刘金涛,等.基于EBP-YOLOv8的葡萄叶病害检测与识别方法研究[J].山东农业大学学报(自然科学版),2024,55(3):322-334.[4]郑建华,朱立学,朱蓉.基于多特征融合与支持向量机的葡萄病害识别[J].现代农业装备,2018(6):54-60.[5]栗苗苗.葡萄种植产业现状与发展对策探讨[J].农业与技术,2020,40(11):110-111.[6]刘媛,冯全.葡萄病害的计算机识别方法[J].中国农机化学报,2017,38(4):99-104.[7]田有文,李天来,李成华,等.基于支持向量机的葡萄病害图像识别方法[J].农业工程学报,2007,23(6):175-180.[8]胡施威,邓建新,王浩宇,等.基于改进EfficientNetB0模型的葡萄叶部病害识别方法[J].现代电子技术,2024,47(15):73-80.[9]乔虹,冯全,张芮,等.基于时序图像跟踪的葡萄叶片病害动态监测[J].农业工程学报,2018,34(17):167-175.[10]王超学,祁昕,马罡,等.基于YOLO V3的葡萄病害人工智能识别系统[J].植物保护,2022,48(6):278-288.[11]樊湘鹏,许燕,周建平,等.基于迁移学习和改进CNN的葡萄叶部病害检测系统[J].农业工程学报,2021,37(6):151-159.[12]苏仕芳,乔焰,饶元.基于迁移学习的葡萄叶片病害识别及移动端应用[J].农业工程学报,2021,37(10):127-134.[13]何欣,李书琴,刘斌.基于多尺度残差神经网络的葡萄叶片病害识别[J].计算机工程,2021,47(5):285-291,300.[14]谢圣桥,宋健,汤修映,等.基于迁移学习和残差网络的葡萄叶部病害识别[J].农机化研究,2023,45(8):18-23,28.[15]张林鍹,巴音塔娜,曾庆松.基于StyleGAN2-ADA和改进YOLO v7的葡萄叶片早期病害检测方法[J].农业机械学报,2024,55(1):241-252.[16]SANDLER M,HOWARD A,ZHU M,et al.MobileNetV2:inverted residuals and linear bottlenecks[C].Salt Lake City:CVPR,2018.[17]DONG X,ZHANG Z,YU R,et al.Extraction of information about individual trees from high-spatial-resolution UAV-acquired images of an orchard[J].Remote Sensing,2020,12(1):133.[18]JIEH,LI S,GANG S,ALBANIE S.Squeeze-and-excitation networks[C].Salt Lake City:CVPR,2018.
相似文献/References:
[1]单守明,平吉成,刘晶,等.6-BA对染卷叶病毒葡萄光合作用和果实品质的影响[J].北方园艺,2013,37(04):5.
[2]梁浩.枣庄地区气候因子对葡萄生产模式选择的影响[J].北方园艺,2013,37(04):55.
[3]张广和,于晓丽,栾炳辉,等.倍创与杀菌剂混用对葡萄果实病害的防效评价[J].北方园艺,2013,37(04):127.
[4]李海峰,王瑞华,韩琛,等.新疆葡萄平衡施肥技术发展现状及展望[J].北方园艺,2013,37(04):185.
[5]史洪琴,蒋丽光.中短剪葡萄结实特性比较[J].北方园艺,2014,38(12):34.
SHI Hong-qin,JIANG Li-guang.Comparison of Grape Fruit Characteristics After Moderate and Short Pruning[J].Northern Horticulture,2014,38(5):34.
[6]张剑侠,吴行昶,杨亚州.葡萄种间杂种的抗寒性评价[J].北方园艺,2014,38(13):1.
ZHANG Jian-xia,WU Xing-chang,YANG Ya-zhou.Cold-resistance Evaluation in Grape Interspecific Hybrids[J].Northern Horticulture,2014,38(5):1.
[7]商佳胤,田淑芬,集 贤,等.果袋颜色对设施葡萄果实品质的影响[J].北方园艺,2014,38(11):42.
SHANG Jia-yin,TIAN Shu-fen,JI Xian,et al.Effect of Fruit Bag Color on Grape Quality in Facility Condition[J].Northern Horticulture,2014,38(5):42.
[8]王文平.肃州区非耕地日光温室葡萄反季节高效节水栽培技术[J].北方园艺,2014,38(10):43.
[J].Northern Horticulture,2014,38(5):43.
[9]周 恒,饶 鹏,田如英,等.不同栽培方式下五个葡萄品种在贵州铜仁地区的生产表现[J].北方园艺,2014,38(14):22.
ZHOU Heng,RAO Peng,TIAN Ru-ying,et al.Five Grape Varieties Under Different Cultivations in the Production Performance in Guizhou Tongren Area[J].Northern Horticulture,2014,38(5):22.
[10]吴久赟,郭峰,杨承时,等.不同叶果比对葡萄果实生长的影响[J].北方园艺,2014,38(04):33.
WU Jiu-yun,GUO Feng,YANG Cheng-shi,et al.Study on the Influence of Different Leaf-fruit Ratio on Growth of Grape[J].Northern Horticulture,2014,38(5):33.
备注/Memo
第一作者简介:程浈浈(1994-),女,博士,讲师,现主要从事图像处理与模式识别等研究工作。E-mail:2021190008@xyafu.edu.cn.责任作者:龚守富(1976-),男,博士,教授,现主要从事农业信息技术等研究工作。E-mail:15137638703@163.com.基金项目:河南省重点研发与推广专项(科技攻关)资助项目(232102111118)。收稿日期:2024-08-20