采用改进的EfficientNet识别苹果叶片病害  

Identifying apple leaf diseases using improved EfficientNet

作　　者：王瑞鹏 陈锋军[1,2,3] 朱学岩 张新伟 WANG Ruipeng;CHEN Fengjun;ZHU Xueyan;ZHANG Xinwei(School of Technology,Beijing Forestry University,Beijing 100083,China;National Key Laboratory-Forest Resource Efficient Production,Beijing 100083,China;Beijing Laboratory of Urban and Rural Ecological Environment,Beijing 100083,China;Key Laboratory of State Forestry Administration for Forestry Equipment and Automation,Beijing,100083,China;Research Center for Intelligent Forestry,Beijing 100083,China)

机构地区：[1]北京林业大学工学院,北京100083 [2]林木资源高效生产全国重点实验室,北京100083 [3]城乡生态环境北京实验室,北京100083 [4]林业装备与自动化国家林业局重点实验室,北京100083 [5]智慧林业研究中心,北京100083

出　　处：《农业工程学报》2023年第18期201-210,共10页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家重点研发计划(2019YFD1002401);北京林业大学科技创新计划项目(2021ZY74);北京市共建项目专项。

摘　　要：该研究针对当前自然环境下的苹果叶片病害识别中病害病斑小、空间分布特征不同以及特征相近病害识别困难的问题,设计DEFL (DenseNet121+EfficientNet with focal loss and label smoothing)模型。首先,该模型以并行的EfficientNet-B0网络和DenseNet121网络为特征提取网络,以提升模型特征提取能力,其次引入结合标签平滑策略的焦点损失函数以加强模型对识别困难样本的关注。经测试,所提模型的识别准确率为99.13%,平均精度均值为98.47%。消融试验表明两项改进分别使模型平均精度均值提高了7.99和3.15个百分点。对比试验结果表明,DEFL模型平均精度均值较于ResNet50、Inception V3、ResNeXt模型以及分别融合这3种模型的EfficientNet-B0模型分别高出14.53、13.17、14.61、 6.4、 7.71以及8.91个百分点,模型规模分别小18.73、 7.7、 12.2、 83.62、 69.6以及60.09MB。Grad-CAM(gradient-weighted class activation mapping)热力图可视化结果表明所提模型重点关注了叶片病变区域。UMAP(uniform manifold approximation and projection)特征降维可视化结果表明所提模型提取的特征更具区分度。实际应用验证取得了97.73%的总体准确率以及95.82%的平均精度均值。综上,该研究提出的DEFL模型能够为苹果病害防治提供有效参考。The traditional manual control measures for apple disease primarily rely on subjective experience,which can lead to arbitrariness and bias.This approach is also prone to pesticide waste and misuse.As apple leaves are one of the high-incidence areas of diseases,achieving natural environment-based identification of apple leaf diseases can provide effective guidance for disease prevention and control.To address the challenge of accurate identification of apple leaf diseases in natural settings,a DenseNet121+EfficientNet model with Focal Loss and Label Smoothing(DEFL)is designed.Given that apple leaf disease lesions are small and exhibit varying spatial characteristics for different diseases,accurately localizing and identifying disease regions is challenging.The DEFL model utilizes two parallel networks:EfficientNet-B0 and DenseNet121 for feature extraction.It combines the semantic and positional information extracted from these networks to enhance the model's finegrained feature extraction capability.To overcome difficulties in identifying apple leaf samples due to the similarity in features among different diseases,a Focal Loss function,combined with label smoothing strategies,is introduced.This prevents the model from being overly confident about the class of apple leaf samples,focusing on those samples with challenging decision boundaries during training,thereby improving the model's ability to recognize boundary cases and enhancing its robustness and generalization.The DEFL model proposed in this study was tested with 3906 apple leaf disease images in natural scenes.The results show that the overall recognition accuracy of the model proposed in this study is 99.13%,and the mean average accuracy is 98.47%.The results of ablation experiments show that the mean average precision of the model is improved by 7.99%after adding DenseNet121 feature extraction branch to EfficientNet-B0.And,after introducing the focus loss function combined with label smoothing strategy to EfficientNet-B0,the mean average precision of th

关 键 词：病害 图像识别 苹果叶片 EfficientNet DenseNet121 

分 类 号：S24[农业科学—农业电气化与自动化]