胡志伟, 杨华, 黄济民, 等. 基于注意力残差机制的细粒度番茄病害识别[J]. 华南农业大学学报, 2019, 40(6): 124-132. DOI: 10.7671/j.issn.1001-411X.201812048
    引用本文: 胡志伟, 杨华, 黄济民, 等. 基于注意力残差机制的细粒度番茄病害识别[J]. 华南农业大学学报, 2019, 40(6): 124-132. DOI: 10.7671/j.issn.1001-411X.201812048
    HU Zhiwei, YANG Hua, HUANG Jiming, et al. Fine-grained tomato disease recognition based on attention residual mechanism[J]. Journal of South China Agricultural University, 2019, 40(6): 124-132. DOI: 10.7671/j.issn.1001-411X.201812048
    Citation: HU Zhiwei, YANG Hua, HUANG Jiming, et al. Fine-grained tomato disease recognition based on attention residual mechanism[J]. Journal of South China Agricultural University, 2019, 40(6): 124-132. DOI: 10.7671/j.issn.1001-411X.201812048

    基于注意力残差机制的细粒度番茄病害识别

    Fine-grained tomato disease recognition based on attention residual mechanism

    • 摘要:
      目的  解决温室环境下细粒度番茄病害识别方法不足问题。
      方法  以早、晚期5种番茄病害叶片为研究对象,提出一种基于注意力与残差思想相结合的新型卷积神经网络模型ARNet。通过引入多层注意力模块,层次化抽取病害分类信息,解决早期病害部位分散、特征难以提取难题;为避免网络训练出现退化现象,构建残差模块有效融合高低阶特征,同时引入数据扩充技术以防止模型过拟合。
      结果  对44 295张早、晚期病害叶片数据集进行模型训练与测试的结果表明,与VGG16等现有模型相比,ARNet具有更好的分类表现,其平均识别准确率达到88.2%,显著高于其他模型。ARNet对早期病害识别准确率明显优于晚期病害,验证了注意力机制在提取细微区域特征上的有效性,且在训练过程中未发生过度抖动的状况。
      结论  本文提出的模型具有较强鲁棒性和较高稳定性,在实际应用中可为细粒度番茄病害智能诊断提供参考。

       

      Abstract:
      Objective  To solve the insufficient identification of fine-grained tomato diseases in greenhouse.
      Method  Taking tomato leaves with five early or late diseases as research objects, we proposed a new convolutional neural network model ARNet based on the combination of attention and residual thought. A multi-layered attention module was introduced to solve the problem of early disease location dispersion and the difficulty of feature extraction by extracting hierarchically disease classification information. In order to avoid the degradation of network training, we constructed a residual module to effectively integrate high- and low-order features. Meantime, we introduced the data expansion technology to prevent model over-fitting.
      Result  Model training and testing results of early and late disease leaf datasets with 44 295 pictures showed that ARNet has better classification performance with an average recognition accuracy of 88.2%, which was significantly higher than those of other existing models. In addition, the identification accuracy of ARNet for early disease was significantly better than that for late disease, which verified the effectiveness of attention mechanism in extracting fine region features, and there was no excessive jitter during training process.
      Conclusion  This model proposed in this paper has strong robustness and high stability, and can provide a reference for intelligent diagnosis of fine-grained tomato diseases in practical application.

       

    /

    返回文章
    返回