当前位置: X-MOL 学术Comput. Electron. Agric. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Few-Shot Learning approach for plant disease classification using images taken in the field
Computers and Electronics in Agriculture ( IF 8.3 ) Pub Date : 2020-08-01 , DOI: 10.1016/j.compag.2020.105542
David Argüeso , Artzai Picon , Unai Irusta , Alfonso Medela , Miguel G San-Emeterio , Arantza Bereciartua , Aitor Alvarez-Gila

Abstract Prompt plant disease detection is critical to prevent plagues and to mitigate their effects on crops. The most accurate automatic algorithms for plant disease identification using plant field images are based on deep learning. These methods require the acquisition and annotation of large image datasets, which is frequently technically or economically unfeasible. This study introduces Few-Shot Learning (FSL) algorithms for plant leaf classification using deep learning with small datasets. For the study 54,303 labeled images from the PlantVillage dataset were used, comprising 38 plant leaf and/or disease types (classes). The data was split into a source (32 classes) and a target (6 classes) domain. The Inception V3 network was fine-tuned in the source domain to learn general plant leaf characteristics. This knowledge was transferred to the target domain to learn new leaf types from few images. FSL using Siamese networks and Triplet loss was used and compared to classical fine-tuning transfer learning. The source and target domain sets were split into a training set (80%) to develop the methods and a test set (20%) to obtain the results. Algorithm performance was evaluated using the total accuracy, and the precision and recall per class. For the FSL experiments the algorithms were trained with different numbers of images per class and the experiments were repeated 20 times to statistically characterize the results. The accuracy in the source domain was 91.4% (32 classes), with a median precision/recall per class of 93.8%/92.6%. The accuracy in the target domain was 94.0% (6 classes) learning from all the training data, and the median accuracy (90% confidence interval) learning from 1 image per class was 55.5 (46.0–61.7)%. Median accuracies of 80.0 (76.4–86.5)% and 90.0 (86.1–94.2)% were reached for 15 and 80 images per class, yielding a reduction of 89.1% (80 images/class) in the training dataset with only a 4-point loss in accuracy. The FSL method outperformed the classical fine tuning transfer learning which had accuracies of 18.0 (16.0–24.0)% and 72.0 (68.0–77.3)% for 1 and 80 images per class, respectively. It is possible to learn new plant leaf and disease types with very small datasets using deep learning Siamese networks with Triplet loss, achieving almost a 90% reduction in training data needs and outperforming classical learning techniques for small training sets.

中文翻译:

使用现场拍摄的图像进行植物病害分类的少样本学习方法

摘要 及时检测植物病害对于预防瘟疫和减轻其对作物的影响至关重要。使用植物田间图像进行植物病害识别的最准确自动算法基于深度学习。这些方法需要获取和注释大型图像数据集,这在技术上或经济上通常是不可行的。本研究介绍了使用小数据集的深度学习进行植物叶片分类的少镜头学习 (FSL) 算法。该研究使用了来自 PlantVillage 数据集的 54,303 个标记图像,包括 38 种植物叶片和/或疾病类型(类别)。数据被分成源(32 个类)和目标(6 个类)域。Inception V3 网络在源域中进行了微调,以学习一般植物叶片特征。这些知识被转移到目标域,以从少量图像中学习新的叶子类型。使用 Siamese 网络和 Triplet loss 的 FSL 与经典的微调迁移学习进行了比较。源域集和目标域集被分成训练集 (80%) 以开发方法和测试集 (20%) 以获得结果。使用总准确率以及每个类别的准确率和召回率来评估算法性能。对于 FSL 实验,算法使用每类不同数量的图像进行训练,并且实验重复 20 次以统计表征结果。源域中的准确率为 91.4%(32 个类别),每个类别的准确率/召回率中位​​数为 93.8%/92.6%。从所有训练数据中学习到的目标域的准确率为 94.0%(6 类),从每类 1 张图像中学习的中位数准确度(90% 置信区间)为 55.5 (46.0–61.7)%。每类 15 张和 80 张图像的中位数准确率分别为 80.0 (76.4–86.5)% 和 90.0 (86.1–94.2)%,在训练数据集中仅用 4 个点就减少了 89.1%(80 张图像/类)准确性的损失。FSL 方法优于经典的微调迁移学习,对于每类 1 和 80 张图像,其准确率分别为 18.0 (16.0–24.0)% 和 72.0 (68.0–77.3)%。使用具有 Triplet 损失的深度学习 Siamese 网络,可以通过非常小的数据集学习新的植物叶片和疾病类型,从而将训练数据需求减少近 90%,并且在小型训练集上优于经典学习技术。每类 15 张和 80 张图像的中位数准确率分别为 80.0 (76.4–86.5)% 和 90.0 (86.1–94.2)%,在训练数据集中仅用 4 个点就减少了 89.1%(80 张图像/类)准确性的损失。FSL 方法优于经典的微调迁移学习,对于每类 1 和 80 张图像,其准确率分别为 18.0 (16.0–24.0)% 和 72.0 (68.0–77.3)%。使用具有 Triplet 损失的深度学习 Siamese 网络,可以通过非常小的数据集学习新的植物叶片和疾病类型,从而将训练数据需求减少近 90%,并且在小型训练集上优于经典学习技术。每类 15 张和 80 张图像的中位数准确率分别为 80.0 (76.4–86.5)% 和 90.0 (86.1–94.2)%,在训练数据集中仅用 4 个点就减少了 89.1%(80 张图像/类)准确性的损失。FSL 方法优于经典的微调迁移学习,对于每类 1 和 80 张图像,其准确率分别为 18.0 (16.0–24.0)% 和 72.0 (68.0–77.3)%。使用具有 Triplet 损失的深度学习 Siamese 网络,可以通过非常小的数据集学习新的植物叶片和疾病类型,从而将训练数据需求减少近 90%,并且在小型训练集上优于经典学习技术。训练数据集中 1%(80 张图像/类别),准确率仅损失 4 点。FSL 方法优于经典的微调迁移学习,对于每类 1 和 80 张图像,其准确率分别为 18.0 (16.0–24.0)% 和 72.0 (68.0–77.3)%。使用具有 Triplet 损失的深度学习 Siamese 网络,可以通过非常小的数据集学习新的植物叶片和疾病类型,从而将训练数据需求减少近 90%,并且在小型训练集上优于经典学习技术。训练数据集中 1%(80 张图像/类别),准确率仅损失 4 点。FSL 方法优于经典的微调迁移学习,对于每类 1 和 80 张图像,其准确率分别为 18.0 (16.0–24.0)% 和 72.0 (68.0–77.3)%。使用具有 Triplet 损失的深度学习 Siamese 网络,可以通过非常小的数据集学习新的植物叶片和疾病类型,从而将训练数据需求减少近 90%,并且在小型训练集上优于经典学习技术。
更新日期:2020-08-01
down
wechat
bug