中国农业期刊集成服务平台

当前位置: 首页 > 文章 > 基于one-shot学习的小样本植物病害识别江苏农业学报 2019,35 (5) 1061-1067

Position: Home > Articles > Plant disease identification for small sample based on one-shot learning Jiangsu Journal of Agricultural Sciences 2019,35 (5) 1061-1067

基于one-shot学习的小样本植物病害识别

作者：

任胜男;孙钰;张海燕;郭丽霞

单位：

国家食品安全风险评估中心;北京林业大学信息学院

关键词：

植物病害识别;深度学习;one-shot学习;焦点损失函数;关系网络

摘要：

针对植物病害小样本问题提出一种基于one-shot学习的植物病害识别方法.以公开数据集PlantVillage中8类样本数量较少的植物病害图像作为识别对象,使用焦点损失函数(focal loss,FL)训练基于关系网络的植物病害分类器.训练过程中,调整FL超参数使模型聚焦于困难样本,从而提高植物病害识别精确率.结果表明:该方法在5-way、1-shot任务中识别精确率达到89.90%,相比原始关系网络模型精确率提高了4.69个百分点.同时,与匹配网络和迁移学习相比,改进后的方法在实验数据集上识别精确率分别提高了25.02个百分点和41.90个百分点.

译名：

Plant disease identification for small sample based on one-shot learning

作者：

REN Sheng-nan;SUN Yu;ZHANG Hai-yan;GUO Li-xia;School of Information Science and Technology, Beijing Forestry University;China National Center for Food Safety Risk Assessment;

关键词：

plant disease identification;;deep learning;;one-shot learning;;focal loss;;relation network

摘要：

A plant disease identification method based on one-shot learning was proposed to solve the problem of small plant disease dataset. Using eight kinds of plant disease images with few samples in the public dataset PlantVillage as recognition objects, the disease classifier based on relation network was trained by using focal loss(FL). To improve the accuracy of plant disease identification, the FL hyper-parameters were adjusted to focus the model on the hard samples during the training process. The results showed that the recognition accuracy of this method was 89.90% in 5-way and 1-shot task, which was 4.69% higher than that of the original relation network model. Besides, compared with matching network and transfer learning, the accuracy of improved method was improved by 25.02% and 41.90% on experimental set.

相似文章

自动对靶施药系统中植物病害识别技术的研究 [冀荣华, 祁力钧, 傅泽田] 农业机械学报 2007,38 (6) 190-192
基于植物叶片图像的植物病害检测方法 [尚怡君, 张善文, 张云龙] 江苏农业科学 2014,42 (4) 340-342
基于特征融合与局部判别映射的苹果叶部病害识别方法 [李超, 彭进业, 张善文] 广东农业科学 2016,43 (10) 134-139+193
机器学习在植物病害识别研究中的应用 [王聃, 柴秀娟] 中国农机化学报 2019 (9) 171-180
面向边缘计算的轻量级植物病害识别模型 [王冠, 王建新, 孙钰] 浙江农林大学学报 2020 (5) 978-985

导出引用

计量

文章访问数: 9

HTML全文浏览量: 0

PDF下载量: 0

基于one-shot学习的小样本植物病害识别

相似文章

所属期刊

江苏农业学报

推荐期刊

江苏农业科学

浙江农林大学学报