摘要
苹果叶部病虫害是影响苹果产量增长的重要因素之一,准确地识别叶病种类以进行及时防治对于苹果增量增产具有重要的意义。本研究提出一种基于深度迁移学习的苹果叶病种类识别方法,提高了模型的泛化性和识别准确率。同时针对数据分布不均衡的问题,改进损失函数,使得模型获得了更高的识别效果。实验结果表明本研究算法进行叶病识别准确率可以达到93.5%以上,验证了本研究算法的有效性,对于苹果叶部病虫害的识别及防治提供了新思路。
Apple leaf diseases and insect pests are one of the important factors affecting the growth of apple production.Accurately identifying the types of leaf diseases for timely control is of great significance for increasing apple production.This research proposes an apple leaf disease type recognition method based on deep transfer learning,which improves the generalization and recognition accuracy of the model.At the same time,aiming at the problem of unbalanced data distribution,the loss function is improved to make the model obtain a higher recognition effect.The experimental results show that the accuracy of the research algorithm for leaf disease identification can reach more than 93.5%,which verifies the effectiveness of the research algorithm and provides a new idea for the identification and control of apple leaf diseases and insect pests.
出处
《智慧农业导刊》
2021年第9期14-17,共4页
JOURNAL OF SMART AGRICULTURE
基金
泰山学院引进人才科研启动基金项目(编号:Y-01-2018006)资助。
关键词
卷积神经网络
深度学习
迁移学习
苹果叶病识别
convolutional neural networks
deep learning
transfer learning
apple leaf disease recognition