摘要
传统的植物叶片分类方法往往难以满足准确性和效率性的要求,本研究引入了VGG16模型作为改进的解决方案,旨在提高药用植物叶片分类的准确性和自动化程度。以基准CNN(Convolutional Neural Networks)模型在数据集上进行评估,用VGG16模型与基准CNN模型比较效果。VGG16模型在训练集上达到了97%的准确率,而验证集的准确率为94%。在相同的训练周期下,训练集和验证集的准确率分别为91.1%和93.4%。这表明VGG16模型在对药用植物叶片进行分类时具有更好的性能和泛化能力。VGG16模型在药用植物叶片分类任务中展现出了优异的性能,为高效精准的植物分类提供了有力的解决方案。未来的研究可以进一步改进和扩展深度学习模型,以应对更广泛和复杂的植物分类挑战。
Traditional plant leaf classification methods often fail to meet the requirements of accuracy and efficiency.This study introduces the VGG16 model as an improved solution,aiming to improve the accuracy and automation of medicinal plant leaf classification.,and evaluates the benchmark CNN(Convolutional Neural Networks)model on the dataset and compare the effectiveness with the VGG16 model and the benchmark CNN model.The VGG16 model achieved an accuracy of 97%on the training set,while the accuracy of the validation set was 94%.Under the same training cycle,the accuracy of the training and validation sets is 91.1%and 93.4%,respectively.This indicates that the VGG16 model has better performance and generalization ability in classifying medicinal plant leaves.The VGG16 model exhibits excellent performance in the task of medicinal plant leaf classification,providing a powerful solution for efficient and accurate plant classification.Future research can further improve and expand deep learning models to address broader and complex plant classification challenges.
作者
刘艺
孙延斌
翟凤国
梁新
LIU Yi;SUN Yan-bin;ZHAI Feng-guo;LIANG Xin(Mudanjiang Medical College,Mudanjiang,Heilongjiang 157011,China;Affiliated Hongqi Hospital,Mudanjiang Medical College,Mudanjiang,Heilongjiang 157011,China)
出处
《新一代信息技术》
2023年第13期6-11,共6页
New Generation of Information Technology
基金
2021年度黑龙江省省属高校基本业务费科研项目(No.2021-KYYWF-0486)