Scholarlink

<p>With the increasing demand for herbal medicine in recent years, the issue of adulteration has posed severe threats to public health and business reputation. The traditional classification of Chinese herbal medicines mainly relies on the manual classification of experienced experts, which is tedious and time-consuming. In view of the above problems, this study proposes a classification method for Chinese herbal medicines based on a convolutional neural network. The RGB images of six similar Chinese herbal medicines were collected by low-cost equipment and imported into the Res-CANet network for training after preprocessing. The Res-CANet network is based on the ResNet18 network and combines the attention mechanism with the residual network. The classifier of the network is improved to further improve the recognition accuracy. The experimental results show that the average recognition accuracy of the Res-CANet network is 99.10%, which is 2.96% higher than that of the original ResNet18 network. This study verifies the feasibility of deep learning algorithms in the classification of medicinal materials, provides a good solution for the rapid identification and classification of Chinese herbal medicines.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Classification of Chinese Medicinal Materials Based on Residual Network and Attention Mechanism

  • 期刊论文

  • 发表日期：

    2026年1月2日

查看全文
  • Bing Liu

摘要

With the increasing demand for herbal medicine in recent years, the issue of adulteration has posed severe threats to public health and business reputation. The traditional classification of Chinese herbal medicines mainly relies on the manual classification of experienced experts, which is tedious and time-consuming. In view of the above problems, this study proposes a classification method for Chinese herbal medicines based on a convolutional neural network. The RGB images of six similar Chinese herbal medicines were collected by low-cost equipment and imported into the Res-CANet network for training after preprocessing. The Res-CANet network is based on the ResNet18 network and combines the attention mechanism with the residual network. The classifier of the network is improved to further improve the recognition accuracy. The experimental results show that the average recognition accuracy of the Res-CANet network is 99.10%, which is 2.96% higher than that of the original ResNet18 network. This study verifies the feasibility of deep learning algorithms in the classification of medicinal materials, provides a good solution for the rapid identification and classification of Chinese herbal medicines.