Chinese herbal medicine recognition network based on knowledge distillation and cross-attention
Abstract In order to reduce the number of parameters in the Chinese herbal medicine recognition model while maintaining accuracy, this paper takes 20 classes of Chinese herbs as the research object and proposes a recognition network based on knowledge distillation and cross-attention – ShuffleCANet...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-025-85697-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841544793821282304 |
|---|---|
| author | Qinggang Hou Wanshuai Yang Guizhuang Liu |
| author_facet | Qinggang Hou Wanshuai Yang Guizhuang Liu |
| author_sort | Qinggang Hou |
| collection | DOAJ |
| description | Abstract In order to reduce the number of parameters in the Chinese herbal medicine recognition model while maintaining accuracy, this paper takes 20 classes of Chinese herbs as the research object and proposes a recognition network based on knowledge distillation and cross-attention – ShuffleCANet (ShuffleNet and Cross-Attention). Firstly, transfer learning was used for experiments on 20 classic networks, and DenseNet and RegNet were selected as dual teacher models. Then, considering the parameter count and recognition accuracy, ShuffleNet was determined as the student model, and a new cross-attention mechanism was proposed. This cross-attention model replaces Conv5 in ShuffleNet to achieve the goal of lightweight design while maintaining accuracy. Finally, experiments on the public dataset NB-TCM-CHM showed that the accuracy (ACC) and F1_score of the proposed ShuffleCANet model reached 98.8%, with only 128.66M model parameters. Compared with the baseline model ShuffleNet, the parameters are reduced by nearly 50%, but the accuracy is improved by about 1.3%, proving this method’s effectiveness. |
| format | Article |
| id | doaj-art-085329bfa6fc46f0bf689bc5b356bc49 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-085329bfa6fc46f0bf689bc5b356bc492025-01-12T12:15:26ZengNature PortfolioScientific Reports2045-23222025-01-0115111510.1038/s41598-025-85697-6Chinese herbal medicine recognition network based on knowledge distillation and cross-attentionQinggang Hou0Wanshuai Yang1Guizhuang Liu2School of Information Engineering, Shandong Huayu University of TechnologySchool of Information Engineering, Shandong Huayu University of TechnologySchool of Information Engineering, Shandong Huayu University of TechnologyAbstract In order to reduce the number of parameters in the Chinese herbal medicine recognition model while maintaining accuracy, this paper takes 20 classes of Chinese herbs as the research object and proposes a recognition network based on knowledge distillation and cross-attention – ShuffleCANet (ShuffleNet and Cross-Attention). Firstly, transfer learning was used for experiments on 20 classic networks, and DenseNet and RegNet were selected as dual teacher models. Then, considering the parameter count and recognition accuracy, ShuffleNet was determined as the student model, and a new cross-attention mechanism was proposed. This cross-attention model replaces Conv5 in ShuffleNet to achieve the goal of lightweight design while maintaining accuracy. Finally, experiments on the public dataset NB-TCM-CHM showed that the accuracy (ACC) and F1_score of the proposed ShuffleCANet model reached 98.8%, with only 128.66M model parameters. Compared with the baseline model ShuffleNet, the parameters are reduced by nearly 50%, but the accuracy is improved by about 1.3%, proving this method’s effectiveness.https://doi.org/10.1038/s41598-025-85697-6 |
| spellingShingle | Qinggang Hou Wanshuai Yang Guizhuang Liu Chinese herbal medicine recognition network based on knowledge distillation and cross-attention Scientific Reports |
| title | Chinese herbal medicine recognition network based on knowledge distillation and cross-attention |
| title_full | Chinese herbal medicine recognition network based on knowledge distillation and cross-attention |
| title_fullStr | Chinese herbal medicine recognition network based on knowledge distillation and cross-attention |
| title_full_unstemmed | Chinese herbal medicine recognition network based on knowledge distillation and cross-attention |
| title_short | Chinese herbal medicine recognition network based on knowledge distillation and cross-attention |
| title_sort | chinese herbal medicine recognition network based on knowledge distillation and cross attention |
| url | https://doi.org/10.1038/s41598-025-85697-6 |
| work_keys_str_mv | AT qingganghou chineseherbalmedicinerecognitionnetworkbasedonknowledgedistillationandcrossattention AT wanshuaiyang chineseherbalmedicinerecognitionnetworkbasedonknowledgedistillationandcrossattention AT guizhuangliu chineseherbalmedicinerecognitionnetworkbasedonknowledgedistillationandcrossattention |