Research on micro-morphological feature recognition method of traditional Chinese medicinal herbs based on deep convolutional neural network
Bibliographic record
Abstract
Based on the excellent achievements of deep learning technology in image recognition in several fields, then the convolutional neural network is expected to play its superior performance in the recognition of micro features of Chinese herbal medicines.The object of analysis in this paper is the microscopic feature images of Chinese herbal medicines, and the residual network will be improved in order to improve the recognition accuracy of the network model on the microscopic images of herbs.On the basis of the traditional CNN network model, CBAM based on mixed domain attention mechanism is added, and residual connection is introduced to increase the transfer of gradient and information flow, preserve image feature data and reduce feature loss.Improved from the traditional residual structure to moving inverted bottleneck convolution (MBConv), the SE module and SAM module are added to the MBConv stage respectively to optimize the feature extraction performance and improve the accuracy of the classification of microscopic features of traditional Chinese medicine.The effect of the addition of the attention mechanism on the network model is analyzed, and the network model is examined in conjunction with the constructed dataset of powdered microscopic images of commonly used Chinese medicinal herbs.The average accuracy of the Attention-TCM-Net network model on the test set reaches 96.47%, which is an improvement of 0.85 percentage points compared with that of the ResNet34 network, and meanwhile, the convergence of the model is significantly better than other models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".