Fresh and Rotten Fruits and Vegetables Classification using Hybrid ResNet-18–ViT Model
Bibliographic record
Abstract
Accurate identification of spoiled versus fresh fruits and vegetables is important for quality control in agriculture. Traditional manual inspection is slow and inconsistent. This work compares a standard ResNet-18 model against a hybrid ResNet-18–Vision Transformer ResNet-18 ViT model for classifying fresh and rotten fruit and vegetable images. This work used a ResNet-18 backbone for feature extraction from the image, and the pre-trained ViT-B16 is used for the classification. Both the models are trained, tested, and validated on the Fresh and Rotten Classification dataset. The model is evaluated with the following performance metrics, such as accuracy, precision, recall, F1-score, and ROC-AUC for training, testing, and validation datasets. The ResNet-18 model achieves 98.77 percent training accuracy, 97.76 percent testing accuracy, and 98.40 percent validation accuracy. While the ResNet-18 ViT model achieves 99.40 percent training accuracy, 99.26 percent testing accuracy, and 99.37 percent validation accuracy. Our hybrid model consistently outperforms the standard ResNet-18, demonstrating that combining convolutional feature extractors with transformer-based global attention can improve classification of fruit and vegetable freshness. These results suggest that hybrid ResNet-18 transformer architectures are promising for image-based food quality assessment.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".