Deep Learning Based Peach Quality and Disease Detection Methods for Research in Intelligent Food Safety Detection
Bibliographic record
Abstract
With the rapid development of China's economic level and the significant improvement of people's living standard, the quality issue of peaches has become more and more strict.In this paper, based on deep learning algorithm, we propose the recognition method of peach fruit color, size and fruit shape features, combined with near-infrared spectroscopy detection technology, to quantify the peach fruit components and discriminate its maturity.Differential algorithm, standard normal transform, and multiple scattering correction are applied to pre-process peach fruit data.Based on M-YOLOv5s target detection framework, spectral analysis and image characterization techniques were used to jointly detect the degree of peach fruit disease.The distribution of peach fruit quality parameters was investigated, and the test results showed that 39.19% of the samples with measured values of fruit size were concentrated at 1.60-6.40cm, and 61.79% of the samples with predicted values were concentrated at 2.50-7.50cm, which was located at around the mean value of 4.763 cm.The classification accuracies of the information modeling set and validation set for the combination of the spectral analysis and image eigenvalue detection techniques were 91.439% and 88.487%, respectively, and the combined use of the two techniques had a high accuracy for the differentiation of diseased peach fruits.Based on the experimental results, the application of spectral detection technology in food freshness detection as well as pesticide residues and illegal additives is explored.
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) | 0.001 | 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 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".