Efficient Commercial Classification of Agricultural Products using Convolutional Neural Networks
Bibliographic record
Abstract
Agricultural products, as essential commodities, are among the most sought-for items in superstores. Barcode is usually utilized to classify and regulate the price of products such as ornamental flowers in such stores. However, the use of barcodes on some fragile agricultural products such as ornamental flowers can be damaged and lessen their life length. Moreover, it is time-consuming and costly and may lead to the production of massive waste and damage to the environment and the admittance of chemical materials into food products that can affect human health. Consequently, we aimed to design a classifier robot to recognize ornamental flowers based on the related product image at different times and surrounding conditions. Besides, it can increase the speed and accuracy of distinguishing and classifying the products, lower the pricing time, and increase the lifetime due to the absence of the need for movement and changing the position of the products. According to the datasheets provided by the robot that is stored in its database, we provide the possibility of identifying and introducing the product in different colors and shapes. Also, due to the preparation of a standard and small database tailored to the needs of the robot, the robot will be trained in a short time (less than five minutes) without the need for an Internet connection or a large hard drive for storage the data. On the other hand, by dividing each input photo into ten different sections, the system can, without the need for a detection system, simultaneously in several different images, decorative flowers in different conditions, angles and environments, even with other objects such as vases, detects very fast with a high accuracy of 97%.
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 | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".