| Autor: |
Mahmoud Al-Ayyoub, Fumie Costen, Mahmoud Ghandour, Mohammad A. Alsmirat, Raffi Al-Qurran, Ali Shatnawi |
| Rok vydání: |
2021 |
| Předmět: |
|
| Zdroj: |
2021 12th International Conference on Information and Communication Systems (ICICS). |
| DOI: |
10.1109/icics52457.2021.9464577 |
| Popis: |
Obtaining the highest quality olive oil (OO) during the milling process is greatly desirable. Since the quality of the produced oil depends mainly on the olive fruits (OF), it is important to manually check each batch of OF before milling them in addition to performing lab tests to verify the quality of the produced OO. The goal of this work is to automate the process of classifying OF based on whether they produce extra virgin OO (EVOO) or not. We collect a large dataset of more than 11K OF images and label them as positive/negative based on whether they produced EVOO or not. We then fine-tune several state-of-the-art deep learning models on this dataset. The results show that most pretrained models are very accurate for this dataset leading the suggestion that we use the most efficient one. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
