| Maturity-based classification of olive fruits by RGB images and transfer learning |
| Paper ID : 1165-NICAME1402 |
| Authors: |
|
Seyed Iman Saedi *, Mehdi Rezaei Faculty of Agriculture, Shahrood University of Technology, Shahrood, Iran |
| Abstract: |
| The use of olive fruit at different maturity stages can lead to the production of various postharvest products with distinct sensory and nutritional characteristics. In addition to the production of table olives, the quality of olive essential oil is also influenced by the cultivar and maturity stage of the olives. Developing an efficient method for sorting and recognizing olive maturity stages can have significant practical implications, particularly in the postharvest processing of olives. Therefore, this study presents an automatic computer vision system that applies state-of-the-art deep learning technology to sort and classify an Iranian olive cultivar, Roghani, in five maturity stages. The developed model utilized deep convolutional neural networks and transfer learning with VGG16 network and ImageNet weights as the base architecture. The model was fine-tuned through sets of well-known layers including convolution, max-pooling, batch normalization, dropout, global average pooling, and others. Data augmentation was used to minimize the risk of overfitting. The model was trained and optimized using categorical cross-entropy loss function and Nadam optimizer. Results showed that the proposed model achieved a total classification accuracy of 75% and a loss of 1.07 on test data. The model's performance was also evaluated using confusion matrix and classification report. The results of this study indicate that the proposed model can be effectively integrated into an olive sorting system, making it easier to distinguish between olives of various levels of maturity for various post-harvest products and better essential oil. |
| Keywords: |
| Olive; Maturity; Transfer Learning; Sorting |
| Status : Paper Accepted (Poster Presentation) |
