QUALITY ASSESSMENT OF COPRA USING DEEP LEARNING AND IMAGE PROCESSING TECHNIQUES

Janrey Elecito; Jannie Fleur Orano; Czarina Ancella Gabi; Narciso Angelo Teofilo; Abelito Lim; Jessica Nava; Cristobal Paraon

Authors

Janrey Elecito Southern Leyte State University, Philippines
Jannie Fleur Orano Southern Leyte State University, Philippines
Czarina Ancella Gabi Southern Leyte State University, Philippines
Narciso Angelo Teofilo Southern Leyte State University, Philippines
Abelito Lim Southern Leyte State University, Philippines
Jessica Nava Southern Leyte State University, Philippines
Cristobal Paraon Southern Leyte State University, Philippines

Keywords:

copra quality, deep learning, image processing, YOLOv11

Abstract

This research investigates the application of the YOLOv11 object detection model in the classification of copra images, specifically categorizing them into three distinct grades: Grade A, Grade B, and Grade C. Through performance metrics such as precision, recall, F1
score, and confusion matrices, the study assesses the model’s effectiveness in accurately identifying and classifying copra grades. The findings reveal that the model achieves an overall accuracy rate of 93.65% on the testing dataset, demonstrating its proficiency in copra grade classification tasks. Furthermore, the model’s performance is validated through a Cohen’s kappa value of 0.90, indicating a high level of agreement with human expert labels and reflecting the model’s robustness and reliability. These results hold significant implications for the agricultural sector, offering insights into the potential integration of advanced deep learning techniques for enhancing quality assessment and optimizing copra production processes.