<![CDATA[Visual assessment of papaya ripeness often leads to inconsistent and low accuracy results. To address this, the study applies Logistic Regression and K-Nearest Neighbor (K-NN) algorithms for automatic classification using digital image processing. The initial dataset consisted of 300 images, which were expanded to 1,200 through preprocessing and augmentation. Features were extracted using the Gray Level Co-occurrence Matrix (GLCM) method, and the data was split into 80% for training and 20% for testing. The study aims to compare the performance of both algorithms and understand their classification mechanisms. Results show that K-NN with k=1 achieved an accuracy of 87%, while Logistic Regression with L2 regularization reached 73%, indicating that K-NN outperforms Logistic Regression in classifying papaya ripeness levels.]]>
Copyrights © 2025
