<![CDATA[Black pod disease is a severe disease affecting cocoa fruit, caused by the Phytophthora Palmivora fungus. This infection turns the fruit's surface dark brown to black, while the inside becomes rotten. Currently, identifying infected cocoa fruits is done manually through visual observation, which is prone to errors and inconsistency. This study aims to implement a Convolutional Neural Network (CNN) algorithm to classify images of black pod disease in cocoa fruits. The dataset consists of 1,500 images obtained through documentation and literature review, with 750 images of healthy cocoa fruits and 750 images of infected fruits. To determine the optimal configuration, the CNN model was tested across 15 scenarios with varying batch sizes and epochs. The results show that the fifth scenario, with a batch size of 32 and 50 epochs, achieved the best performance, with an accuracy of 97.33%, precision of 97.41%, recall of 97.33%, and an f1-score of 97.33%. Additionally, the model was further tested using 20 original images, achieving an accuracy of 90%. These results demonstrate that the CNN model developed effectively classifies cocoa fruit images affected by black pod disease, highlighting its potential for use in developing more accurate and efficient cocoa disease detection applications]]>
Copyrights © 2025
