DETECTION OF DISEASES ON BANANAS (Musasp.) USING IMAGE PROCESSING AND MACHINE LEARNING TECHNIQUES

Bananas, whose demand is very high in the global market, are considered one of the best agricultural export products in the Philippines - a country where agriculture plays a significant role in economic development. However, diseases in bananas have caused significant losses to farmers over the years due to low yields, as it significantly affects the growth and quality of the fruits. As a solution to the problem, research highlights the importance of identifying banana diseases at an early stage, enabling local farmers to implement cost-efficient control methods that can minimize, or even eliminate, the spread of infestations. Given the effectiveness of image processing in classification and analysis tasks, this study centered on its application hence, a dataset of 3,000 images depicting common banana diseases, primarily manifesting on the leaves, was compiled and divided into training, validation, and testing subsets. These images underwent preprocessing before being input into four pre-trained convolutional neural network architectures-VGG-19, InceptionV3, ResNet50, and EfficientNet-all of which were configured with identical optimization techniques and model parameters.Performance evaluation metrics such as accuracy results, confusion matrix, and classification report were used to identify the model with the highest performance in a test dataset.The results have shown that among the identified model architectures, the EfficientNet model obtained the highest accuracy of 91%.