One particular area that provides food to life is agriculture. The Philippines is in a tropical country, and its climate causes the variation of plant diseases that affect the yield of agriculturalists, and the growth of mango trees. This study presents an application of image processing and a convolutional neural network (CNN) for carabao mango leaf disease identification. Rectified Linear Unit (ReLU) activation function was applied in the CNN model and were implemented using tens or flow deep learning libraries. The model was retrained to classify the leaves into four categories; anthracnose, sooty mold, red rust, and healthy leaf. There were 2080 leaf images captured using smartphone camera from the real cultivation conditions area. Experimental results revealed that the convolutional neural network model achieved 81.01% accuracy in the identification of carabao mango leaf diseases and could be applied to the plant diagnosis application.
Volume 12 | 01-Special Issue
Pages: 152-158
DOI: 10.5373/JARDCS/V12SP1/20201058