Development of a pineapple leaf disease diagnostic model by extracting visual features together with data mining.

Abstract

The purpose of this research was to study and develop a diagnostic model in pineapple using extracting image features together and data mining techniques. The disease of pineapple leaves data was collected in 150 leaves, categorized as 50 wilts, 50 root rot, and 50 normal leaves. Using algorithm Color layout filter, Simple color histogram filter, together with Multilayer Perceptron, Naive Bayes, LMT and Random Forest. The operations were compared between image feature extraction. using the accuracy, precision, recall, and F-measure to measure the performance of the model. Results showed that Simple Color Histogram Filter with Naive Bayes 84.66% was the best, followed by the Simple Color Histogram Filter algorithm with LMT 84.66%, and the Simple Color Histogram Filter algorithm with the Random Forest 84.66%, respectively.