Identification of diseases in rice plant (oryza sativa) using back propagation Artificial Neural Network

Abstract

In this study, digital image processing was incorporated to eliminate the Subjectiveness of manual inspection of diseases in rice plant and accurately identify the three common diseases to Philippine's farmlands: (1) Bacterial leaf blight, (2) Brown spot, and (3) Rice blast. The image processing section was built using MATLAB functions and it comprises techniques such as image enhancement, image segmentation, and feature extraction, where four features are extracted to analyze the disease: (1) fraction covered by the disease on the leaf; (2) mean values for the R, G, and B of the disease; (3) standard deviation of the R, G, and B of the disease and; (4) mean values of the H, S and V of the disease. The Backpropagation Neural Network was used in this project to enhance the accuracy and performance of the image processing. The database of the network involved 134 images of diseases and 70% of these were used for training the network, 15% for validation and 15% for testing. After the processing, the program will give the corresponding strategic options to consider with the disease detected. Overall, the program was proven 100 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> accurate.