FEATURE SELECTION FOR LEAF DISEASES DETECTION USING OPTIMIZATION ALGORITHMS

Abstract

Modern techniques of computer science and machine learning become more and more important and useful in recent years. The cutting-edge techniques of artificial intelligence assisted humanity in different aspects of life, such as health-care, military and agricultural fields. However, a current solution of identifying leaves diseases totally based on visual inspection of farmers and agricultural engineers. Because this is a very time consuming manual method, its cost is also high as it requires a lot of personnel and risks a lot of plants. This work proposes a novel solution to identify the location and type of diseases on plant leaves, using imperialist competitive algorithm (ICA) for feature selection, and an efficient artificial neural network (ANN) algorithm for recognition. Moreover the comparison between two meta-heuristic optimization algorithms namely imperialist competitive algorithm (ICA) and particle swarm optimization (PSO) is given to demonstrate the effectiveness of the ICA.