Classification of leaf diseases using modified genetic algorithm and normalized sum square deviation

Abstract

In real world scenario, large number of features represents a data, but all these features are not useful. In this paper, hybrid algorithm comprising of modified genetic algorithm (GA) for segmentation and normalized sum square deviation (NSSD) for feature selection is proposed. The proposed algorithm is tested on a standard leaf disease dataset to classify diseased leaf and healthy leaf.The proposed algorithm yields an average accuracy of 94.3% for neural network classifier.