Plant Disease Prediction Using Feature Selection Based Multilayered ANN with Genetic Algorithm

This paper introduces a novel method for detecting plant diseases by combining a multilayer Artificial Neural Network (ANN) with a feature selection technique based on Genetic Algorithms (GA). The proposed system aims to enhance the accuracy and efficiency of disease detection in plants by leveraging machine learning techniques. The feature selection process, powered by GA, optimizes the input feature set, eliminating irrelevant or redundant features and focusing on the most significant attributes for disease classification. This reduces the dimensionality of the data and improves the performance of the ANN, which is employed to classify the health status of plants based on visual symptoms or sensor data. The GLCM technique is used for the feature extraction process of input image. The multilayer ANN is trained to recognize patterns indicative of various plant diseases, with the GA facilitating the identification of the most relevant features for each disease classification task. Experimental results demonstrate that the integration of GA-based feature selection with ANN significantly improves classification accuracy, compared to traditional approaches that rely on exhaustive feature sets. This method shows promise for real-time, automated plant disease detection, offering potential applications in precision agriculture and crop management. Overall, the proposed system provides a robust, scalable solution for early disease detection, potentially reducing crop losses and supporting sustainable agricultural practices.