This Research explores the application of Genetic Algorithm (GA) for the identification and segmentation of diseased regions in leaf images. By treating pixels of the input image as a population and using their intensity values as fitness metrics, the GA iteratively traces pixels with fitness values above 160 and below 41, effectively identifying diseased regions. With a population size of 100, the proposed method achieves an average accuracy of 92%, demonstrating efficacy even for small diseased areas. GA, a probabilistic search algorithm inspired by Darwinian evolution, efficiently searches for near-optimal solutions by simulating natural selection processes. It involves initializing a population of possible solutions, evaluating their fitness, and using crossover and mutation to generate new solutions, which replace the previous generation until a stopping criterion is met. The proposed method segments diseased leaf regions by considering image segmentation as a global optimization problem. Preprocessed input images are resized to 256x256 pixels for computational efficiency. Chromosomes, represented by row and column indices of pixels, undergo fitness evaluation based on pixel intensity. The GA converts RGB images to grayscale for faster processing, and iteratively retains pixels with relevant intensity values, discarding the rest. Analysis on 500 apple leaf images from the PlantVillage repository, including diseases like apple scab, black rot, and cedar apple rust, shows that the method effectively segments lesion regions. Classifiers such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and neural networks are used to classify the segmented regions, with neural networks achieving the highest accuracy of 92%. The proposed GA method, while performing robustly under normal lighting conditions, shows slight performance degradation in the presence of specular noise. Nonetheless, it offers a computationally efficient, accurate approach to leaf disease detection, supporting precision agriculture through enhanced plant health management.

genetic algorithm ,machine learning,plantdisease,detection ,optimization