MATLAB-Based Plant Pest Infestation Detection System

This project develops a MATLAB-based plant pest detection system using color analysis to classify leaves into four severity levels: mild infestation, moderate infestation, severe infestation, and normal. The algorithmic workflow is structured as follows: 1. Organize leaves by severity level into separate directories, where each folder contains samples of a specific infestation category. 2. Train the system by processing leaves from each folder to extract their characteristic color components. This typically involves using MATLAB's image processing functions like imread() for image loading and rgb2hsv() for color space conversion to obtain Hue, Saturation, and Value components. 3. Compile the extracted color features into a training dataset stored as color.mat using MATLAB's save() function, which contains reference values for each infestation category. 4. During testing, preprocess the target leaf image through brightness normalization using imadjust(), convert to appropriate color spaces (e.g., HSV or LAB), and compute dominant color components through color histogram analysis or mean color value calculation. 5. Perform classification by comparing the test leaf's color features against the trained color.mat dataset using distance metrics (Euclidean or Mahalanobis) or machine learning classifiers (k-NN or SVM) implemented via MATLAB's classification learner app or custom code. For any technical questions regarding the implementation workflow or code development, the author provides comprehensive support and detailed explanations of the MATLAB programming approach.