Combining GLCM Features with SVM Classification for Improved Accuracy in Cotton Crop Disease Detection

Different diseases affecting cotton plants can be diagnosed and controlled to obtain high yields of good quality cotton. In this study, we propose a methodology for categorizing crop disease images into two groups: diseased and healthy images using feature extraction based on Gray-Level Co-Occurrence Matrix (GLCM) and classification using a Support Vector Machine (SVM). The goal of this study is to improve disease diagnosis using image processing by applying the texture co-occurrence matrix, which gives information on the relationship of pixel intensity of the images. The proposed model is then tested on a cotton crop dataset, and the accuracy score is 86%. These findings enrich the literature by proving the concept that integrating texture-based feature extraction with sophisticated machine-learning approaches to the classification of crop disease highlights diagnostic precision to boost the efficacy of precision agriculture.