Tomato Leaves Disease Detection Using Gray Level Co-Occurrence Matrix (GLCM) and Image Processing Techniques

Crop diseases pose a serious danger to agricultural productivity and food security, making accurate detection techniques urgently needed. Convolutional Neural Networks (CNN) are used in this project's automated approach to better detect damaged tomato leaves using image processing. First, a sizable dataset with tagged photos of both healthy and diseased leaves is gathered from sources such as the Plant Village collection. To increase the diversity and quality of data, preprocessing techniques like data augmentation, scaling, and standardization are used. To guarantee accuracy and robustness, the CNN model is trained on well-labeled datasets after extracting complex features from the images. Techniques like batch normalization and dropout help reduce overfitting. Metrics including accuracy, precision, recall, and F1-score are used to evaluate the model's performance, showing significant improvements over conventional methods. With potential for wider uses in precision farming, the findings show that CNNs are a viable tool for the early and precise diagnosis of tomato leaf diseases. In order to enhance performance across a range of crops and diseases, future research will concentrate on transfer learning and growing datasets. The proposed model achieved 97.2% ACC, 95.7% Spec, 94.5% Recall, 98% Prec, and 97.5% F1-Score.