A Hybrid Framework for Detection and Analysis of Leaf Blight Using Guava Leaves Imaging

Fruit is an essential element of human life and a significant gain for the agriculture sector. Guava common fruit found in different countries. It considered fourth primary Pakistan. Several bacterial fungal diseases guava decrease production daily. Leaf Blight disease that affects growth fruit. Automatic detection leaf blight can help avoid decreases its production. In this research, we proposed CNN-based deep model named SidNet. The contains thirty-three layers. We used dataset early recognition blight, which consists two classes. Initially, YCbCr color space was employed as preprocessing step detecting blight. As original small, data augmentation performed. DarkNet-53, AlexNet, SidNet were feature acquisition. features fused to get best-desired results. Binary Gray Wolf Optimization (BGWO) on selection. optimized given variants SVM KNN classifiers classification. experiments performed 5- 10-fold cross validation. highest achievable outcomes 98.9% with 5-fold 99.2% validation, confirming evidence identification accurate, successful, efficient.