An Intelligent Grape Leaf Disease Identification Model Using WOA-Optimized XG- Boost Classifier

Grape Leaf Disease (GLD) identification have become a significant agricultural issue with an alarming rise in recent years, necessitating effective prediction algorithms. In this paper, Whale Optimization Algorithm based XG Boost classifier is proposed for prediction of GLD such as Black Rot, ESCA, healthy, Leaf blight. Initially, image resizing is applied to the grapevine disease dataset to resize the image, noise reduction to reduce the unwanted noise and improve the clarity of unclear image using Contrast Limited Adaptive Histogram Equalization (CLAHE). Next, the processed image is given to the Otsu’s segmentation, to calculate the threshold value. After that grape leaf image is featured using Local Binary Patten (LBP) for analysing local texture structure. Finally, a WOA optimized XG Boosting framework is processed to enhance the classification of GLD for analysis. Using python software proposed framework have accuracy of 92.41% is accomplished when compared to other techniques.