Automatic grape leaf disease detection using improved attention residual net classifier

Grape plant diseases are typically severe and spread quickly. Henceforth, this paper proposes a method for GLD detection using an Improved Attention ResidualNet for diagnose early-stage diseases by analyzing leaf images. This paper incorporates the preprocessing step utilizes a Median Filter (MF) for reduce noise, K-means clustering based segmentation used to identify Regions of Interest (ROI) for further analysis of grape leaf. For feature extraction, the Scale-Invariant Feature Transform (SIFT) is utilized to capture important characteristics of the leaf images. The proposed model, developed using Python software and trained on a grapevine disease dataset, validates high classification of 94 % accuracy on the test dataset. For accurate and efficient GLD identification.