Optimized Betel Leaf Disease Detection Using Improved CNN Model for Precision Agriculture

Abstract

Plant diseases are a major threat to the production of betel leaves, causing large financial losses and endangering the expansion of the betel leaf market. This paper examines how deep learning methods can be used to analyze massive plant image datasets in order to tackle this problem. The processes covered in the study include feature extraction, feature selection, pre-processing, data collecting, and classification. Data collection involves gathering an extensive database of plant images depicting various diseases affecting betel plants. Pre-processing methods, like resizing,data augmentation,noise reduction, enhance image quality. Principal Component Analysis (PCA) is employed as a feature extraction technique to extract relevant data from the images. For classification, deep learning models like VGGNet and MobileNet based CNN architectures are employed, and it offers improved performance in image recognition tasks. These models use the features that have been retrieved and chosen to effectively and precisely categorize images. Through the identification of distinct patterns and symptoms linked to betel plant diseases, these algorithms provide prompt classification, thereby averting possible economic damage and preserving food security. The proposed model is implemented using Python platform.