Evaluating AI-Based Methods for Sugarcane Leaf Disease Detection and Classification using Image Processing Techniques

Abstract

The health of sugarcane crops is fundamentally significant for the rural economy, yet they are habitually compromised by different leaf illnesses that can altogether decrease yield and quality. Powerful and opportune location and arrangement of these sicknesses are fundamental for carrying out suitable administration methodologies. A comprehensive comparison of various artificial intelligence (AI) methods for identifying and classifying sugarcane leaf diseases is presented in this study. Leaf diseases impact food security and profitability by lowering agricultural output and quality. In most country India, agribusiness is the principal kind of revenue. Consequently, farming plant infections should be naturally analyzed and arranged utilizing exceptional and exact man-made intelligence (Computerized reasoning) strategies. This permits ideal preventive counsel. Image processing, Machine Learning, and Deep Learning are utilized in AI strategies. The most recent studies on sugarcane leaf disease detection and classification were thoroughly examined. The investigation centers around DL or ML calculations, explore datasets, execution pointers, and model precision. SVM (Support Vector Machine) was used in 45 percent of sugarcane leaf image classification studies, while KNN (K-Nearest Neighbors) was used in 22%. In 22% of queries, K-means clustering and ANN (Artificial Neural Network) classifiers were utilized. Pre-trained models were used in 44% of studies, while CNN (Convolutional Neural Network) models were used in 56%. When compared to controlled laboratory images, the performance of machine learning and deep learning models on realworld image datasets is subpar. Deep learning models, on the other hand, were 98.84 percent more accurate than machine learning models. Image categorization with ML models is inferior. Prior research's detection and classification techniques had significant flaws. New crop disease diagnosis and plant leaf disease detection and classification methods will be discovered thanks to this study.