Hybrid Machine Learning Model for Early Detection of Cucumber Leaf Curl Disease in Smart Agriculture

Efficient Accurate and timely detection of plant diseases plays a crucial role in maintaining crop health and ensuring sustainable agricultural productivity. Manual identification of leaf-based diseases is often labor-intensive, time-consuming, and subject to human inconsistency. To overcome these limitations, this study introduces a hybrid machine learning-based framework for the early detection of Cucumber Leaf Curl Disease (CLCuD) within a smart farming environment. The proposed system integrates Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Decision Tree (DT) classifiers as base learners to construct a robust ensemble model capable of high-precision classification. A real-time cucumber leaf image dataset was used for model training and evaluation. Experimental outcomes demonstrate that individual classifiers achieved accuracies of 96.37% (SVM), 97.00% (KNN), and 98.00% (DT), while the hybrid ensemble attained the highest accuracy of 98.29%. The results confirm that the proposed hybrid model offers superior detection accuracy, reliability, and efficiency compared to individual classifiers, providing a promising tool for early disease identification, yield preservation, and the advancement of precision agriculture through smart and sustainable farming practices.