<p>Cotton Leaf Curl Disease (CLCuD) presents a major threat to cotton production on a global scale, leading to substantial economic losses. Early and accurate detection of CLCuD is critical for timely intervention and effective disease management. In this study, we propose a novel approach that leverages the combined strengths of InceptionV3 and YOLOv8 models to automatically detect and classify CLCuD. InceptionV3 is utilized to extract high-level features from cotton leaf images, capturing intricate visual patterns, while YOLOv8, a state-of-the-art real-time object detection model, is employed to identify and classify disease-related changes in the images. Additionally, a statistical prediction model is incorporated to measure the distances between different visual representations of leaf curl disease and the designated affected areas, further enhancing detection accuracy. The model was trained and tested on a large dataset comprising images of both healthy and diseased cotton leaves. Performance metrics, including precision, recall, and F1 score, were used to evaluate the model’s effectiveness in identifying CLCuD symptoms. The results demonstrate the superiority of the hybrid approach, achieving an overall accuracy of 98%. Specifically, the precision, recall, and F1 score for healthy leaves were 99%, 97%, and 96%, respectively, while for diseased leaves, they were 97%, 96%, and 98%. The weighted and macro averages across all categories were consistent at 98%. The findings highlight the robustness of the proposed model in early and accurate detection of CLCuD, outperforming individual models and providing a reliable, automated solution with promising real-world applications for cotton crop disease management.</p>