Machine Learning Applications for Precise Nutrient Deficiency Detection in Paddy Farming

Abstract

This study tackles the critical issue of nutrient deficiencies in paddy crops by leveraging machine learning for accurate detection, while also considering the economic implications for farmers. Since agriculture is fundamental to global food security—and rice is a key staple crop—sustainable cultivation depends on effective nutrient management. Deficiencies in paddy crops can significantly reduce yield and quality, posing a major challenge for farmers. Traditional detection methods, which rely on subjective visual inspection, are labor-intensive and often delay corrective measures. However, advances in machine learning and data-driven approaches present a promising solution. The research focuses on using machine learning to identify nutrient deficiencies in paddy leaves, employing the K-means clustering algorithm for precise and efficient detection. It explores practical applications in agriculture, emphasizing robust feature selection and model validation. The proposed method integrates image processing techniques, including image capture, RGB-to-HSI conversion, and segmentation, building upon prior work in nutrient deficiency detection. Performance is evaluated using key metrics such as accuracy, precision, recall, and F1 score. The model is trained and tested on datasets from the International Rice Research Institute (IRRI), targeting deficiencies in nitrogen (N), potassium (K), phosphorus (P), manganese (Mn), and zinc. By combining machine learning with image analysis, this approach offers a faster, more reliable alternative to traditional methods, ultimately supporting better crop management and food security.