Crop Disease Prediction and Yield Prediction using Machine Learning (Sugarcane)

Abstract

Abstract— Sugarcane cultivation faces significant challenges due to biotic stresses such as fungal and viral diseases, as well as abiotic factors including climate variability and soil nutrient imbalance. Accurate disease diagnosis and yield estimation are critical for improving productivity and ensuring sustainable agricultural practices. This study proposes an integrated machine learning framework for automated sugarcane disease classification and yield prediction.A Convolutional Neural Network (CNN)-based deep learning model is implemented for image-based disease detection, enabling automatic feature extraction and high-accuracy classification of major sugarcane diseases. For yield prediction, supervised regression algorithms including Random Forest Regressor and Gradient Boosting are employed to model the relationship between environmental parameters, soil properties, and historical yield data.Experimental evaluation demonstrates improved predictive performance in both classification and regression tasks. The proposed system provides a data-driven decision support tool for farmers, enhancing crop management efficiency and reducing agricultural losses. Keywords: Sugarcane, Disease Prediction, Yield Prediction, Machine Learning, Convolutional Neural Network (CNN), Random Forest, Support Vector Machine (SVM), Artificial Neural Network (ANN), Precision Agriculture, Crop Monitoring.

Plant phenotyping relevance

CNNによるサトウキビ病害の画像ベース分類は、植物の病徴・状態を観測から推定する中心的手法であり、方法適用として収録対象です。収量予測も含まれますが、環境・土壌・履歴データに基づく農業予測で、病害画像分類部分が主な適格要素です。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.