Rice Leaf Disease Detection and Crop Yield Prediction Using Random Forest

Abstract

Rice production contributes a considerable amount to national income. Rice production can be affected by various diseases like brown spots, bacterial leaf blight, leaf smut caused by fungi, bacteria, etc. In this research, the diagnosis of rice plant leaf disease is done using random-forest classification and Digital image processing. The random forest classifier is efficient and accurate on a large dataset. The image is uploaded to the system by following digital image processing steps and using a random forest algorithm to perform on the processed image which outputs disease name, cause, symptoms, and remedy respectively. The proposed method also predicts the crop yield based on temperature, rainfall, humidity, and soil pH level. Overall, the model achieves 90% of accuracy.