Potato Leaf Disease Prediction: A Machine Learning Perspective

Abstract

Potato leaf disease has mostly two categories; early blight and late blight disease. The disease may be more prevalent in certain weather patterns and have a catastrophic effect on potato crops. In summary, warm, humid weather with frequent rain or heavy dew, temperatures between 15°C and 20°C, and a lack of sunshine are the weather conditions that can cause potato late blight. Drier weather conditions favour early blight, unlike late blight. Warm and dry weather with a lack of rain or irrigation, temperatures between 21°C and 29°C, and high humidity in the morning are the weather conditions that can cause potato early blight. A modified dataset is used for climate-influenced prediction, and the testing accuracy using random forest models is 97%. Analysis of experimental results shows that the suggested potato-leaf disease prediction based on the weather data framework outperforms the outcome of frameworks.