Sustainable Pesticide usage in Agriculture

Human monitoring cannot efficiently and effectively detect the pest or the disease that has infected the crop, even with the familiarity and expertise in the field of agriculture. Machine learning algorithms in recent days have been able to predict the output with utmost accuracy and precision based on the input. With the help of techniques such as deep neural networks, one can predict the pest or disease attack affecting the crop. The excessive usage of pesticides is detrimental to the overall farm yield. The appropriate knowledge of the quantity and quality of pesticides is of paramount importance for paving the way for sustainable agriculture. The work in this paper enunciates the implementation of predicting the countermeasures for a few disease classes of two vegetables, namely tomato and potato which are some of the most widely grown routine vegetables that are susceptible to frequent pest attacks. The model of this solution outputs a 94% accuracy rate with customized local language prediction and countermeasure suggestions to the farming community to employ pragmatic measures in the near future to protect their agricultural produce.