Paddy Disease Detection using a Hybrid Machine Learning Model

The shortcomings of RNN in the detection are lower speed, complex in nature, and not able to remember long sequences. To address these, RNN is coupled with GRU, whose architecture is simple, ensuring speed, and is also coupled with LSTM, which remembers long sequences. Most southern states that are dependent on or have a high priority on eating paddy rice as part of their daily food. The integrated approach is applied over the paddy crop for disease identification, then takes appropriate steps, through the usage of a hybrid model for detection, and then makes necessary steps for pesticide spraying for the removal of fungus in the paddy crop. The disease is to be detected early, so that losses to be minimized as well as increase productivity. The purpose of using a hybrid model is to ensure that paddy farmers increase their revenue and income by early identification of the disease. The effectiveness of the hybrid model is through specific measures and analyzed against other models in terms of accuracy and performance.