EarthGrow: A Hybrid Deep Learning Architecture for Sustainable Agriculture

The current technology era has connected all the parts of the world, leading to the crisis of the agriculture, which is considered to br the most essential item for mankind. Thus, gradually leading to the question on the sustainability of the agriculture. Therefore, a novel approach for creating sustainable agriculture by detecting the diseases of various crops is proposed. The proposed methodology utilized a hybrid deep learning approach, combining the two state-of-the-art methods, the CvT (Convolutional Vision Transformer) and GRU (Gated Recurrent Unit) for disease detection. The proposed methodology has proved the worth of it by gaining an accuracy of 95.86% during the testing. And by suppressing the current methods with exceptional margin.