A Hybrid Approach for Corn Leaf Diseases Detection Based on the ResNet and Capsule Neural Network

Hybrid model of Residual Networks (ResNets) and Capsule neural networks (CapsNets) for corn leaf disease detection represents an innovative advancement in the field of agricultural technology. Corn leaf disease detection helps promote sustainable agricultural practices, guarantee farmers' financial stability, and promote global food security in addition to safeguarding crops and maximizing yields. The ability to detect diseases early, improve crop quality, and encourage environmental stewardship in agriculture are all made possible by advancements in disease detection technologies. Proposed hybrid CapsNets model achieved remarkable F1_score and accuracy is 99.5% and 99.6% on plant village dataset, respectively. In contrast to the capsule network (CapsNet), the proposed model's accuracy has increased by 12%.CapsNet only needed 110 training epochs to attain its optimise state and proposed model needed 140 training epochs to do the same. The experimental results demonstrate the efficiency of this technology in recognizing and classifying corn leaf illnesses, and it can provide early defence against these diseases.