Diagnose crop disease using Krill Herd optimization and convolutional neural scheme

Develop a novel Krill Herd-based Convolutional Neural (KHbCN) scheme to identify and diagnose crop diseases accurately. Using an improved krill herd fitness function, the proposed model can identify crop disease reliably and improve detection performance. The krill herd fitness is updated to the convolutional neural network (CNN) to diagnose crop damage accurately. The created framework is executed in Python, and the system is evaluated and trained using the plant villa dataset. After removing mistakes during preprocessing, feature extraction is used to extract texture features from the crop. Finally, the constructed model uses the fitness of the krill herd to identify crop illness. By recognizing and detecting agricultural diseases, the primary goal of building a convolution-based optimization model is to enhance the growth of agriculture. The experimental findings of the framework's development are contrasted with those of other cutting-edge methods that achieve 99.85% accuracy, 98.98% recall, 99.85% precision, and 40 ms execution time.