Chapter 34 - Whale-optimized convolutional neural network for potato fungal pathogens disease classification

Abstract

Artificial intelligence (AI), along with its subfields of machine learning (ML) and deep learning (DL), allows computational models to process and learn the data representations with various levels of abstraction. AI models are widely used in finance, healthcare, technology, business, banking, science, and agriculture. In agriculture, utilizing ample human resources at each stage of cultivation and challenging tasks involved in predicting diseases and other factors using traditional methods is ineffective and expensive. Potato (Solanum Tuberosum) is a highly consumed and cultivated vegetable crop across the globe. The potato crop yield is highly affected by the fungal pathogens which cause leaf diseases: Early Blight (EB) and Late Blight (LB). Early detection of fungal pathogens in the crop can reduce yield loss. To detect the fungal pathogens in the plants, a continuous monitoring system in the field must be established. Computer vision and artificial intelligence can help farmers monitor plant diseases at every stage. Their application in monitoring the diseased leaves of the crop and predicting the diseases correctly has received great attention from researchers. This chapter deals with applying a nature-inspired Whale Optimization Algorithm (WOA) in optimizing the Convolutional Neural Network (CNN) performance in classifying the EB and LB diseases in potato leaves.