JCSO‐⋮DMN: Jellyfish Competitive Shuffled Optimisation Enabled Deep Maxout Architecture for Maize Leaf Disease Detection in IoT

Agriculture is one of the significant sectors and has been considered the backbone of the Indian economy. Crops, like rice, cotton and maize, are commonly cultivated in India. Among these, maize is the pivotal crop and serves a vital part in crop production. The huge issue that affects the production of crops is plant infections. Automatic crop disease identification is a hot research discussion in today's agriculture because of the appearance variations and messed up environment. To counterpart such constraints, a deep learning model has come into the stage that refines visual features and detects crop diseases; however, such methods work well only under simple backgrounds. It achieves poor accuracy while processing the images acquired from agricultural areas that include unavoidable calamities. In such circumstances, the Internet of Things (IoT) has gained increased concentration with IoT choices to gather high‐level contextual data that assists recognition models to efficiently detect crop infections in the wild. For maize leaf disease identification in IoT, this research proposes a Deep Maxout Network (DMN) enhanced by Jellyfish Competitive Shuffled Optimisation (JCSO). This algorithm novel is the hybridisation of Competitive Shuffled Shepherd Optimisation (CSSO) as well as Jellyfish Search (JS) optimiser. However, CSSO derived is by merging the features of the Shuffled Shepherd Optimisation algorithm (SSOA) and Competitive Swarm Optimiser (CSO). The proposed JCSO‐DMN has delivered remarkable results regarding energy, accuracy, sensitivity, specificity and delay of 1.000 J, 0.949, 0.976, 0.946 and 0.126 s, respectively.