Hybrid optimized deep quantum neural network in Internet of Things platform using routing algorithm for detecting smart maize leaf disease

• Published in: International journal of adaptive control and signal processing 2024-08, Vol.38 (8), p.2873-2892
• Main Authors: Gunisetti, Loshma, Koduri, Shirin Bhanu, Jagannathan, Veeraraghavan, Chundru, Raja Ramesh
• Format: Article
• Language: English
• Subjects: Adaptive algorithms; adaptive concept; Algorithms; base station; Corn; Disease; energy texture features; Internet of Things; maize leaf; Neural networks; Optimization; Plant diseases; routing; statistical features
• ISSN: 0890-6327; 1099-1115
• DOI: 10.1002/acs.3836

Summary The productivity in the agricultural sector is minimized due to the disease in plants. In general, the ailments that affect plants are identified by the farmers and the losses are minimized, when the diseases are identified early. The early identification of leaf diseases is difficult in the traditional approaches. Hence, in this article, for detecting maize leaf disease, an adaptive competitive shuffled shepherd optimization‐driven deep quantum neural network (adaptive CSSO‐based deep QNN) is implemented. Here, the initial process is the simulation of the IoT nodes and the leaf data are collected. This data are transferred to base station (BS) via the best routes. The optimal routes are identified using the adaptive CCSO algorithm. The adaptive concept, shuffled shepherd optimization algorithm (SSOA) and competitive swarm optimizer (CSO) are merged for forming the adaptive‐CSSO algorithm. The leaf detection is done in the BS and initially, the data is preprocessed using region of interest (ROI). Then, the relevant features are extracted. Finally, the disease in the maize leaf is detected using Deep QNN and the training is done by adaptive CSSO. The devised approach has maximum accuracy of 96.04%, sensitivity of 97.41%, specificity of 94.35%, energy of 0.01 J, and minimum delay of 0.9596 s.