E-waste management using hybrid optimization-enabled deep learning in IoT-cloud platform

Abstract

Electronic waste (E-waste) is generated at a quick pace due to technological expansion and thereby reducing the obsolescence age of electrical and electronic equipment (EEE). This ever-increasing e-waste is not only dangerous for the environment but also to the health of human beings. A robust and effective E-waste classification approach is devised in this paper, named Fractional Horse Herd Gas Optimization-based Shepherd Convolutional Neural Network (FrHHGO-based ShCNN) for classifying E-wastes in the Internet of Things (IoT)-cloud platform. Here, the E-waste images are collected by the IoT nodes and then stored in the cloud data storage which then performs the process of routing using the proposed FrHHGO algorithm. Moreover, effective features are extracted and then augmented to perform the process of E-waste classification. The E-waste classification is performed using FrHHGO, which is the combination of Fractional Henry Gas Optimization (FHGO), and Horse Herd Optimization Algorithm (HOA). The developed method outperformed various existing approaches with minimum energy and delay of 0.301 J, and 0.666 s, maximum accuracy, sensitivity, and specificity of 0.950, 0.934, and 0.967, respectively. Thus, the developed E-waste classification system enhances the social, environmental, and economic sustainability using FrHHGO-based ShCNN in emerging economies.