OPTIMIZED MIXSTYLE NEURAL NETWORK-BASED INTELLIGENT WASTE MANAGEMENT SYSTEM WITH INTERNET OF THINGS
International Research Journal of Modernization in Engineering Technology and Science · 2024
Waste management results in the destruction of waste through landfilling and recycling.Waste creation is increasing in most civilizations as a result of population growth, as seen in the recent faster rate of increase in urban population density.Any smart waste management system would need to be able to process this aggregate data in order to identify the trash status in each bin, aggregate bin locations, and detect bin locations.In this manuscript Optimized Mixstyle Neural Network-based intelligent waste management system with Internet of Things (OMNN-IMS-IoT) is proposed.Initially the data is collected from TrashNet database.The collected data is pre-processed using Information Exchange Multi-Bernoulli Filter (IEMBF) to remove noise also resize the pictures.The pre-processed images are classified using Mixstyle Neural Network (MNN) as digestible and indigestible waste like cardboard, glass, metal, paper, plastic and trash.In general, MNN does not express any modification to optimization method to find the ideal parameters to ensure precise categorization of waste images.Hence, Namib beetle optimization algorithm (NBOA) is suggested to improve weight parameter of MNN classifier, which precisely classifies the waste images.The anticipated technique is executed in Kaggle docker container.The preferment of OMNN-IMS-IoT approach attains 20.28%, 28.22%, and 29.27% higher accuracy analysed with existing techniques like intelligent waste management system using deep learning with IoT (IWMS-DL-IoT), deep learning-based waste detection in natural and urban environments.(DL-WD-NUE) and A lightweight multiscale convolutional neural network for garbage sorting.(LM-CNN-GSSC) respectively.