AI Driven Waste Classification for Smart Recycling

Abstract

Having advanced technologies is critical to optimize the efficiency of sorting, as global waste generation continues to increase. This study categorizes waste types (plastic, glass, metal, organic, and hazardous waste) - the VGG16 architecture, a type of Convolutional Neural Network (CNN), is commonly utilized for image classification on images of discarded substances, which in turn minimizes human error, lowers labor costs, and improves retrieval efficiency. Furthermore, there are IoT deep learning applications that can be used on the internet, which enable real-time access to data for monitoring and collecting, offer insights into waste disposal patterns. These technologies help address urban waste challenges by improving recy-cling rates, minimizing landfill usage, and advocating for sustainable waste management options. This study focuses on deep learning frameworks for waste categorization, data collection obstacles, and their relevance in urban waste management, highlighting the use of AI in propelling for-ward environmental sustainability and resource recovery. This study utilizes data up to October 2023 to train the model.

Topics & Concepts