Research on Intelligent Garbage Classification Device Based on the Fusion Technology of Differential Classification and Artificial Intelligence Recognition
Year: 2025
Type: article
Abstract: To solve the problem of the "last mile" in garbage classification, this study designs an intelligent garbage classification device based on artificial intelligence, machine vision and automation technologies. By constructing a multi-modal garbage identification model (YOLOv5s) and integrating it with a differential sorting mechanism, the functions of garbage type identification, sorting and compression are achieved. The experimental results show that the device's recognition accuracy rate is ≥98%, the single-piece processing cycle is ≤300ms, the compression rate of recyclable waste is ≥70%, the operating cost is reduced by 60%, and the sorting rate of hazardous waste is increased to 99%. This device overcomes the bottlenecks of traditional classification efficiency and accuracy, providing core infrastructure support for the construction of "waste-free cities" and contributing to the achievement of the "dual carbon" goals.
DOI: 10.54097/xq8bfs97
Authors: Ru Li, Ziyi Yang, Xinyu Liu, Tianyou Yu
Institution: Beijing University of Chemical Technology
Language: English
Subfield: Waste Management and Disposal
Field: Environmental Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7090711449