Intelligent Garbage Sorting Device Based on an Improved YOLOv5 Detection Algorithm

Bowu Ren; Tansheng Wang; Xilin Ge; Mile Li; Yihan Gao; Hongming Tang

Intelligent Garbage Sorting Device Based on an Improved YOLOv5 Detection Algorithm

Bowu Ren
, Tansheng Wang
, Xilin Ge
, Mile Li
, Yihan Gao
, Hongming Tang^*

^*Corresponding author for this work

Department of Physics

Xi'an Jiaotong-Liverpool University
Soochow University
New York University Shanghai

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

Abstract

The rapid growth of municipal solid waste poses significant challenges to sustainable urban development, necessitating intelligent and efficient waste sorting solutions. This paper presents a smart waste sorting system integrating deep learning visual recognition, edge computing, and automated control technologies. Through the optimization of the YOLOv5 object detection model, the system attains realtime, accurate identification of recyclable waste, food waste, hazardous waste, and other waste categories. Even under difficult illumination and occlusion scenarios, the system, which is based on the NVIDIA Jetson Nano edge computing platform and uses STM32 microcontrollers for coordinated control, retains a high recognition accuracy. In most settings, test results show strong performance and real-time capability. In order to provide a comprehensive and scalable solution for smart city waste management, the system also includes a human-machine interface and ultrasonic sensors to precisely monitor the fill level of smart garbage cans.

Original language	English
Title of host publication	2025 International Conference on Computer, Internet of Things and Smart City (CIoTSC)
Publisher	IEEE
ISBN (Electronic)	979-8-3315-5522-1
ISBN (Print)	979-8-3315-5523-8
Publication status	Published - 3 Mar 2026

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities
SDG 12 Responsible Consumption and Production

Cite this

@inproceedings{734b2684ccf94f939fa72f31348885bc,

title = "Intelligent Garbage Sorting Device Based on an Improved YOLOv5 Detection Algorithm",

abstract = "The rapid growth of municipal solid waste poses significant challenges to sustainable urban development, necessitating intelligent and efficient waste sorting solutions. This paper presents a smart waste sorting system integrating deep learning visual recognition, edge computing, and automated control technologies. Through the optimization of the YOLOv5 object detection model, the system attains realtime, accurate identification of recyclable waste, food waste, hazardous waste, and other waste categories. Even under difficult illumination and occlusion scenarios, the system, which is based on the NVIDIA Jetson Nano edge computing platform and uses STM32 microcontrollers for coordinated control, retains a high recognition accuracy. In most settings, test results show strong performance and real-time capability. In order to provide a comprehensive and scalable solution for smart city waste management, the system also includes a human-machine interface and ultrasonic sensors to precisely monitor the fill level of smart garbage cans.",

author = "Bowu Ren and Tansheng Wang and Xilin Ge and Mile Li and Yihan Gao and Hongming Tang",

year = "2026",

month = mar,

day = "3",

language = "English",

isbn = "979-8-3315-5523-8",

booktitle = "2025 International Conference on Computer, Internet of Things and Smart City (CIoTSC)",

publisher = "IEEE",

}

TY - GEN

T1 - Intelligent Garbage Sorting Device Based on an Improved YOLOv5 Detection Algorithm

AU - Ren, Bowu

AU - Wang, Tansheng

AU - Ge, Xilin

AU - Li, Mile

AU - Gao, Yihan

AU - Tang, Hongming

PY - 2026/3/3

Y1 - 2026/3/3

N2 - The rapid growth of municipal solid waste poses significant challenges to sustainable urban development, necessitating intelligent and efficient waste sorting solutions. This paper presents a smart waste sorting system integrating deep learning visual recognition, edge computing, and automated control technologies. Through the optimization of the YOLOv5 object detection model, the system attains realtime, accurate identification of recyclable waste, food waste, hazardous waste, and other waste categories. Even under difficult illumination and occlusion scenarios, the system, which is based on the NVIDIA Jetson Nano edge computing platform and uses STM32 microcontrollers for coordinated control, retains a high recognition accuracy. In most settings, test results show strong performance and real-time capability. In order to provide a comprehensive and scalable solution for smart city waste management, the system also includes a human-machine interface and ultrasonic sensors to precisely monitor the fill level of smart garbage cans.

AB - The rapid growth of municipal solid waste poses significant challenges to sustainable urban development, necessitating intelligent and efficient waste sorting solutions. This paper presents a smart waste sorting system integrating deep learning visual recognition, edge computing, and automated control technologies. Through the optimization of the YOLOv5 object detection model, the system attains realtime, accurate identification of recyclable waste, food waste, hazardous waste, and other waste categories. Even under difficult illumination and occlusion scenarios, the system, which is based on the NVIDIA Jetson Nano edge computing platform and uses STM32 microcontrollers for coordinated control, retains a high recognition accuracy. In most settings, test results show strong performance and real-time capability. In order to provide a comprehensive and scalable solution for smart city waste management, the system also includes a human-machine interface and ultrasonic sensors to precisely monitor the fill level of smart garbage cans.

M3 - Conference Proceeding

SN - 979-8-3315-5523-8

BT - 2025 International Conference on Computer, Internet of Things and Smart City (CIoTSC)

PB - IEEE

ER -