实验科学与技术2025，Vol.23Issue(5)：54-59,6.DOI:10.12179/1672-4550.20240074
基于GoogLeNet卷积神经网络的智能垃圾分类系统设计
Design of an Intelligent Garbage Classification System Based on the GoogLeNet Convolutional Neural Network
摘要
Abstract
In view of the problems of existing garbage classification methods,such as tedious sorting,low efficiency and supervision difficulties,an intelligent garbage classification system based on the GoogLeNet convolutional neural network is designed to overcome the shortcomings of market products concerning automatic identification and disposal,simple operation and suitability for household use.Based on GoogLeNet convolutional neural network,a four-category recognition classification algorithm based on garbage classification is designed.The collected garbage data set is fine-tuned and trained to achieve effective recognition of garbage types.A Raspberry PI as the core controller is employed in this system,combined with a CSI camera to obtain images and equipped with image recognition algorithm to achieve garbage classification.An independently designed mechanical structure and control unit,integrating classification information,enables automatic garbage disposal.The test analysis shows that the system can accurately perform garbage type identification and disposal,thereby effectively enhancing the convenience and efficiency of garbage classification.This holds great significance for the practical implementation of garbage classification at the user level.关键词
GoogLeNet卷积神经网络/智能垃圾分类/图像识别算法/机械设计/自动投放Key words
GoogLeNet convolutional neural network/intelligent garbage classification/image recognition algorithm/mechanical design/automatic disposal分类
信息技术与安全科学引用本文复制引用
赵一璇,原晓楠,王戈,丁颖,刘镒搏..基于GoogLeNet卷积神经网络的智能垃圾分类系统设计[J].实验科学与技术,2025,23(5):54-59,6.基金项目
2022年西安交通大学课程思政专项研究项目(KCSZ202235) （KCSZ202235）
2023西安交通大学国家级大学生创新创业训练项目(XJ202310698022) （XJ202310698022）
西安交通大学2024年第二批产学合作协同育人项目(24CXHZ031). （24CXHZ031）
