A Research on Front-End Garbage Classification Based on Machine Vision

被引：0

作者：

Gao, Longyu ^{[1
]}

Xiao, Zhiqing ^{[1
]}

Hao, Junlong ^{[1
]}

Shen, Luqi ^{[1
]}

Hu, Manqian ^{[1
]}

机构：

[1] Tianjin Univ Sci & Technol, Tianjin, Peoples R China

来源：

PROCEEDINGS OF THE 2021 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS (ICAROB 2021) | 2021年

关键词：

machine vision; intelligent classification; convolutional neural network;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Adding a machine vision recognition module to the traditional smart trash can can effectively improve the efficiency of trash recognition. The intelligent garbage classification model constructed by the convolutional neural network can accurately identify the types of garbage, with an average accuracy rate of 0.87. Deploy the trained model on openMV and test it on the produced physical trash can. After the system is stable, the average time to complete a sorting and recovery is 2s. Experiments show that the system can effectively identify the types of garbage and complete garbage classification and recycling.

引用

页码：720 / 723

页数：4