Intelligent classification trash bin based on deep learning

Guanyi Li; Yuqi Ren; Na Chu; Xinyue Zhou; Jiaheng Wan; Qiongyan Li

doi:10.1117/12.2646632

14 November 2022 Intelligent classification trash bin based on deep learning

Guanyi Li, Yuqi Ren, Na Chu, Xinyue Zhou, Jiaheng Wan, Qiongyan Li

Proceedings Volume 12299, International Conference on Sustainable Technology and Management (ICSTM 2022); 122990R (2022) https://doi.org/10.1117/12.2646632
Event: International Conference on Sustainable Technology and Management (ICSTM 2022), 2022, Macao, China

Abstract

In order to solve the problems of low accuracy and low labor efficiency in garbage classification, an intelligent classification trash bin equipped with robot arm and 110-degree wide-angle camera is designed. A fast identification and positioning algorithm based on deep learning is proposed. The PASCAL VOC 2007 dataset is applied, which consists of 600 images containing 4000 different types of garbage in different situations. 20% of the images is randomly selected as the validation set, and the remaining 80% as the training set. The dataset is trained using the YOLOv5 model. After training, the model is tested on the garbage can system, automatically identification and classification the type of garbage is achieved and the robotic arm realizes the automatic positioning, grabbing and sorting. The experimental test results show that the error of the positioning algorithm is less than 0.8 mm, the fastest recognition speed is 2.6ms/piece, and the recognition accuracy can be maintained above 90% in the experimental environment, indicating that the algorithm has good accuracy and stability.

Citation Download Citation

Guanyi Li, Yuqi Ren, Na Chu, Xinyue Zhou, Jiaheng Wan, and Qiongyan Li "Intelligent classification trash bin based on deep learning", Proc. SPIE 12299, International Conference on Sustainable Technology and Management (ICSTM 2022), 122990R (14 November 2022); https://doi.org/10.1117/12.2646632

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available