Share Email Print
cover

Proceedings Paper • new

Material classification technology based on convolutional neural networks
Author(s): Dailin Li; Guilei Li; Baojun Wei; Dan Yang; Ning Wang; Huafeng Zhu; Hao Ni
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The contact measurement techniques are typically used in the field of object material classification. It has a lot of disadvantages, such as the complex operation and time-consuming. In this paper, a new non-contact object material identification method based on Convolutional neural networks (CNNs) and polarization imaging is proposed. Firstly, the relationship between the complex refractive index of object and the polarization information is simulated, and then the structure of the CNNs is constructed according to the specific conditions of the polarization imaging system. The accuracy of the identification method is measured by repeated test using 7 materials. The experimental results show that the CNNs model can quickly realize the object material classification with the polarization images, and the classification accuracy is above 92%.

Paper Details

Date Published: 12 March 2019
PDF: 10 pages
Proc. SPIE 11023, Fifth Symposium on Novel Optoelectronic Detection Technology and Application, 1102331 (12 March 2019); doi: 10.1117/12.2521860
Show Author Affiliations
Dailin Li, China Univ. of Petroleum (China)
Guilei Li, China Univ. of Petroleum (China)
Baojun Wei, China Univ. of Petroleum (China)
Dan Yang, China Univ. of Petroleum (China)
Ning Wang, China Univ. of Petroleum (China)
Huafeng Zhu, China Univ. of Petroleum (China)
Hao Ni, China Univ. of Petroleum (China)


Published in SPIE Proceedings Vol. 11023:
Fifth Symposium on Novel Optoelectronic Detection Technology and Application
Qifeng Yu; Wei Huang; You He, Editor(s)

© SPIE. Terms of Use
Back to Top