Material classification technology based on Convolutional neural networks

Volume 4, Issue 5, October 2019     |     PP. 176-189      |     PDF (595 K)    |     Pub. Date: August 21, 2019
DOI:    259 Downloads     6583 Views  

Author(s)

Dailin Li, College of Science, China University of Petroleum (East China), Qingdao 266580, China
Guilei Li, College of Science, China University of Petroleum (East China), Qingdao 266580, China
Baojun Wei, College of Science, China University of Petroleum (East China), Qingdao 266580, China
Dan Yang, College of Science, China University of Petroleum (East China), Qingdao 266580, China
Ning Wang, College of Science, China University of Petroleum (East China), Qingdao 266580, China
Huafeng Zhu, College of Science, China University of Petroleum (East China), Qingdao 266580, China
Hao Ni, College of Science, China University of Petroleum (East China), Qingdao 266580, China

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%.

Keywords
Material classification; Convolutional neural networks; Polarization imaging; HSV color model

Cite this paper
Dailin Li, Guilei Li, Baojun Wei, Dan Yang, Ning Wang, Huafeng Zhu, Hao Ni, Material classification technology based on Convolutional neural networks , SCIREA Journal of Physics. Volume 4, Issue 5, October 2019 | PP. 176-189.

References

[ 1 ] Song W. H., Huo J. R., “Research on computer object recognition based on shape,” Information Technology. Papers (6):188-190, (2016).
[ 2 ] Fan, Y., Lang , B., “An Object Shape-matching Method Using Contour Orientation Feature,” Computer Technology and Development. Papers (4):82-86 (2018).
[ 3 ] Zhang H. P., Jiang Z. G., “Multi-view space object recognition and pose estimation based on kernel regression,” Chinese Journal of Aeronautics. Papers 27(05):1233-1241 (2014).
[ 4 ] Han P. L., Liu F., Zhang G., et al, “Multi-scale analysis method of underwater polarization imaging,” Acta Physica Sinica. Papers (5):124-134 (2018).
[ 5 ] Zhou T., Huang D. F., Wang H. M., et al, “Detection Method of Hepatocellular Carcinoma Based on Polarization Properties of Biological Tissue,” Science Technology and Engineering. Papers (7):91-95 (2018).
[ 6 ] Sun Z., Han T. S., Jiang J. Y., et al, “Study on Surface Reflectance Light Elimination of Biological Tissue with Cross-Polarization,” Spectroscopy and Spectral Analysis. Papers (11):3520-3524 (2017).
[ 7 ] Wang X., Zhou X. F., Jin W. Q., “Study of Polarization Properties of Radiation Reflected by Roughness Objects,” Transactions of Beijing Institute of Technology. Papers 31(11):1327-1331 (2011).
[ 8 ] Lu S. J., Li J. Q., Zhang X. L., “Outline Enhancement Technology of Target Based on Polarization Imaging,” Ordnance Industry Automation. Papers (9):66-67,96(2017).
[ 9 ] Li X. L., Li Y. Y., Xie X. H., Xu L. J., "Laser polarization imaging models based on leaf moisture content," Infrared and laser engineering. Papers 46(11):121-126 (2017).
[ 10 ] Peng B., Huang S. L., Li D. J., “Detection of colorless plastic contaminants hidden in cotton layer using chromatic polarization imaging,” Chinese Optics Letters. Papers 13(09):81-85 (2015).
[ 11 ] Cai W. J., Wang L. M., “Recognition of Chinese characters using deep convolutional neural network,” Journal of Image and Graphics. Papers (3):410-417 (2018).
[ 12 ] Lu Z. G., Liu Q. S., Sun Y. B., “Large-Scale Face Image Retrieval based on Deep Residual Embedding Feature,” Journal of Taiyuan University of Technology. Papers (1):106-112 (2018).
[ 13 ] Cai S. Q., “A Study on Rapid Image Super-resolution,” Infrared Technology. Papers (3):269-274 (2018).
[ 14 ] Lu H., Fu X., Liu C., et al, “Cultivated land information extraction in UAV imagery based on deep convolutional neural network and transfer learning,” Journal of Mountain Science. Papers 14(04):731-741 (2017).
[ 15 ] Zhang G. Q., Li Z. M., Li X. W., “Research on color Image segmentation in HSV space,” Computer Engineering and Aoolications. Papers (26):179-181 (2010).
[ 16 ] Li D. L., Yu Y., Li G. L., et al, “The Study of Underwater Material Recognition Techonlogy,” Laser & Optoelectronics Progress. Papers 55(7):071010 (2018).
[ 17 ] Yin S. H., Deng J. C., Zhang D. W., et al, “Traffic sign recognition based on deep convolutional neural network,” Optoelectronics Letters. Papers 13(06):476-480 (2017).
[ 18 ] Chen H. C., “A Method of vehicle color recognition based on deep convolutional neural networks,” Journal of the Hebei Academy of Sciences. Papers 34(02):1-6 (2017).