Research on Pearl Detecting and Grading Based on Monocular Multi-view Machine Vision

Yi-ping Tang *

College of Information Engineering, Zhejiang University of Technology, Hangzhou, 310023, China

Shao-jie Xia

College of Information Engineering, Zhejiang University of Technology, Hangzhou, 310023, China

Zhi-liang Zhu

College of Information Engineering, Zhejiang University of Technology, Hangzhou, 310023, China

*Author to whom correspondence should be addressed.


Abstract

Aims: Pearl is a non-planar object and its surface has a certain degree of curvature. It is necessary to obtain the image of the entire surface of a pearl for the aim to achieve the grade based on size, shape, luster, blemish and color. This paper designed a detecting device to solve this problem and complete the pearl’s grading automatically.
Methodology: In this paper, an imaging apparatus of monocular multi-view was designed by putting a symmetric bucket cavity, which was made of 4 planar mirrors in front of the camera. The surface of a pearl will be captured from different perspectives using this device. Then a series of views which reflect the quality of a pearl could be achieved by image processing. Finally feature fusion will be utilized to determine the quality of the pearl.
Conclusion: Experimental results show that the image of the entire surface of a pearl could be obtained in a unified color system using the proposed device. Besides, the device accomplishes pearl online detecting and grading according to the quality indicators such as size, shape, luster, blemish and color.

Keywords: Machine vision, monocular multi-view, digital image processing, pearl quality detection, online grading


How to Cite

Tang, Yi-ping, Shao-jie Xia, and Zhi-liang Zhu. 2014. “Research on Pearl Detecting and Grading Based on Monocular Multi-View Machine Vision”. Current Journal of Applied Science and Technology 4 (15):2136-51. https://doi.org/10.9734/BJAST/2014/5544.

Downloads

Download data is not yet available.