ML Anthology

Diamond Color Grading Based on Machine Vision

Zhiguo Ren, Jiarui Liao, Lilong Cai
ICCVW 2009 pp. 1970-1976
doi:10.1109/ICCVW.2009.5457523 /iccvw/2009/ren2009iccvw-diamond/

Abstract

This paper presents an effective method for diamond color grading based on machine vision. In order to acquire satisfactory diamond images, a special light source based on an integrating sphere is employed. After compensating the fluctuation of the light source, the compositive color features, including independent and joint distribution features of Hue and Saturation, are extracted in segmented uniform regions. Then, depending on a trained BP Neural Network, diamonds can be graded by color. Experiment results show that the proposed method can reach a satisfactory accuracy to replace manual grading for real diamonds. The proposed method can also be used to classify other objects by small color difference.

ICCVW Semantic Scholar

Cite

Text

Ren et al. "Diamond Color Grading Based on Machine Vision." IEEE/CVF International Conference on Computer Vision Workshops, 2009. doi:10.1109/ICCVW.2009.5457523

Markdown

[Ren et al. "Diamond Color Grading Based on Machine Vision." IEEE/CVF International Conference on Computer Vision Workshops, 2009.](https://mlanthology.org/iccvw/2009/ren2009iccvw-diamond/) doi:10.1109/ICCVW.2009.5457523

BibTeX

@inproceedings{ren2009iccvw-diamond,
  title     = {{Diamond Color Grading Based on Machine Vision}},
  author    = {Ren, Zhiguo and Liao, Jiarui and Cai, Lilong},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2009},
  pages     = {1970-1976},
  doi       = {10.1109/ICCVW.2009.5457523},
  url       = {https://mlanthology.org/iccvw/2009/ren2009iccvw-diamond/}
}