Extracting Golden Area from Image Based on Otsu Algorithm

Abstract:

The paper suggested an algorithm to extract the golden region with tone change in an image as a printing plate information of metallic ink. It is based on HSV color space. A statistical hue range of golden color was used to separate the golden regions from the image, and Otsu algorithm was applied to acquire the tone of golden regions. The golden target regions are acquired by multiplication pixel by pixel between processed H values and the processed S values. The experiments were carried out on various images with different textures and background characteristics, and the segmentation effect were compared and analyzed also. They showed that the suggested algorithm is suitable for both simple and complex texture area. The color difference between image’s background and target influences the result of extraction, he large color difference is more suitable for segmentation.

• Periodical: Applied Mechanics and Materials (Volume 469)
• Pages: 260-264
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.469.260
• Online since: November 2013
• Authors: Ying Wu*, Xiu Ping Zhao, Yang Jin, Yang Xu
• Keywords: Golden Color, HSV, Otsu Algorithm, Threshold Segmentation

References

[1] Zhang Zheng, Wang Yan Ping, Xue Gui Xiang, in: Digital Image Processing and Machine Vision —Visual C + + and Matlab to Achieve, Post &Telecom Press, Beijing (2010), in press.

[2] Li Zheng Ming, Wang Sen, Sun Jun: Transactions of the Chinese Society of Agricultural Machinery Vol. 40 (2009), p.105.

[3] Deng Hong Gui, Gao Xiang Rong: Microcomputer Information Vol. 24 (2008), p.199.

[4] Zhang Meng Tong. A Study of Spot Color Separating in Hue Centralized Image [D]. Wuhan: Wuhan University, (2005).

[5] Information on http: /zh. wikipedia. org.