Original Article

Performance analysis of color matching technique for teeth classification based on color histogram

Justiawan Justiawan , Riyanto Sigit, Zainal Arief, Dian A. Wahjuningrum

Justiawan Justiawan
Department of Electrical Engineering, Politeknik Elektronika Negeri Surabaya, Indonesia. Email: dian-a-w@fkg.unair.ac.id

Riyanto Sigit
Department of Electrical Engineering, Politeknik Elektronika Negeri Surabaya Kampus PENS, Raya ITS, Sukolilo, Indonesia

Zainal Arief
Department of Electrical Engineering, Politeknik Elektronika Negeri Surabaya Kampus PENS, Raya ITS, Sukolilo, Indonesia

Dian A. Wahjuningrum
Department of Conservative Dentistry, Faculty of Dental Medicine, Universitas Airlangga, Surabaya, Indonesia
Online First: August 01, 2017 | Cite this Article
Justiawan, J., Sigit, R., Arief, Z., Wahjuningrum, D. 2017. Performance analysis of color matching technique for teeth classification based on color histogram. Journal of Dentomaxillofacial Science 2(2): 95-99. DOI:10.15562/jdmfs.v2i2.525


Objective: Color matching technique is one of the requirement in clinical dentistry. Using dental shade selection can help the dentist to determine the suitable color for the patients during fabrication of prosthesis. However the lack of dentists’ knowledge in color science due to many kinds of shade guide becomes a problem in the field of dentistry. So color matching technique by using digital images are feasible solution when suitable color features have been properly manipulated.

Material and Methods: Separating the color features of digital images into RGB and HSV feature spaces are the first step of this system. Due to many features in this step, it requires some classifier algorithm according to the shade type of teeth. In this paper, we proposed color teeth classification for dental shade selection using DT, NN, and K-nearest neighbors algorithm based on color histogram feature spaces.

Results: The result showed that on using the KNN algorithm it achieved 96.67% in learning process and 90% in testing process which had error level (0.183).

Conclusion: It proves that our proposed system have high similarity in color matching according to the learning and testing process.

No Supplementary Material available for this article.
Article Views      : 34
PDF Downloads : 15