Title: Efficient fractal dimension calculation method for feature extraction of skin images

Authors: Ali Parchamy Araghy, Mie Sato, Masao Kasuga

Addresses: Graduate School of Engineering, Utsunomiya University, 7-1-2 Yoto, Utsunomiya, Tochigi 321-8585, Japan. ' Graduate School of Engineering, Utsunomiya University, 7-1-2 Yoto, Utsunomiya, Tochigi 321-8585, Japan. ' Graduate School of Engineering, Utsunomiya University, 7-1-2 Yoto, Utsunomiya, Tochigi 321-8585, Japan

Abstract: We applied the method of power spectrum to evaluate the state of the human skin using the analysis of textured images by fractal dimension. We extracted features from these skin images to classify the age and gender. For a good estimation, achieving pictures that include accurate information among taken pictures is necessary. However, in practical use the time required to find pictures that include accurate information is a problem. This paper presents an efficient method to calculate accurate fractal dimensions by extracting only relevant information from a picture. The fractal dimension is calculated continuously after cropping sections of a picture to include only the accurate information.

Keywords: fractal dimensions; power spectrum; skin age; skin gender; noise; image cropping; feature extraction; skin images; textured images; biometrics.

DOI: 10.1504/IJBM.2011.040814

International Journal of Biometrics, 2011 Vol.3 No.3, pp.175 - 188

Received: 08 Dec 2010
Accepted: 09 Dec 2010
Published online: 24 Jan 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article