Home // International Journal On Advances in Software, volume 10, numbers 3 and 4, 2017 // View article

Accurate and Robust Skin Feature Extraction Scheme for Aging Estimation

Authors:
Hyungjoon Kim
Jisoo Park
Woogeol Kim
Eenjun Hwang

Keywords: Skin analysis; Feature extraction; Wrinkle feature; Contrast stretching; Microscopy image.

Abstract:
In this paper, we revise our method for skin feature extraction based on cell segmentation to improve its accuracy, efficiency and robustness that are very critical in the realistic skin condition analysis. In order to achieve such goals, we enhance the contrast of wrinkles on the skin by using the contrast limited adaptive histogram equalization (CLAHE) and highlight the depth of wrinkles by using the extended-minima transform. By performing watershed transform, we can segment the skin image into labelled skin cells and calculate various skin features from the labelled cells. We focus on two types of skin features that play a key role in assessing the degree of skin aging; cell features and wrinkle features. To evaluate the performance and robustness of our revised method, we collected skin images using three different types of microscopy cameras and extracted their cell and wrinkle features using the revised method. Through various experiments, we show that our revised method achieves 10% increase in the skin feature extraction accuracy and 50% decrease in the skin

Pages: 335 to 344

Copyright: Copyright (c) to authors, 2017. Used with permission.

Publication date: December 31, 2017

Published in: journal

ISSN: 1942-2628