Home // MMEDIA 2017, The Ninth International Conferences on Advances in Multimedia // View article

Improving Feature Extraction Accuracy for Skin Analysis

Authors:
Woogeol Kim
Hyungjoon Kim
Eenjun Hwang

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

Abstract:
In this paper, we revise our skin feature extraction method based on cell segmentation to improve the accuracy and efficiency of skin feature analysis. Accurate skin feature extraction is critical for the evaluation of skin conditions. In order to improve the accuracy of analyzing skin features, we use the contrast limited adaptive histogram equalization (CLAHE) method for contrast enhancement. Also, we use the extended-minima transform in order to enhance the depth of wrinkles on the skin. After conducting watershed transform for cell segmentation, we utilize the labeled information of skin cells to extract skin features. We consider two types of skin features that are important for estimating the skin age of users. They are cell features and wrinkle features. To evaluate the performance of our revised method, we collected diverse images using two types of microscopy cameras and estimated the skin age based on their skin features. Through various experiments, we show that our revised method achieves 11% increase in analysis accuracy and 53% decrease in feature extraction time compared to our previous work.

Pages: 26 to 31

Copyright: Copyright (c) IARIA, 2017

Publication date: April 23, 2017

Published in: conference

ISSN: 2308-4448

ISBN: 978-1-61208-548-7

Location: Venice, Italy

Dates: from April 23, 2017 to April 27, 2017