Please use this identifier to cite or link to this item:
Title: Age prediction based on skin
Authors: Asvega.
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Issue Date: 2010
Abstract: Skin is the most easily seen organ. The different visual properties of skin could be attributed to physio-anatomical differences in between individuals. There are several factors that could affect the skin visual properties, including but not limited to environment and hormone. The work presented in this thesis is an experiment to determine whether it is feasible to classify if a skin sample belongs to a child below up to 13-year-old or to an adult above 13-year-old. The sum of hair and average hair density on a skin sample is used to classify. This is because hair is noticeably affected by the change in hormone during puberty. For determining the classification performance of the system two databases of digitalised skin sample photographs were used. The photographs were manually processed to produce the desired skin sample databases. The hair map was automatically extracted from a skin sample to be processed to produce sum of hair and average hair density from the respective skin sample. The sum of hair and average hair density will then be used to classify whether the skin sample belongs to children or adult using Support vector machine classification. Probability distribution function and Gaussian function were then employed to find out whether the probability of correct classification is at least 60% accurate. The experiment was performed on 160 skin samples photographs. The probability of correct classification of the age group was at least 68.75%.
Schools: School of Computer Engineering 
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
2.43 MBAdobe PDFView/Open

Page view(s) 50

Updated on Sep 29, 2023

Download(s) 50

Updated on Sep 29, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.