A PATTERN RECOGNITION SYSTEM FOR HUMAN AGE IDENTIFICATION
Vidyashree H S
, Pradeep N
Face images, Features, Feature Descriptors, Classifier, Data sets.
Identification of human age can be accessed based on their facial characteristics. Each face contains special unique information which can be used in the identification of an individual’s age. To identify age of an individual from the input, some of fundamental steps to be performed are pre-processing, feature extraction and a classification technique. In our work, we have used the traditional KNN algorithm as classifier. The inputs are acquired from publically available dataset and by constructing local dataset by capturing faces of individuals’ from different orientation and under different illumination using good resolution camera. In our work, the popular GLCM and LBP feature extraction techniques have been adopted. The extracted features are maintained in the feature matrix which will be fed as input to the classifier for identifying person age. In this paper, two datasets are considered for our experiments. One is publically available MORPH dataset and another is local dataset constructed by capturing different person’s face images using mobile camera. An attempt is made to develop a usable system that will identify the person age based on the face features.
"A PATTERN RECOGNITION SYSTEM FOR HUMAN AGE IDENTIFICATION", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.3, Issue 7, page no.242 - 249, July-2018, Available :https://ijsdr.org/papers/IJSDR1807041.pdf
Volume 3
Issue 7,
July-2018
Pages : 242 - 249
Paper Reg. ID: IJSDR_180500
Published Paper Id: IJSDR1807041
Downloads: 000346998
Research Area: Engineering
Country: Davanagere, Karnataka, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave