Invariant Moment Based Neural Network Classifier for Face Recognition
Satya Sreedevi Redla1, Vamsi Krishna Mangalampalli2, Benitamani Mallik3
1Satya Sreedevi Redla, Jackson State University, John R. Lynch Street, Jackson, MS & Research Scholar, Centurion University of Technology & Management, (Odisha), India.
2Dr. Vamsi Krishna Mangalampalli, Chaitanya Institute of Science & Technology, Kakinada (Andhra Pradesh), India.
3Dr. Benitamani Mallik, Centurion University of Technology & Management, (Odisha), India.
Manuscript received on 12 February 2019 | Revised Manuscript received on 02 March 2019 | Manuscript Published on 08 June 2019 | PP: 52-58 | Volume-7 Issue-5S4, February 2019 | Retrieval Number: E10110275S419/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Comparing the selected features of a digital facial image with those images in database is considered as face recognition system. Several methods have been developed in last few decades based on biometric artificial intelligence. The variability in the angle of facial images and facial expression posed challenging problem in recognition system. This paper develops face recognition methodology based on 7 invariant moments with respect to rotation, translation and scaling using Neural Network classifier. Methodology and demonstrations are being provided. Bio id face benchmark data base is used for the proof of concept. It is noticed that 97% accuracy is attained on randomly selected sample of 10 individual’s faces.
Keywords: Face Recognition, Face Detection, Feature Extraction, Hu Moments, Neural Network Classifier.
Scope of the Article: Pattern Recognition