PCA based improved face recognition system

Fayaz Ali Dharejo, Munsif Ali Jatoi, Zongbo Hao, Majid Ali Tunio

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

Due to the rapid evolution of modern image processing and pattern recognition techniques, there exists a variety of biometric techniques like fingerprints, iris (retina) scans, and speech recognition etc. nowadays. However, among them, face recognition is still the most common technique which is in use due to the fact that it is easy to install and has less complexity. It has been a prominent research field for security applications such as video surveillance, fraud detection, person tracking and crowd recognition. This research work discusses and implements the facial recognition by using MATLAB environment in real time. In this work, the face recognition system is implemented using Principal Component Analysis (PCA) and Eigenface approach by dealing with large dataset. A methodology is proposed to produce more accuracy and efficiency by removing the unrelated space from the image. Also, a graphical Interface System (GUI) is developed in order to make our system clearer and to measure the training time. Furthermore, large databases (ORL and private Database named face-100) are tested through PCA and Eigenfaces approach and then the person identification is carried out. The system successfully recognized the human faces and worked better in different conditions like illumination and blur conditions of the face. The rate of male and female accuracy is also calculated.

Original languageBritish English
Title of host publicationInformation Technology and Intelligent Transportation Systems - Proceedings of the 2nd International Conference on Information Technology and Intelligent Transportation Systems, ITITS 2017
EditorsValentina Emilia Balas, Xiangmo Zhao, Fuqian Shi, Lakhmi C. Jain
Pages429-440
Number of pages12
ISBN (Electronic)9781614997849
DOIs
StatePublished - 2017
Event2nd International Conference on Information Technology and Intelligent Transportation Systems, ITITS 2017 - Xi'an, China
Duration: 10 Jun 2017 → …

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume296
ISSN (Print)0922-6389

Conference

Conference2nd International Conference on Information Technology and Intelligent Transportation Systems, ITITS 2017
Country/TerritoryChina
CityXi'an
Period10/06/17 → …

Keywords

  • Cropped function
  • Facial images
  • Gender recognition
  • GUI
  • Large database
  • PCA
  • Recognition

Fingerprint

Dive into the research topics of 'PCA based improved face recognition system'. Together they form a unique fingerprint.

Cite this