STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Image Processing & Pattern Recognition Progress

Automatic Facial Recognition Using Adaptive Modelling of Illumination Field for Shape Recovery of Micro Facial Features

  1. Rajveer Kour
  2. Tripti Sharma

Abstract

In this study, we present an innovative framework to resolve the issue of confounding effects caused by variation of illumination in the scenario of face recognition. The proposed framework involves a mathematically modelled adaptive algorithm which augments the pixels sets of facial images with realistically simulated illumination field. This enhances the recognition rate and subsequently its performance in a way independent of classifier mode for each of its facial parameters. Earlier methods usually require the manual intervention for setting the threshold parameters for each variant image with different illumination styles. However, the proposed method is advantageous as it enables automatic setting functions for such variants and promises to be superior to that of recent automatic methods. The performance of the framework is carried through state of the art benchmark tools available in the research community like: Multi-PIE and CMU PIE. Finally, we showed the comparison analysis of our model with that of the principal papers cited in the literature. Cite this ArticleRajveer Kour, Tripti Sharma. Automatic Facial Recognition Using Adaptive Modelling of Illumination Field for Shape Recovery of Micro Facial Features. Journal of Image Processing & Pattern Recognition Progress. 2015; 2(1): 19–24p.

Keywords

Support