Review of transactions on illuminated facial image preprocessing for user authentication

摘要

The problem in 2D face verification system is poor illumination condition which causes variation in facial image and degrades the performance of system, in such cases this system is a reliable solution for time crucial, high volume traffic locations that require tightened security, for instance airports, border checkpoints etc. Therefore, in this paper first of all we will use IRM (Illumination Reflection Model) to detach illumination and reflectance components and then secondly we will perform Discrete Wavelet Transform (DWT) and inverse of DWT (iDWT) on the image and then finally we will be doing the fusion of wavelet image (wIm) obtained from DWT and Wavelet Local Binary Patterns Image (wLBPIm) obtained after applying LBP in order to enhance the quality of the image and get the preprocessed image which ultimately decides whether the user is authenticated to access a particular system or not.