The aim of this paper is to implement facial recognition for a dataset that has different illuminated images with variant poses for the purpose of human authentication using an low resolution camera available on the day-to-day electronic gadgets such as mobile phones, tablets and other hand-held devices The basic step in the face recognition from an image is to acquire only the face portion. The acquired image is then taken to the next process called feature extraction. After feature extraction of the image, the face recognition will be performed. The process of face recognition requires many very important aspects to be considered, i.e., the illumination and poses and angle. To accomplish this task, lighting controllers must tactically be employed to ensure that the correct current and timing controls are applied to obtain the probable lighting. A combination of the latest feature extraction and illumination compensation algorithm are used to encode micro-patterns giving an efficient description for face recognition, i.e., Oriented Local Histogram Equalization (OLHE), which has proven to perform exceptionally high under extreme lighting conditions along with the previous state-of-the-art algorithms such as Bit Plane Slicing, Gabor Filter, Local Binary Pattern (LBP) and Local Gradient Oriented Binary Patterns (LGOBP) at the same time as they encode micro-patterns which gives an efficient description for face recognition. The goal is to explore and analyze the performance of the following face Recognition algorithms, namely PCA (Principal Component Analysis), LDA (Linear Discriminate Analysis), CCA (Canonical Correlation Analysis), AAM (Active Appearance Model) and SVM. (Support Vector Machine). It has been proposed that the illumination compensation algorithm OLHE for face recognition is to be utilized. The combination of CCA and OLHE will give the highest recognition rates among the five feature extraction and face recognition algorithms taken into consideration. Upon analysing their performance on the datasets such as FERET, ORL, CMU-PIE, EXTENDED YALE B and newly created VIT DATABASE, this combination is found to be the handpicked illuminated compensation algorithm. These images from the VIT dataset along with the other images from the various open-source databases are subjected to various pre-processing and post-processing methods of feature extraction and the results are tabulated and they are compared. © 2006–2016. Asian Research Publishing Network (ARPN). All rights reserved.