Technology based on face recognition to identify the person unique identity has been seeking the attention of the people in recent era. Face recognition system has wide range of application, so it is necessary to increase its the robustness. Various techniques have been discovered to overcome various factors like scale, pose, expression and illumination in order achieve better output. But there is no technique which is efficient for various practical cases, also which may serve factors simultaneously. Illumination was the biggest challenge for the techniques when applied in real world. Face recognition system has three stages: First stage is Face detection, second is feature extraction and the last stage is classification. A proper set of feature extraction may improve the performance of the system. Our proposed solution uses Viola Jones for face detection, PCA with Multilevel Grid search method is used for facial data processing and feature extraction. Further SVM is applied on data for classification. Proposed model identifies faces more accurately with the accuracy of 85% compared to current traditional method. Experimental results are measured based on the sensitivity, specificity and precision using the lfw-deep funneled dataset. © BEIESP.