Enhancing Face Identification Based On LBP-KNN And CNN-SVM

The human face plays an important role in our social interaction,conveying people's identity.Using the human face as a key to security,biometric passwords technology has received significant attention in the past several years due to its potential for a wide variety of applications.Faces can have many variations in appearance(aging,facial expression,illumination,inaccurate alignment and pose)which continue to cause poor ability to recognize identity.While very promising results and considerable progresses have been shown(made)on face recognition related problems,age-invariant face recognition still remains a major challenge in real world applications of face recognition systems with less research.The purpose of our research work is to provide an approach that contributes to resolve face identification issues with large variations of parameters such as pose,illumination,age-invariant and expression.In order to resolve those problems,the next sections elaborate and explain carefully the different algorithms use to demonstrate the effectiveness of the proposed model.For provable outcomes across pose,illumination,and expression,we combined two algorithms:(a)robustness local binary pattern(LBP),used for facial feature extractions;(b)k-nearest neighbor(K-NN)for image classifications.Our experiment has been conducted on the CMU PIE(Carnegie Mellon University Pose,Illumination,and Expression)face database and the LFW(Labeled Faces in the Wild)dataset.The proposed identification system shows higher performance,and also provides successful face similarity measures focus on feature extractions.This study addition the novel use and effectiveness of Deep Learning in general and Convolutional Neural Networks(CNN),in particular with Support vector Machine(SVM)approach to face recognition subject for aging invariant.The images for each subject taken at various times are treated as a single set,which is then compared to sets of images belonging to other subjects.The facial features are extracted using a Convolutional Neural Networks characteristic of Deep Learning.Here,a CNN architecture using the VGG-Face deep(neural network)is found to produce highly discriminative and interoperable features that are robust to aging variations even across the different biometric datasets(large or small,constraint or unconstraint).The facial appearance of a person changes over time because of the aging process,which results in significant intra-class variations;therefore,the extracted features show high inter-class and low intra-class variability leading to low generalization errors on aging datasets using ensembles of subspace discriminant classifiers.SVM-based face recognition method then perfectly achieve the image classifications.Extensive experiments are conducted on two well-known public domain face aging datasets: MORPH Album2 and FGNET show the effectiveness of the proposed approach.The obtain outcomes show that,it is easier to recognize older subjects from younger ones rather than younger subjects from older ones.The ultimate aims of this study are to present concrete facts related to research activities in facial identification across several parameters such as pose,illumination,age-invariant and expression.The adopted main methodologies during the past decade are presented,a comprehensive list of benchmark results and most importantly provide roadmaps for future trends are also highlighted,in order to give some requirements and research directions in facial recognition.