Face Recognition Based On Sparse Coding With Feature Enhancement

Compared to other biometric identification technologies,face recognition has obvious advantages.Face recognition system has a wide range of applications in identification,self-help service,information security and video surveillance.In ideal conditions,the existing face recognition systems have good robustness.But,if test images are affected by uncontrolled factors,including obvious light and posture changes,sunglass occlusion etc,performance of the recognition system will degrade abruptly.Over the last two decades,the research of face recognition under uncontrolled scene have been a hotspot in the field of pattern recognition.In recent years,face recognition based on sparse representation classification has received attention due to its good robustness.Although the sparse representation classification has a significant effect on solving face recognition that is contaminated by Gaussian noise,its performance when the image is damaged by strong noise is greatly reduced.For robust face recognition under strong noise interference conditions,we present a face recognition method based on sparse representation dictionary and low-rank representation.The main work and contributions of this article include:(1)Face recognition method based on sparse coding and low-rank representation.Mainly completed the experimental simulation of the relevant classification algorithm and low-rank matrix recovery algorithm,and laid the foundation for the follow-up research work.First,the basic theory and specific algorithms of several feature extraction algorithms,sparse coding,low rank decomposition are deeply studied;Then the global or local feature extraction algorithm is applied to the unobstructed face image,and the corresponding classification algorithm is used to simulate the system.The performance of each classification algorithm is compared;Finally,the simulation of occlusion face recognition is performed and the robustness of each algorithm is analyzed.(2)Face Recognition Based on Sparse Coding with Feature Enhancement.In view of the fact that low-rank matrix recovery can effectively remove image noise,and for noise-contaminated face recognition problems,a sparse-coded face recognition method based on low-rank matrix recovery feature enhancement is proposed.The key to the algorithm is to find a better denoising algorithm to recover a clean face image.For robust principal component analysis only applies to a single face subspace,while low rank indicates that only the matrix's main information is recovered from the column direction and the matrix's significant information is not taken into account.This paper uses an implicit low-rank representation to obtain a more comprehensive Face information.A large number of experimental results show that the proposed method has a better recognition effect and is more robust to noise.