The extraction of effective recognition feature is important for high dimensional data pattern recognition.In order to solve the problem of low accura cy and high feature dimension for the present face recognition methods,a new face recognition algorithm based on exponential regularized null space linear discriminant analysis (NSLDA) is proposed in this paper.The NSLDA shows desirable performance.In this paper,the wi thin class sample scatter is regularized at first in order to avoid the small sample size p roblem,and then we utilize the matrix exponential to the discriminant criterion to enhance the perf ormance.The proposed method integrates the advantages of NSLDA in dimension reduction and re gularized linear discriminant analysis (RLDA) in pattern recognition,and effectively improves the accuracy of face recognition.