Abstract: In order to improve the performance of the traditional face recognition algorithm based on PCA
( Principal Component Analysis ) and SVM ( Support Vector Machines ) ，we introduce an improved crossvalidation
algorithm with grid search method to optimize SVM parameters，which combins with PCA and SVM
algorithm． The improved algorithm uses the K-CV ( K-fold Cross Validation ) algorithm to optimize SVM
parameters，minimize the impact of individual sample errors on the prediction model，shorten the search time and
improve the face recognition rate． Compared to other PCA( Principal Component Analysis) and SVM( Support
Vector Machine) salgorithms，this algorithm has 9． 08% higher performance than tradition algorithm．

Key words: face recognition, cross-validation, grid search, principal component analysis ( PCA) , support vector machines ( SVM)