The invention provides a face recognition system based on PCA and linear discriminant analysis. The system realizes face detection based on Haar-like features and an AdaBoost algorithm by means of anOpenCV function library. The invention provides a face recognition algorithm with principal component analysis (PCA) and linear discriminant analysis (LDA) being combined, thereby not only preventingthe defects that a principal component analysis method does not carry out primary and secondary distinction on image information and ignores category information, but also reducing limitation of largeerrors and small samples caused by high computation amount of a linear discriminant analysis algorithm; and a simple and convenient face recognition system is realized.