Abstract: The number of features in the face recognition has an important impact in the recognition stage. Hence, such important issue in the face recognition methods need to be solved by choosing accurate features. Here, a new face feature extraction method with optimal features extraction via. PCA and Genetic algorithm techniques is proposed. The proposed method primarily performs features extraction using PCA and the Genetic algorithm is used to obtain the optimal features. The optimal features from PCA+GA is used efficiently to perform the face recognition process. To validate the proposed feature selection mechanism, the neural network Feed forward back propagation neural network has been used for the recognition purpose. The human face dataset called CASIA is used to examine the performance of our proposed PCA+GA technique. The results showed the proposed method is very effective in choosing the optimal features and has been reflected that in the recognition process.