[face recognization] How to exclude the input image that isn't in the training set

hi everyone

I use PCA for recognizing face . here is my code shortly:

[evectors, score, evalues] = princomp(images_train_eig');    
num_eigenfaces = 10; 
n_evectors = evectors(:, 1:num_eigenfaces);

% project the images into the subspace to generate the feature vectors
features = n_evectors' * shifted_images;

% calculate the similarity of the input to each training image
feature_vec = n_evectors' * (input_image(:) - mean_face);
similarity_score = arrayfun(@(n) 1 / (1 + norm(features(:,n) - feature_vec)), 1:num_images);

% find the image with the highest similarity
[match_score, match_ix] = max(similarity_score);

how can i recognize people that doesn't in my training set ,

Please answer me with keywords or any clues. Thank you