A face recognition method based on aggregated image was proposed. In this paper, aggregated images were formed by stacking images with multiple face images. The aggregated images were used as the training images for feature detection and description. Finally, the features were matched with features extracted from the test images. Compared with other face recognition methods, the experiments indicate that this method has obvious advantages in the recognition rate. Even though the using a single face image is still able to get a higher recognition rate.Using aggregated images, the number of feature points is increased with the increase of the number of images used by the aggregated images. That enables the test images to match more appropriate feature points in the training images and increase the recognition rate.