Improvement of Face Recognition Rate under uneven Illumination depending on NPCA

In the last years, researches have started developing the facial recognition systems after it has been widely used in both private and business environments. Despite the continuous work on the development of face recognition methods and the reduction of the error percentage, they are still influenced by many factors such as the age of the people, the orientation of the faces, the accuracy of the images and their illumination. Therefore, this paper proposes a system for processing images with poor lighting and images with uneven illumination, by creating processed images based on both the original image in the database and the modified images using Histogram Equalization after performing an N operation on them, where these images are used in Eigenfaces algorithm for the purpose of distinguishing faces, where the experimental results show on the Extended Yale B database that this method is highly effective for addressing the problem of poor lighting and uneven lighting of images, that using Histogram Equalization to process images with Normalized Eigenfaces gave results equal to (90.2255%) when used with images with uneven lighting, while the proposed method gave results of up to (97.7443%) in the same lighting conditions.