Face recognition using non-negative matrix factorization with a single sample per person in a large database
- PDF / 1,023,938 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 30 Downloads / 158 Views
Face recognition using non-negative matrix factorization with a single sample per person in a large database F. Nikan 1 & H. Hassanpour 1 Received: 20 July 2019 / Revised: 29 June 2020 / Accepted: 21 July 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
There are various face recognition techniques in literature, which are faced with challenges such as occlusion, pose variation, illumination, and facial expressions. Existing methods often perform well when their database is small, or multiple samples per person exist. However, face recognition methods with just one reference sample per person may not work well, especially on a large database. To address this problem, this paper proposes a scheme to extract features from facial images. Using Non-negative Matrix Factorization (NMF), basic features are extracted from the face structure. The matrix of images is decomposed into basis matrix (W) and weight matrix (H). The basis matrix contains several versions of mouths, noses and other facial parts, where the various versions are in different locations or forms. Hence, to recognize a facial image in the database, searching is done on the weight matrices feature set. In this research, to more precisely form the structural elements, a separate basis matrix is constructed for the upper and lower parts of the facial images from the database. Also the images are enhanced using pre-processing techniques including histogram equalization, image intensity, and contrast limited adaptive histogram. The FERET database with 990 single images per person was used to evaluate the proposed method. Experimental results show that the proposed method can achieve a recognition rate close to 93%. Keywords Face recognition . Non-negativematrix factorization . Single sample per person . Large database . Recognition rate
1 Introduction Face recognition (FR) is used in a variety of applications such as security maintenance, image processing, video surveillance, and criminal identification. Single Sample per Person (SSPP),
* H. Hassanpour [email protected]
1
Department of Computer Engineering, Shahrood University of Technology, Shahrood, Iran
Multimedia Tools and Applications
especially in a large database such as one used for passport identification, is one of the most important challenges in FR [31]. When only one sample per person is available for training, FR becomes more difficult [30]. Although several studies have been done on this topic, it is still an active challenge [7, 10]. As mentioned, one of the active subjects in face recognition is SSPP [7], and some SSPP face recognition methods were proposed in the literature. These methods can be divided into five categories: direct methods, generic learning methods, patch-based methods, expanding sample methods, and deep learning methods [35]. Direct methods perform based on SSPP by applying an algorithm directly on the facial images [35]. Generic learning methods use an auxiliary dataset to make multiple samples for each person to extrac
Data Loading...
