Professional Documents
Culture Documents
ABSTRACT: A facial recognition system is one of the biometric application for automatically identifying or
verifying a person from a digital image or a video frame from a video source. It is one of the way of doing comparison from selected facial features from the face image and face database. Mostly it is used in security systems and can be compared to other biometrics such as fingerprint or eye iris recognition systems. Face recognition technology is the least intrusive and fastest biometric technology. It works with the most obvious individual identifier the human face. It measures the overall facial structure, including distances between eyes, nose, mouth, and jaw edges. These measurements are retained in a database and used as a comparison. Face recognition can be largely classified into two different classes of approaches, the local feature-based method and the global feature-based method. Human faces can be characterized both on the basis of local as well as of global features global features are easier to capture they are generally less discriminative than localized features local features on the face can be highly discriminative, but may suffer for local changes in the facial appearance or partial face occlusion. In this survey paper, an overview of some of the Multi-view face recognition methods in each of these categories is provided and some of the benefits and drawbacks of the methods mentioned therein are examined.
KEYWORDS: Face Recognition, Human Recognition, Biometrics, Multi-view face recognition. I. INTRODUCTION
Image processing is widely used in many applications, including medical imaging, industrial manufacturing, and security systems, face recognition, human recognition, figure print recognition in cyber crime, military application, and medical diagnosis. Now days we need to maintain global security Information, in every organization or individual wants to improve their existing security system. Most of the people need better security system which gives complete security solution. From time to time we hear about the crimes of credit card fraud, computer break-in by hackers, or security breaches in company, in shops, in government buildings. In most of these crimes the criminals were taking advantage of that hacking the information from commercial or academic access control system. The systems do not grant access by who we are, but by what we have, such as ID cards, keys, passwords, PIN numbers. These means they are really defining us or they just want to authenticate us. It goes without Permission of owners, duplicates, or acquires these identity means, he or she will be able to access our data or our personal property any time they want. Recently, technology became available to allow verification of true individual identity. This technology is based in a field called "biometrics". Biometrics is a technique for identifying people by using a unique physiological characteristic, such as a fingerprint, eye, face, etc. or behavioral characteristics, e.g., voice and signature etc. Biometrics is the use of computers to recognize people, considering all of the across-individual similarities and within-individual variations. Among the various biometric ID methods, the physiological methods such as fingerprint, face, DNA are more stable than methods in behavioral category like keystroke, voice print etc.. Face recognition is one of the biometric methods that to have the merits of both high accuracy and low intrusiveness. It has the accuracy of a physiological approach without being intrusive. For this reason, the face recognition has drawn the attention of researchers in fields from security, Psychology, and image processing, to computer vision. Many algorithms have been proposed for face recognition, Face recognition has also proven useful in other multimedia information processing areas. Facial recognition analyzes the characteristics of a person's face images input through a digital video camera or online face capturing. In this paper, the section II gives Structure and Procedure of Face Recognition System, section III gives Literature Survey of Face Recognition Techniques and section IV gives the conclusion and future scope for the paper.
www.ijesi.org
6 | Page
Fig 1 Architecture of face recognition system [1] As shown in fig 1 the first task of the face recognition system is capturing image by video or by camera and this image is given to the further step of face recognition system that is: Face Detection: The main function of this step is to determine whether the human faces appear in given image or not and where the face is located at. The outputs of this step are patches containing each face in the input image, face alignment are performed to justify the scales and orientations of these patches. These step is working as the preprocessing for face recognition, face detection could be used for region of interest detection, retargeting, video and image classification, etc. Feature Extraction: In this step feature extractions are performed to do information packing, dimension reduction, salience extraction, and noise cleaning. After this step, a face patch is usually transformed into a vector with fixed dimension or a set of fiducial points and their corresponding locations. Face Recognition: Feature extraction step analyzes the representation of each face; this last step is used to recognize the identities of these faces. In order to achieve automatic recognition, a face database is required to build. For each person, several images are taken and their features are extracted and stored in the database. Then when an input face image comes in, then perform face detection and feature extraction, and compare its feature to each face class stored in the database. There have been many researches and algorithms proposed to deal with this classification problem. There are two general applications of face recognition, one is called identification and another one is called verification. Face identification means given a face image, can be used to determine a person's identity even without his knowledge or consent. While in face verification, given a face image and a guess of the identification, the system must to tell about the true or false about the guess. Face recognition can be largely classified into two different classes of approaches, the local feature-based method and the global feature-based method. The Human faces can be characterized both on the basis of local as well as of global features global features are easier to capture they are generally less discriminative than localized features local features on the face can be highly discriminative, but may suffer for local changes in the facial appearance or partial face occlusion. Now a days face recognition system is recognize the face using multiple-views of faces, these Multi-view face recognition techniques has proposed by some authors for detecting each view of face such as left, right, front, top, and bottom using some methods like, PCA[2] , Elastic Bunch Graph Matching [3], Neural Network[4], Support Vector Machine[5], Spherical Harmonic Representation[6], Class-based, imagebased, rerendering and recognition method[7], Self-quotient image (SQI) method[8], Hybrid PCA-correlation filter Method[9], Min-Max and Z-score methods by a simple sum of score fusion[10], Class-Dependence feature analysis(CFA)[11], Kernel Discriminative Common Vector method, Gabor wavelets and Local Binary Patterns (LBP)[12], score fusion techniques[13], Color Image Discriminant(CID)model[14], Color information with the following three-fold contributions: RC r Q color space, RGB, YC b C YIQ color spaces[15], Fishers linear discriminant (FLD)[16], Haar-like Features for Visual Tracking[17], a novel illumination-insensitive preprocessing method[18], PCA, LDA, DWT+LDA algorithms[19] etc.The method proposed by Matthew Turk and Alex Pentland[2] suggest that, the PCA method works better on noise reduction and image compression but there are some issues in this method that is it has difficulty in modifying and evaluation of face parts and also there is a problem of recognizing the faces under different illumination condition.
www.ijesi.org
7 | Page
www.ijesi.org
8 | Page
www.ijesi.org
9 | Page
G. Kernel Discriminative Common Vector method Xiaoyang Tan and Bill Triggs[12] has proposed, The Kernel Discriminative Common Vector method, Gabor wavelets and Local Binary Patterns (LBP),In this method combing the two local appearance descriptor Gabor wavelets and Local Binary Patterns (LBP). LBP is basically a fine scale descriptor that captures small texture details. Local spatial invariance is achieved by performing local histogram on the resulted texture codes. LBP is a good for coding fine details of facial appearance and texture. And Gabor features encode facial shape and appearance information over a range of coarser scales although they have also been used as a preprocessing stage for LBP feature extraction. First the face image is captured, and its Gabor and LBP representations are extracted, projected into their PCA subspaces, it normalized separately and integrated into a combined feature vector, which is then projected into the optimal discriminant space, after that the optimal projection matrix is calculated and each training vector entry is mapped into training samples. The projected test feature vector test is then classified using the nearest neighbor rule and the cosine distance where template is a face template in the gallery set. Other similarity metrics such as Mahalanobis distances could be used, but found that the cosine distance and performed best among the metrics it tested on this database. Both methods give more information and computational efficiency is good. But neither scalable to large numbers of individuals nor easy to extend to additional feature sets. H. Likelihood Ratio Based Biometric Score Fusion Likelihood Ratio Based Biometric Score Fusion is proposed by Karthik Nandakumar[13], The combination of match scores that is based on the Likelihood ratio test. Fusion can be performed at four different levels of information, namely, sensor, feature, match score, and decision levels. Score level fusion is generally preferred because it offers the best trade-off in terms of the information content and the free in fusion. Combining match scores is a difficult task because the scores of different matchers can be either distance or similarity measure; it follows different probability distributions, or provides different accuracies and may be correlated. Score fusion techniques can be divided into the three categories: Transformation-based score fusion, classifier-based score fusion, Density based score fusion. The match scores are first normalized to a common domain and then combined. This approach is based on the likelihood ratio test and it requires explicit estimation of genuine and impostor match score densities. the finite Gaussian mixture model(GMM) is quite effective in modeling the genuine and impostor score densities, fusion based on the resulting density estimates achieves consistently high performance.
www.ijesi.org
10 | Page
www.ijesi.org
11 | Page
IV.
CONCLUSION
Face recognition is a challenging and important recognition technique, among all the biometric techniques. The computational time, accuracy and the performance of the face recognition system under uncontrolled illumination with different posses is important issue in multiple face model recognition system. The various methods and issues in multiple face model recognition are discussed in this paper. There is need of efficient method for face recognition system which will reduce computational time and increase efficiency.
ACKNOWLEDGEMENT
This paper is part of the project initiatives for identification of problem definition. Author thanks to Dr. M. V. Deshpande, Associate Dean, Mukesh Patel School of Technology Management & Engineering, NMIMS, Shirpur, for his valuable guidance in performing literature survey
REFERENCES
[1]. [2]. [3]. [4]. [5]. Wei-Lun Chao, Face Recognition, GICE, National Taiwan University, March 2007. Matthew Turk and Alex Pentland. Eifenface for Recognition." Journal of Cognitive Neuroscience, Vol. 3, No 1, 1991. Laurenz Wiskott, Jean-Marc Fellous, Norbert Kriiger, and Christoph von der Malsburg. Face Recognition by Elastic Bunch Graph Matching, IEEE Transactions On Pattern Analysis And Machine Intelligence, Vol. 19, No. 7. 1997. Henry A. Rowley. Neural Network-Based Face Detection IEEE Transaction on Pattern Analysis and Machine Intelligence, vol 20, No.1, 1998. Yongmin Li, Shaogang Gong, Jamie Sherrah and Heather Liddel. Multiview Face Recognition Using Support Vector Machine and Eigen space Modeling, Fourth International Confieme on knowledge-Based Infelligent Engineering Systems 6 Allied Technologies, 30Aug-1, Brighbon, UK, 2000. Ravi Ramamoorthi and Pat Hanrahan, On the relationship between radiance and irradiance: Determining the illumination from images of a convex Lambertian object, J. Opt. Soc. Amer., vol. 18, no. 10, pp. 24482459,2001. Amnon Shashua and Tammy Riklin-Raviv, The quotient image: Class-based re-rendering and recognition with varying illuminations, IEEE Trans. Pattern Anal. Mach. Intell., vol. 23, no. 2, pp. 129139,2002. H. Wang, S. Li, and Y. Wang, Generalized quotient image, IEEE. Comput. Vis. Pattern Recognit., vol. 2, pp. 498505,2004. M. Savvides, B. Kumar, and P. Khosla, Corefaces Robust shift invariant PCA based correlation filter for illumination toleran t face recognition, IEEE, Comput. Vis. Pattern Recognit., vol. 2, pp. 834841, 2004. Anil Jain, Karthik Nandakumar , Arun Ross, Score normalization in multimodal biometric systems, Science Direct,Pattern Recognition 38,2270 2285,2005. C. Xie, M. Savvides, and B. V. Kumar, Kernel Correlation Filter Based Redundant Class-Dependence Feature Analysis (KCFA) on FRGC2.0 Data, Springer-Verlag Berlin Heidelberg, AMFG 2005, LNCS 3723, pp. 3243, 2005. Xiaoyang Tan and Bill Triggs, Fusing gabor and LBP feature set for kernel-based face recognition, IEEE Int.Workshop Anal. Model. Face Gestures, pp. 235249, 2007. Karthik Nandakumar, Likelihood Ratio Based Biometric Score Fusion, To appear in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007. Jian Yang , Chengjun Liu, Color image discriminant models and algorithms for face recognition, IEEE Trans. Neural Netw., vol. 19, no. 12, pp.20882098,2008. Zhiming Liu, Chengjun Liu, Robust face recognition using color information, Advance Biometric, vol. 5558/2009, pp. 122 131,2009. Yu Su, Shiguang Shan, X. Chen, and W. Gao, Hierarchical ensemble of global and local classifiers for face recognition, IEEE Trans. Image Process., vol. 18, no. 8, pp. 18851896,2009. Seunghoon Park, Bohyung Han, Adaptive Feature Extraction with Haar -like Features for Visual Tracking, Pohang University of Science and Technology, Department of Computer Science and Engineering,2012. Wonjun Hwang, Haitao Wang, Hyunwoo Kim, Face Recognition System Using Multiple Face Model of Hybrid Fourier Feature Under Uncontrolled Illumination Variation, IEEE Transactions On Image Processing, Vol. 20, No. 4,2011 M. M. Mohie El-Din1, M.Y. El Nahas2 and H. A. El Shenbary, Hybrid Framework for Robust Multimodal Face Recognition , IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 2, No 2,2013..
[6]. [7]. [8]. [9]. [10]. [11]. [12]. [13]. [14]. [15]. [16]. [17]. [18]. [19].
www.ijesi.org
12 | Page