International Journal of Computer Trends and Technology- March to April Issue-2011

SOM and PCA Approach for Face Recognition - A Survey

Jayshree Ghorpade1, Siddhant Agarwal2
1

Assistant Professor, Department of Computer Engineering, MITCOE, Pune University, India 2 Student, Department of Computer Engineering, MITCOE, Pune University, India

Abstract wouldnt you love to replace password based access control to avoid having to reset forgotten password and worry about the integrity of your system? Wouldnt you like to rest secure in comfort that your healthcare system does not merely on your social security number as proof of your identity for granting access to your medical records? Because each of these questions is becoming more and more important, access to a reliable personal identification is becoming increasingly essential .Conventional method of identification based on possession of ID cards or exclusive knowledge like a social security number or a password are not all together reliable. ID cards can be lost, forgotten or misplaced. But a face is undeniably connected to its owner. It cannot be borrowed stolen or easily forged. Face recognition has always been a fascinating research area. Face recognition is the preferred mode of identification by humans: it is natural, robust and non-intrusive. Face recognition being the most effective and natural technique to identify a person since it is the same as the way human does and there is no need to use special equipments. A lot of algorithms have been proposed for solving face recognition problem. Principal component analysis (PCA) is a classical and successful method for face recognition. Self organizing map (SOM) has also been used for face space representation. This paper makes an attempt to enhance these two techniques and describe the use of PCA and SOM in facial recognition. It also explains the performance result when the two algorithms are combined together. The advantage in combining the two techniques is that the reduction in data is higher but at the cost of recognition rate. Keyword Face Recognition, SOM (Self Organizing Map), PCA (Principal Component Analysis), LDA (Linear Discriminant Analysis), ORL (Olivetti Research Laboratory).

PINs and passwords: birthdays, phone numbers and social security numbers. Face recognition technology may solve this problem since a face is undeniably connected to its owner except in the case of identical twins. This unique feature of an individual is non-transferable. The system can then compare scans to records stored in a central or local database or even on a smart card. Biometrics: Biometrics refers study of methods for uniquely recognizing human based upon one or more intrinsic physical or behavioural characteristics. Biometrics is used to identify the input sample when compared to a template used in cases to identify specific people by certain characteristics.

I. I NTRODUCTION The information age is quickly revolutionizing the way transactions are completed. Everyday actions are increasingly being handled electronically, instead of with pencil and paper or face to face. This growth in electronic transactions has resulted in a greater demand for fast and accurate user identification and authentication. Access codes for buildings, banks accounts and computer systems often use PIN's for identification and security clearances. Using the proper PIN gains access, but the user of the PIN is not verified. When credit and ATM cards are lost or stolen, an unauthorized user can often come up with the correct personal codes. Despite warning, many people continue to choose easily guessed

Fig. 1 Types of Biometrics.

Face recognition is one of the few biometric methods, which is very complicated system since the human faces change depending on their age, expressions etc. Face recognition is the preferred mode of identification by humans: it is natural, robust and non-intrusive [3, 4]. Security systems, human computer interaction, entertainment, criminal identification, videos surveillance etc. are among the most common

ISSN: 2231-2803

-1-

IJCTT

International Journal of Computer Trends and Technology- March to April Issue-2011

application of it [1, 2].Number of supervised and unsupervised learning techniques or methods have been reported for the face recognition. These techniques have been broadly divided into two categories namely Structured-based and Statisticsbased. Structured-based Face Recognition: In the structured-based approaches (Brunelli & Poggio, 1993; Wiskoot et al, 1997) recognition is based on the relationship between human facial features such as eye, mouth, nose, profile silhouettes and face boundary [8]. Statistics-based Face Recognition: Statistics-based approaches (Turk & Pentlands 1991; Zhao, 2000) attempt to capture and define the face as a whole. Under this approach, the face is matched through finding its underlying statistical regularities [8]. verification of the user and determine whether the match declared is right or wrong. D. Verification:

The verification module also consists of a pre-processing system. Verification means the system checks as to who the person says he or she is and gives a yes or no decision. In this module the newly obtained sample is pre-processed and compared with the sample stored in the database. The decision is taken depending on the match obtained from the database. Correspondingly the sample is accepted or rejected. Instead of verification module we can make use of identification module. For each comparison made a match score is given. The decision to accept or reject the sample depends on this match score falling above or below a predetermined threshold.

II. COMPONENTS OF FACE RECOGNITION SYSTEMS The Face Recognition system includes: An automated mechanism that scans and captures a digital or an analog image /of living personal characteristics (Enrollment module). Another entity which handles compression, processing, storage and compression of the captured data with stored data (Database). The third interfaces with the application system (Identification module).

The various Modules of Face Recognition System as shown in the Figure. 2 are: A. User Interface:

User interface captures the analog or digital image of the person's face. In this we take number of photos of a single person either in analog or in a digital form. B. Enrollment Module:

In the enrollment module the obtained sample is pre-processed and analyzed. This analyzed data is stored in the database for the purpose of future comparison. C. Database:

Fig. 2 Components of Face Recognition system.

The database compresses the obtained sample and stores it properly. It should have retrieval property also so that it compares all the stored sample with the newly obtained sample and retrieves the matched sample for the purpose of

This paper makes an attempt to compare the two different techniques that are SOM (an unsupervised learning algorithm) and the popular, successful classical method PCA to see the performance of face recognition system when the two techniques are combined together. After a brief discussion of

ISSN: 2231-2803

-2-

IJCTT

International Journal of Computer Trends and Technology- March to April Issue-2011

SOM and PCA in section III, section IV describes the combined approach of SOM and PCA for face recognition and experiments analysis, V contains the application and finally VI contain the conclusion. most common ones are either one-dimensional or twodimensional maps. The number of input connections depends on the number of attributes to be used in the classification. The neuron with weights closest to the input data vector is declared the winner during the training. Then the weights of all of the neurons in the neighbourhood of the winning neuron are adjusted by an amount inversely proportional to the distance. It clusters and classifies the data set based on the set of attributes used. The algorithm is summarized as follows [4, 5]: Step 1. Initialization: Choose random values for the initial weight vectors wj(0) , the weight vectors being different for j = 1, 2,...,l where 1 is the total number of neurons. Step 2. Sampling: Draw a sample x from the input space with a certain probability. Step 3. Similarity Matching: Find the best matching (winning) neuron i (x) at time steps n by using the minimum Distance Euclidean criterion

III. SOM AND PCA A. lf Organizing Map In this section, we will give a brief introduction on SelfOrganizing Map (SOM) [3, 4] (Kohonen, 1985). SOM has been proposed by Kohonen in the early eighties (Kohonen, 1985) [5]. Since that time, it has been used most widely for data analysis in some areas such as economics physics, chemistry or medical applications. The SOM provides an orderly mapping of an input high-dimensional space in much lower dimensional spaces, usually one or two dimensions. As it compresses information while preserving the most important topological and metric relationships of the primary data items, it can be thought to produce some kind of abstractions of information. So it can be utilized in a number of ways in complex tasks such as pattern classification, process analysis, machine perception, control, and communication [3, 4].

i(x) = arg min||x(n) wj||,

j

j = 1,2,...,l

Step 4. Updating: Adjust the synaptic weight vector of all neurons by using the update formula

Wj (n + 1) = wj (n) + (n)hj, i(x) (n)(x(n) - wj (n)) where, (n) is the learning rate parameter, and hj, i(x)(n) is the neighbourhood function centred around the winning neuron i(x).
Both (n) and hj, i(x) (n) are varied dynamically during learning for best results. Step 5. Continue with step 2 until no noticeable changes in the feature map are observed.

Fig. 3 An example of a Two-Dimensional SOM.

T. Kohonen introduced the Self-Organizing Map (SOM) [3, 4, 6]. It is an unsupervised learning process, which learns the distribution of a set of patterns without any class information. It has the property of topology preservation. There is a competition among the neurons to be activated or fired .The result is that only one neuron that wins the competition is fired and is called winner-takes all neuron. SOMs may be onedimensional, two-dimensional or multidimensional, but the

B. Principal Component Analysis The Principal Component Analysis (PCA) [10] is one of the most successful techniques that have been used in image

ISSN: 2231-2803

-3-

IJCTT

International Journal of Computer Trends and Technology- March to April Issue-2011

recognition and compression. PCA is a statistical method under the broad title of factor analysis. The purpose of PCA is to reduce the large dimensionality of the data space (observed variables) to the smaller intrinsic dimensionality of feature space (independent variables), which are needed to describe the data economically. This is the case when there is a strong correlation between observed variables. The jobs which PCA can do are prediction, redundancy removal, feature extraction, data compression, etc. Because PCA is a classical technique which can do something in the linear domain, applications having linear models are suitable, such as signal processing, image processing, system and control theory, communications, etc. Kirby and Sirovich [7] developed a technique based upon which the idea of eigenfaces was used. The main idea of using PCA for face recognition is to express the large 1-D vector of pixels constructed from 2-D facial image into the compact principal components of the feature space. This can be called eigenspace projection. Eigenspace is calculated by identifying the eigenvectors of the covariance matrix derived from a set of facial images (vectors). These eigenfaces represented the faces efficiently using Principal Component Analysis. An image was treated as a vector in a very high dimensional space. Only the best eigenfaces (eigenvectors of the covariance matrix of a set of images) those that had the largest eigenvalues were used to approximate the face [5]. Consider a set of N sample images {1, 2 ... N} taking values in an n-dimensional image space. Let us also consider a linear transformation mapping the original n-dimensional image space to m-dimensional feature space where m < n. The new feature vectors Yk are defined by the following linear transformation [5]:

Normal (original) images

Eginefaces images

Fig. 4 Example of Eginefaces [1, 10].

IV. COMBINING SOM AND PCA As discussed earlier that SOM [3, 4] is an unsupervised learning process and this technique has the property of topology preservation. It defines a mapping from an input space onto a set of nodes in a space that has dimension much lower than that of the input space. The set of nodes is topologically ordered. An image, divided into sub blocks, is mapped to a lower dimensional space with topologically ordered set of nodes thereby providing dimensionality reduction. Further feature extraction is provided with the method known as Karhunen - Loeve (KL) [7] transform via Principal Component Analysis (PCA). A set of orthogonal axes of projections called as principal components or the eigenvectors are generated by PCA. PCA is applied to the weight matrix generated by mapping the image onto lower dimensional space using SOM. For the further reduction of dimensionality, the eigenvectors with smaller eigenvalues are ignored and the eigenvectors corresponding to the largest eigenvalues are retained for image reconstruction. In PCA method [1, 9, 10] the pixels of each row of the training images, taken one at a time, are concatenated vertically to form a single vector containing all the pixel values of an image thereby producing a matrix, each column of which represents an image and there are as many numbers of columns as the number of training images [7]. Nearly half of the total number of eigenvectors, corresponding to the largest eigenvalues, obtained from the covariance matrix of the training images, is retained. The test images are reconstructed after finding out its KL coefficients using the retained eigenvectors and matching is done using Euclidean norm as the similarity measure. Whereas in SOM method [5], the training images are mapped to lower dimension using SOM and the weight matrix of each training image is stored. At the time of recognition, the training images are reconstructed using the weight matrices and matching is done with the test image using Euclidean norm as the similarity measure. The same procedure is adopted for the third technique when the two SOM & PCA are

Yk = T k
where,

is a matrix with orthonormal columns

. The covariance matrix is defined as

N C = (k - )(k - ) T K=1
where,

is the mean image of all the samples. Only mnumber of n-dimensional eigenvectors [V1, V2, ... Vm] of C is chosen that correspond to the m largest eigenvalues [5].
Example of eginefaces:

ISSN: 2231-2803

-4-

IJCTT

International Journal of Computer Trends and Technology- March to April Issue-2011

combined. The eigenvectors of the weight matrix are found which are obtained through SOM training and sixty four percent of the total number of eigenvectors corresponding to largest eigenvalues and hence the KL coefficients is retained and at the time of recognition, the images are reconstructed and matching is done with the test images. In the second case, the weight matrix was obtained by using the eigenvectors and the KL coefficients and the weight vectors corresponding to first twenty neurons were considered for reconstruction of the image thereby reducing the memory space required to store the image.
TABLE 1 RECOGNITION RATE OF THE FACE RECOGNITION SYSTEM WITH VARYING NUMBER OF CLASSES [5]

varied from 10 to 20 to 40 and then the recognition rate was found. Table 1 clearly show the recognition rate (%) as the number of classes is varied. Table 1observations indicates that as the number of classes is increased then the recognition rate is decreases and hence automatically performance of the system is decreased. The results shows that the recognition rate of SOM is slightly better that that of PCA technique. While for the combined SOM & PCA (1) method, sixty four percent of eigenvectors corresponding to the largest eigenvalues were retained for reconstruction of the image. B. In the second experiment the performance of the face recognition system is checked when the sub block size is changed. Table 2 shows the recognition rates as the size of the sub block is changed. It is clear from Table 2 that there is a little change in the recognition rate for SOM and SOM & PCA combined (1) techniques whereas for SOM & PCA combined (2), it is less as compared with the first two techniques and it reduces more rapidly as the size of sub block is increased.

Recognition Rate (%) Number of Classes Method SOM (5 5) SOM (10 10) PCA SOM+PCA (1) 10 94.06 94.06 93.39 77.75 20 90.72 91.86 90.25 72.08 40 89.92 90.82 89.51 62.64

V. APPLICATIONS The natural use of face recognition technology is the replacement of PIN, physical tokens or both needed in automatic authorization or identification schemes. Additional uses are automation of human identification or role authentication in such cases where assistance of another human needed in verifying the ID cards and its beholder. There are numerous applications for face recognition technology: A. Government Use: Law Enforcement: Minimizing victim trauma by narrowing mugs hot searches, verifying identify for court records, and comparing school surveillance camera images to know child molesters. Security/Counterterrorism. Access control, comparing surveillance images to Know terrorist. Immigration: Rapid progression through Customs.

TABLE 2 RECOGNITION RATE OF THE SYSTEM WITH CHANGING SUB BLOCK SIZE [5]

Recognition Rate (%) Size of Sub Block Method SOM (5 5) SOM +PCA (1) SOM +PCA (2) ( 4 4) 94.06 77.75 68.29 ( 8 8) 94.06 77.17 62.17 (1616) 95.95 72.83 54.89

Experimentation Analysis For the experimentation [5, 9] on combined SOM and PCA the ORL [4, 11] (Olivetti Research Laboratory) face database has been used. This ORL database basically contains 10 different images of 40 distinct subjects. Now Two dimensional SOM was trained and finally one weight matrix of 2516 was generated and after that PCA was applied to the transpose of the weight matrix and eigenvalues were retained for reconstruction of image. The experiments are as follows: B.

Commercial Use: Day Care: Verify identity of individuals picking up the children. Residential Security: Alert homeowners of approaching personnel. Voter verification: Where eligible politicians are required to verify their identity during a voting process this is intended to stop proxy voting where the vote may not go as expected. Banking using ATM: The software is able to quickly verify a customers face.

A. In the first experiment to see the performance of face recognition system the number of classes was varied from one value to another value. The ORL databases were

ISSN: 2231-2803

-5-

IJCTT

International Journal of Computer Trends and Technology- March to April Issue-2011

Physical access control of buildings areas, doors, cars or net access.
[8] Face Recognition Using Self-Organizing Maps, Qiu Chen, Koji Kotani, Feifei Lee and Tadahiro Ohmi, Tohoku University, Japan. Face Recognition using PCA versus ICA versus LDA cascaded with the Neural Classifier of concurrent Self-Organizing Maps, VictorEmil Neagoe, Senior Member IEEE, Ioan Anton Stanculescu, University of Bucharest. Face Recognition Using Improved Fast PCA Algorithm, Neerja, Ekta Walia, University Patiala (Punjab), INDIA. AT&T Laboratories Cambridge, The database of faces at ttp: //www.cl. cam.ac.uk/ research/ dtg/attarchive/facesataglance.html.

[9]

VI. CONCLUSION Thus Face recognition is a fascinating research area. In this paper information about two very innovative techniques i.e. SOM and PCA is given in detail. And also explained how these techniques can be used for face recognition. Both techniques are very useful in face recognition. In this paper we have explained SOM and PCA approach for face recognition. With the help of this paper a new idea was proposed. New idea is nothing but combining both SOM and PCA together. PCA was combined with SOM for dimensionality reduction and feature extraction. This is true for all the three methods i.e. SOM, PCA and SOM & PCA combined. If we compare combined SOM & PCA method with other two methods then we will come to know that the decrease is more in case of combined method. The experimental results also shows that if the size of the sub block is varied or changed, then there is only small significant change in the performance of the recognition system for SOM and for the technique where the two SOM & PCA are combined. For the combined technique where lesser number of neurons of the weight matrix is used for image reconstruction, the recognition rate is less as compared to the earlier two techniques and decreases more rapidly. The reduced number of neurons and hence reduced number of features result in decrease in the recognition rate, but at the same time, it results in saving in the memory space required for storing the images.
[10]

[11]

References
[1] W.Zhao, R. Chellapa, A. Rosenfeld, P.J Phillips,, "Face Recognition: A Literature Survey ," ACM Computing Surveys, Vol.35, No.4, pp. 399-458, 2003. R. Chellapa, C.L.Wilson, S.Sirobey, "Human and Machine Recognition of Faces: A Survey," Proc. IEEE, Vol.83, No.5, pp. 705740, 1995. Santaji Ghorpade, Jayshree Ghorpade, Shamla Mantri, Pattern Recognition using Neural Networks, IJCSIT, Vol 2, No 6, Dec- 2010, pg.92-98, India. Santaji Ghorpade, Jayshree Ghorpade, Shamla Mantri, Dhanaji Ghorpade Neural Network For Face Recognition using SOM, IJSCT, Vol. 1, Iss ue 2, Dec-2010, pg.65-67, India. Face Recognition using Self-Organizing Map and Principal Component Analysis, Dinesh Kumar, C.S. Rai and Shakti Kumar, 07803-9422-4/05, 2005 IEEE. Kohonen, T.(1985). Self - Organizing Maps, Springer Verlag, Berlin. M.Kirby and L.Sirovich, "Application of the Karhunen-Loeve Procedure for the Characterization of Human Faces," IEEE Trans.

[2]

[3]

[4]

[5]

[6] [7]

ISSN: 2231-2803

-6-

IJCTT