A New Method for Face Recognition with Fewer Features under Illumination and Expression Variations

Size: px
Start display at page:

Download "A New Method for Face Recognition with Fewer Features under Illumination and Expression Variations"

Transcription

1 A New Method for Face Recognition with Fewer Features under Illumination and Expression Variations Chandan Tripathi Dept. of Computer Science Engg. Sharda University Greater Noida, India t chandan@yahoo.com Dr. K.P.Singh Dept. of Information Technology IIIT Allahabad Allahabad, India Abstract In this study, a new adaptive feature extraction method has been presented based on multi-dimensional discriminant analysis MLDA) over multi-dimensional principal components. Proposed work has been aimed to design a method that can predict required number of features for a particular dataset. This method use only effective features which have better discriminant power in different dimensions of an image. In order to ease the pre-processing we controlled the variance in each mode to make the feature selection adaptive in different datasets with facial variance present in the image. The Experiments with different datasets has been performed in order to check suitability for larger dataset, with lesser computational cost and higher efficiency. Moreover, when support vector machine operated as classifier, proposed algorithm shows its superiority of recognition over previous known methods like PCA,PCA-LDA,MPCA. Keywords-Principal Component AnalysisPCA);Linear Discriminant AnalysisLDA); Multilinear Principle Component AnalysisMPCA); K-Nearest Neighborhood ClassifierKNN); Support Vector Machine SVM); I. INTRODUCTION Face Recognition is one the field that extensively attracted the concentration of the researchers as in case of its methodology or its application in computer based vision and pattern recognition [1]. Several algorithms have been introduced and renovated to achieve higher recognition rate based on their own points of strength and restrictions. Still there are several challenges remains to improve the correctness of their systems under different illuminations, poses and image resolutions. Many of the researchers designed their algorithms that focuses over such issues but all of them required to adjust their features dimensions manually or based on some thresholds. In three decade research history of appearance based first face recognition algorithm was introduced by Turk and Pentlands [2] with use of Eigenfaces based principal component analysis PCA). Principal idea behind PCA was reduction of those dimensions which are non-repetitive in the image dataset i.e., with lower Eigen values. Comon Pierre [3] introduced a kernel Hilbert space based ICA which was further used to work with non-gaussian nature of the faces [4]. K. Kim, K. Jung and H. K. Kim introduced a kernel based PCA[5] to achieve optimal recognition speed. All these methods require reshaping of the series of k number images with dimension m n into matrix having higher dimensional size of mn k. It causes high computational and processing cost in terms of both memory and time. Yeng et al.[6] proposed a 2D-PCA approach which successfully reduces the computation complexity with an increase in recognition rate[7][8]. 2D-PCA approach was further modified by Nguyen et al. [9] with the help of random subspace technique in feature selection. This was the first time when some Eigenvectors with non-zero Eigenvalues were also considered with Eigenvectors with highest Eigenvalues. It enhanced the recognition rate when compared with traditional 2D-PCA. A Gabor Features based 2D) 2PCA was proposed by X. Pan and Q. Ruan [10]. W. Yu et al. [11] proposed an approach using LDA for feature reduction over both Upright vertical) and Level horizontal) directional information subspace of 2D-PCA. Three dimensional PCA was explored by Yu and Bennamoun [12] as nd-pca algorithm which used to test its efficiency on FRGC-3D face dataset as well as gray images. H. Lu et al. [13, 14] modified the MPCA framework for Tensor based object feature extraction with an application for gait recognition. D. Xu, S. Yan [15] proposed a solution of MPCA based algorithm but was only concerned with reconstruction not the recognition where the data is not centred. They were also not concerning with initialization issues, termination problems, convergence, and subspace dimensionality determination problems. After several experimentation it has been found that 2D-PCA and MPCA based methods designed to achieve computational efficiency but with an increase in feature dimension in order to retain recognition accuracy, when compared with traditional PCA. In process of classification these higher feature dimension of those methods were leading towards higher computational time. Keeping above views in this study, a new adaptive method has been proposed with optimal reduction in feature dimension without compromising with both, recognition accuracy as well as computational time. In order to justify the quality of proposed method we compared the results with different feature selection processes different PCA and LDA /12/$ IEEE

2 based methods and with their combinations) and classification techniques classifiers based on nearest neighbourhood and support vectors) examined in our experimentations. II. PRINCIPAL COMPONENT ANALYSIS PCA was the first algorithm used as feature selection based on Eigenfaces [2]. Eigenfaces is projection of input images into that subspace where principal components appear orthogonal to each other. PCA was designed under unsupervised learning model to perform its feature selection. Suppose the training images have P number of images F with size M N. Algorithm grounded on PCA is explained below: 1) Reshaping of F i into a column matrix with size of MN 1 in order to form a 2D matrix of size MN P. As for an i th image. F i M N to F i MN 1 1) 2) Obtain the mean image of reshaped image matrix as F MEAN = 1 P P FMN 1 i 2) i=1 3) Center the images by making difference between reshaped input image FMN i i and F MEAN as F i Diff = F i MN 1 F MEAN 3) 4) Construct the co-variance matrix C = 1 P P FDiff i )FDiff i ) T 4) i=1 5) Find out the Eigen Decomposition on covariance matrix C. Let X a matrix that contains the selected eigenvectors based on the largest n Eigenvalues, where the X can be imagined as X = [x 1,...,x n ] 6) Project the testing sample into the acquired subspace to find the features which will forwarded for classification as Y i = X T F i Diff ) T est 5) 7) Test the classifier with features obtained in Y i III. MULTILINEAR PRINCIPAL COMPONENT ANALYSIS This method works on tensor representation [13] of the faces in order to preserve the relationship among neighboring pixels as absent in PCA. A tensor object is a higher dimensional matrix of order N stand as T NS R T 1 T 2... T N ), where each L N represents each mode of that tensor. For M number of images where image is symbolized as F i, can be projected into a third dimensional tensor as T NS R T1 T2 T3). In TNS T 1 stands as row size height) of image as mode-1 of tensor TNS, T 2 is column size width) of image as mode-2 of tensor TNS and T 3 is number of images P used in training as mode-3 of tensor TNS. Multilinear PCA algorithm is explained below: 1) Obtain the mean Image value of all training images F M ean as F Mean = 1 T 3 FM N t where T 3 = P 6) T 3 t=1 2) Center the tensor object representation of training images {}}{ T NS = [F 1 M N F Mean, F 2 M N F Mean,..., F P M N F Mean ] 7) 3) Unfold the tensor for different n-modes into matrix for different modes. Unfolding of tensor TNS along the n-mode can be defined as T N Sn) R Tn T1... T N )) i.e.; for { T NS }}{ n ) indexval = f t1t 2t 3), where f is value present at t 1, t 2, t 3 ) th index value where index value selection can be done as n+1 n+1 ) 1 ) index val = [t n, t m 1) T q T q m=3 + 1 m=n 1 q=m t m 1) q=n 1 1 q=m 1 T q )] 8) 4) Evaluate the covariance matrix mentioned as C n and C n = T {}} NS { T ranspose {}}{ T NS of each n-mode and find the n) n) set of eigenvectors X n = [x 1,..., x kn) ]founded on the selected eigenvectors according to the largest k n) Eigenvalues. One can obtain optimal features set using any of the subsequent feature selection process: i) Using Y t = Fm n) t Input F Mean ) n different value for n for different modes), which will give k 1) T 2 + k 2) T 1 features. X T ranspose n) ii) Using Y t = X T ranspose 1) AB M N) t ) Input AB Mean )X 2) different value for n for different modes), which will givek 1) k 2) features. 5) Test the classifier with obtained feature set Y t IV. LINEAR DISCRIMINANT ANALYSIS LDA [16] [17] [18] aims to search the projection P a, a set of basis vectors which are normal vectors to discriminant hyper-plane of feature vector y m that can be obtained using any of the feature selection method. LDA intend to use Gaussian information present in image data using class information. For an m th training sample image having class C m LDA intended to maximize the ratio of the S between class between-class scattering matrix) and S within class withinclass scattering matrix), where S within class = M y m y cm )y m y cm ) T 9) m=1

3 where and S between class = where y c = 1 N c C Class=1 y = 1 M m,c m =C y m N Class y Class y)y Class y) T m y m 10) The purpose of LDA is find maximized set of basis vector P as P LDA = argmaxp P T S between class P ) P T S within class P ) 11) as above equation results P LDA = [p 1 p 2...p HZ ] Here p hz, where h Z = 1, 2,...H Z, is the set of generalized eigenvectors of S between class and Swithin class for H Z C 1 largest generalized Eigenvalues. The discriminant feature vector for m th training image can be obtained as yd m = P T LDA y m. V. PROPOSED ALGORITHM Based on above observations we are introducing our proposed algorithm based on n-mod Tensor unfolding features as input to Discriminant Analysis to find optimal features in each mod. The algorithm has been presented as below. 1) Obtain the mean Image value of all training images F M ean as F Mean = 1 T 3 FM N t where T 3 = P 12) T 3 t=1 2) Center the tensor object representation of training images {}}{ T NS = [F 1 M N F Mean, F 2 M N F Mean,..., F P M N F Mean ] 13) 3) Unfold the tensor for different n-modes into matrix for different modes. Unfolding of tensor TNS along the n-mode can be defined as T N Sn) R T n T 1... T N )) i.e.; for T { NS }}{ n ) indexval = f t1 t 2 t 3 ), where f is value present at t 1, t 2, t 3 ) th index value where index value selection can be done as n+1 n+1 ) 1 ) index val = [t n, t m 1) T q T q m=3 + 1 m=n 1 q=m t m 1) q=n 1 1 q=m 1 T q )] 4) Evaluate the covariance matrix mentioned as C n and C n = { T }} NS { T ranspose {}}{ T NS of each n-mode and find the set n) n) of eigenvectors X n = [x 1,..., x kn) ] according to their k n) Eigenvalues. 5) Search LDA subspace projection of each mod-n: i) A new projection P an) set of basis vectors for each mod-n which are normal vectors to discriminant hyper-plane of feature vector x kn) of m th training sample image in mod-n having class c m. ii) ii) Calculate within-class scattering matrix S within class using S within class = P class=1 x kn)class x kn)class ) x kn)class x kn)class ) T 14) here x kn) is within-class average image, where x kn) = 1 N C n,c n=c x kn) iii) Calculate between-class scattering matrix S between class using S between class = C class=1 N Class x kn)class x kn) ) x kn)class x kn) ) T 15) Here x kn) is between-class average image, where x n) = 1 P n x kn) iv) Calculate Eigenvaluesλ ldan) and Eigenvectors V ldan) corresponding each n-mod within-class scattering matrix S within class and between-class scattering matrix S between class S between class λ ldan) = λ ldan) S within class V ldan) 16) v) Find set of eigenvectors X ldan) = [x lda1,, x ldakn) ] from V ldan),as return by step iv), for each mod n according to their discrimination power over generalized k n) Eigenvalues obtained in step 5. 6) One can obtain optimal features set using Y t = X T ranspose lda 1) F t M N ) Input F Mean )X lda2) 17) Experimental outcomes of the proposed method have been compared with different previous known methods as Eigen face based PCA [2], fisher faces based combined approach of PCA-LDA [16], Eigen tensor based MPCA. In order to

4 TABLE I: Results along with their respective dimensions based on E-1 on AT&T Dataset based on SVM classifiers Sample Ratio training/testing) PCA PCA-LDA MPCA MLDA 1/ %40) 73.05%39) 75.28%69) 79.16% 39) 2/ %86) 86.87%39) 87.50%65) 89.38%39) 3/ %64) 91.43%39) 92.50%55) 92.34%39) 4/ %78) 94.17%39) 93.75%59) 95.42%39) 5/ %43) 97.00%39) 97.00%50) 98.50%39) 6/ %69) 96.25%39) 98.13%60) 99.32%39) 7/ %76) 98.33%39) 97.50%79) 98.97%39) 8/ %63) 97.50%39) 98.75%71) 100%39) 9/ %21) 97.50%39) 100%55) 100%39) TABLE II: Results along with their respective dimensions based on E-2 on AT&T Dataset based on SVM classifiers Sample Ratio training/testing) PCA PCA-LDA MPCA MLDA 1/ %40) 58.89%39) 61.94%69) 56.67% 39) 2/ %86) 71.88%39) 83.13%65) 77.50%39) 3/ %64) 78.21%39) 83.21%55) 88.57%39) 4/ %78) 79.58%39) 85.42%45) 91.67%39) 5/ %43) 79.50%39) 85.00%50) 93.50%39) 6/ %69) 80.63%39) 88.75%60) 96.88%39) 7/ %76) 83.33%39) 89.17%79) 95.00%39) 8/ %63) 81.25%39) 87.50%71) 96.25%39) 9/ %21) 85.00%39) 90.00%55) 95.00%39) Fig. 1: Example images of cropped AT&T dataset with size as used in E-1 Fig. 2: Example images of original AT&T data-set with as used in E-2 understand the roll of classifiers on the features based on the proposed method we tested it with the neighborhood classification approach and redial basis kernel function based support vector machine classifier. VI. EXPERIMENTAL ANALYSIS Proposed method has been tested over windows-7 platform of 32-bit based machine with Intel core-2 duo 2.5 GHz and 2GB RAM. The code has been simulated using Matlab Two data-sets AT&T and YALE has been involved to test the performance of proposed algorithm. The data-sets were partitioned into two random sets as one for training and other for testing. In order to ease of representation the data-set sample ratio has been represented as training/ testing. Pre processing methods have been used for face finding based on coordinate extraction with eye localization for alignment and normalization of the same. Proposed algorithm has been tested under two different experimental conditions of the original data-sets. The first experiment E-1) was tested over cropped and normalized facial area in both the data-sets. The motive behind E-1 was to test the strength of features selected through proposed algorithm. The second experiment E-2) tested over original data-set without any preprocessing. The idea behind E-2 was to test the selection strength of the proposed algorithm in presence of hair and white glass spectacles as occlusion followed by some shadow part in background. A. Experiment on AT&T data-set AT&T dataset is consists of 400 images of size 92x112 pixels of 40 individuals with 10 faces for each individual. Example images for E-1 with size 44x44 and for E-2 with 92x112 have been given in Figure-1 and Figure-2 respectively. The experiments have been performed by changing training and testing ratio from 1/9 to 9/1. The proposed algorithm has been comes up with percentage score 79.32% in E-1 and 56.67% in E-2 with ratio 1/9 when combined with SVM. Based on several trials it has been found that E-2 gives good result then the algorithm operated with 80% of variance in each mode MPCA feature dimension. This variance control was maintained in E-2 was just to give a guess of facial area to the algorithm. In E-1 there was no need of any variance

5 a) b) c) Fig. 3: Recognition rate of all four methods over ORL dataset: A) with two images taken for training B) with three images taken for training C) with four images taken for training D) with five images taken for training of each training sample d) control as we have taken the facial part only. Further, a change in sample ratio shows that recognition accuracy of proposed method gradually increases and found its superiority when it comes with its optimal feature size. Initially when fewer features have been used with small training size proposed method shows lower recognition accuracy. The reason behind this lies in way of feature selection of these methods. As PCA based methods seeks to find the features based on higher Eigen-values, whereas LDA based methods search those features which may not have higher Eigen-values but the higher discrimination power. The results based on Experiment- 1 and E-2 has been given in table-i and in table-ii. One comparison has been done to in order to understand the effect of feature dimension over recognition rate with different sample ratio in experiment-1 as shown in Figure-III. It shows that after reaching an optimal features dimension and a gradual increment upto the number of required features supposed by the system the proposed method shows higher recognition rate against rest of the methods. When the sample ratio of 5/5 has been operated then the proposed method shows its superiority at every feature dimension. A comparative study has been done over different combinations of classifiers and previous known algorithms along with the proposed algorithm. This comparison is to understand the effect of classifiers recognition rate as well as total processing time. Here total processing time contains sum of following time informations: i) Time in feature generation form raw facial images. ii) Time taken in training the SVM/ Time taken in finding K nearest neighbourhoods iii) Time taken in classification of testing images This test includes KNN and SVM classifiers in combination with PCA, PCA-LDA, MPCA, and proposed algorithm as shown in table-iii. Based on above comparisons we observed that time taken in proposed algorithm is quiet promising than the other methods. In both cases proposed algorithm recognize faster than other methods. B. Experiment on YALE data-set YALE database is consists of 165 images of 15 individuals with 11-type images per person in GIF format. Example images for E-1 with size and for E-2 with as original size, have been given in Figure-4 and Figure-5 respectively. The motive behind use of this database is to

6 TABLE III: Processing time comparisons of the methods based on KNN classifier and SVM classifier in E-1 on AT&T Dataset SVM KNN Sample Ratio Training/Testing) 1/9 2/8 3/7 4/6 5/5 6/4 Features MPCA Accuracy%) Total Processing Times) Features MLDA Accuracy%) Total Processing Times) Features MPCA Accuracy%) Total Processing Times) Features MLDA Accuracy%) Total Processing Times) Fig. 4: Example images of cropped YALE dataset with size as used in E-1 Fig. 5: Example images of original YALE dataset with size as used in E-2 TABLE IV: Results along with their respective dimensions based on E-1 on YALE Dataset based on SVM classifiers Sample Ratio training/testing) PCA PCA-LDA MPCA MLDA 2/ %30) 48.15%14) 49.63%25) 54.07%14) 3/ %25) 55.83%14) 76.30%20) 83.70%14) 4/ %28) 64.76%14) 79.17%15) 89.17%14) 5/ %28) 68.89%14) 83.81%23) 91.43%14) 6/ %20) 81.33%14) 87.78%25) 97.33%14) 7/ %25) 88.33%14) 89.33%20) 96.66%14) 8/ %21) 93.33%14) 93.33%18) 97.77%14) 9/ %29) 90.00%14) 90.00%21) 93.33%14) TABLE V: Results along with their respective dimensions based on E-2 on YALE Dataset based on SVM classifiers Sample Ratio training/testing) PCA PCA-LDA MPCA MLDA 2/ %30) 36.30%14) 46.67%25) 52.59%14) 3/ %25) 39.17%14) 74.17%20) 81.67%14) 4/ %28) 49.52%14) 76.19%15) 84.76%14) 5/ %28) 58.89%14) 80.00%23) 78.89%14) 6/ %20) 74.67%14) 77.33%25) 86.67%14) 7/ %25) 85.00%14) 76.67%20) 88.33%14) 8/ %21) 88.89%14) 82.22%18) 91.11%14) 9/ %29) 90.00%14) 83.33%21) 90.00%14)

7 TABLE VI: Processing time comparisons of the methods based on KNN classifier and SVM classifier in E-1 on YALE Dataset SVM KNN Sample Ratio Training/Testing) 2/9 3/8 4/7 5/6 6/5 Features MPCA Accuracy%) Total Processing Times) Features MLDA Accuracy%) Total Processing Times) Features MPCA Accuracy%) Total Processing Times) Features MLDA Accuracy%) Total Processing Times) a) b) c) Fig. 6: Recognition rate of all four methods over YALE dataset: A) with two images taken for training B) with three images taken for training C) with four images taken for training D) with five images taken for training of each training sample d)

8 understand the feature selection strength with change in illumination variation with different directional focus of light. This database is more challenging due to presence of occlusion as glasses and different facial appearance normal, sleepy, wink, surprised, happy and sad). The experiment has been done with different sample ratio varying from 2/9 to 9/2. Proposed algorithm scored 89.17% in E-1 and 84.76% in E-2 when operated with sample ratio of 4/7. The highest recognition rate of 97.77% in E-1and 91.11% in E-2 has been shown by proposed method. On further increment in sample ratio we found slight decrement in recognition accuracy because of some variations in hair style of last images of each person class, still higher than other methods. The results analysis has been given in table-iv. After several experimentations it has been found that in E-2 keeping 60% of variance in each mode with MPCA feature dimension gives best result as shown in table-v. As in E-2 when shadow part was higher in image part the system failed to understand true face variations and causes a decrement in recognition accuracy as compared with results based on E-1. One test has been done in order to understand the time complexity issue of proposed method over YALE dataset. When the total processing time has been compared then the proposed method proved its superiority than previous methods. Another test has been prepared with KNN and SVM classifiers in order to understand effect of classifiers on processing time with higher recognition rate. In both cases proposed MLDA works faster than other methods as shown in table-vi. We compared the recognition accuracy based on feature dimension variations in different methods with proposed methods with sample ratio starting from 2/9 to 5/6. Based on the results we found better accuracy of proposed method than previously known methods. All the comparisons have been performed with E-1. VII. CONCLUSION We have proposed a new algorithm for face recognition using fewer features. Proposed algorithm has been tested on ORL and YALE datasets with and without pre-processing techniques. Based on observations we concluded that proposed MLDA algorithm have more discrimination power under different illumination and expression conditions than previous existing methods such as PCA, PCA-LDA and MPCA. When data size has been increases the recognition accuracy also increases as information in the tenor space is also increases. In each case with increase in training size, the recognition accuracy of the proposed system shows its superiority when compared with other existing methods. It ensures that proposed method is suitable with larger datasets. Although it shows better results than existing methods, under pose and illumination variations but a slight degrade in accuracy when there is a larger variation in illumination is present. Another experiment that contains background with some shadow effect has been performed. We found encouraging results when compared with the other methods. SVM and KNN classifiers have been used to understand the effect of classifiers on the recognition rate. In both datasets proposed method gives better accuracy than other combinations when operated with SVM classifier. Finally it can be conclude that proposed MLDA method work well under illumination variation, pose variations and under shadow effect with promising accuracy and efficiency even with larger dataset size. REFERENCES [1] W. Jhao, R. Chellappa, P.J. Phillips, A. Rosenfeld, Face Recognition: A Literature Survey,ACM Computing Surveys CSUR, vol. 35,no. 4, pp , [2] M. Turk, A. Pentland, Eigen Faces for Recognition, Journal of cognitive neuroscience, vol. 3, no. 1, pp , [3] Comon Pierre Independent component analysis, A new concept?,signal Processing, vol. 36, no. 3, pp , [4] M.S. Bartlett, JavireR, Movellan, T.J. Sejnowski, Face Recognition by Independent Component Analysis,IEEE Transactions on Neural Networks, vol. 13, no. 6, pp , November [5] K. Kim, K. Jung, H. K. Kim, Face recognition using Kernel Principal Component Analysis,IEEE Signal Processing Letters, vol. 9, no. 2, pp , August [6] J. Yung, D. Jhang, et al., Two-dimensional PCA: A new approach to appearance based face representation and recognition,ieee Transaction on pattern analysis and Machine Intelligence, vol. 26, no. 1, pp , Jan [7] J. Ye., Generalized Low Rank Approximation of matrices, Machine Learning,vol. 61, no. 1-3, pp , [8] Liu, J. and Chen, S. and Zhou, Z.H. and Tan, X.,Generalized lowrank approximations of matrices revisited,ieee Transactions on Neural Networks, vol. 21, no.- 4, pp ,2010. [9] N. Nguyan, W. Liu, S. Venkates, Random subspace two-dimensional PCA for Face Recognition,Lecture Notes in Computer Science, Advances in Multimedia Information Processing-PCM 2007, vol. 4810, p.p , December [10] Pan, Q Ruan, Palmprint Recognition using Gabor Features-Based 2D)2PCA,Neurocomputing, vol. 71, no , pp , August [11] W. Yu., Z. Wang, W. Chen, A new Framework to Combine Vertical and Horizontal Information for Face Recognition, Neurocomputing, vol. 72, no. 4-6,pp , January [12] H. Yu, M. Bennamoun, 1D-PCA 2D-PCA to nd-pca, 18th International Conference on Pattern Recognition, ICPR 2006, vol. 4, pp ,August [13] H. Lu, K. N. Plataniotis, A. N. Venetsanopoulos, Multilinear Principal Component Analysis of Tensor Object for Recognition, 18th International Conference on Pattern Recognition, ICPR 200, vol. 2, pp , [14] D. Xu, S. Yan, L. Zhang, H. Zhang, Z. Liu, and H. Shum, Concurrent Subspace Analysis,IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. -2, pp , [15] H. Lu, K. N. Plataniotis, A. N. Venetsanopoulos, MPCA: Multilinear Principal Component Analysis of Tensor Object,IEEE Transaction on Neural Networks, vol. 19, no. 1, pp , [16] S. Yan, D. Xu, Q. Yang, L. Jhang, X. Tang, and H.J. Jhang, Discriminant Analysis with Tensor Representation,IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. 1, pp , June [17] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, Eigenfaces vs. Fisherfaces: Recognition using Class specific linear Projection, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp , July [18] W. Zhao, R. Chellappa, A. Krishnaswamy, Discriminant Analysis of Principal Components for Face Recognition, Third IEEE International Conference on Automatic Face and Gesture Recognition, Proceedings, pp , April [19] J. Wang, K. N. Plataniotis, and A. N. Venetsanopoulos, Selecting Discriminant Eigenfaces for Face Recognition, Pattern Recognition Letters, vol. 26, no. 10, pp , 2005.

9 [20] K. Delac, M. Grgic, S. Grgic, Statistics in Face Recognition: Analyzing Probability Distributions of PCA, ICA and LDA Performance Results,Proceedings of the 4th International Symposium on Image and Signal Processing and Analysis, ISPA 2005, Zagreb, Croatia, pp , September [21] Vapnik. V.,An overview of statistical learning theory, IEEE Transactions on Neural Networks, vol. 10, no. 5, pp , [22] Gates, K. E., Fast and Accurate Face Recognition Using Support Vector Machines, IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 3, pp. 163, 25 June [23] Daoqiang Zhang, Zhi-Hua Zhou, 2D)2PCA: Two-directional twodimensional PCA for efficient face representation and recognition, Neurocomputing,vol. 69, pp , December [24] Wangxin Yu, Zhizhong Wang, Weiting Chen, A new framework to combine vertical and horizontal information for face recognition, presented at Neurocomputing, Vol. 72, no. 4-6, pp , January [25] Jin Wang, Armando Barreto, Lu Wang, Yu Chen, Naphtali Rishe, Jean Andrian and Malek Adjouadi, Multilinear principal component analysis for face recognition with fewer features, Presented at Neurocomputing, vol. 73, no , pp , June [26] C. Gold, P. Sollich, Model Selection for Support Vector Machine Classification, Presented at Neurocomputing, vol. 55, no. 1-2, pp , September [27] F. Lauer, G. Bloch, Incorporating Prior Knowledge in Support Vector Machine for Classification: A Review, Presented at Neurocomputing, vol. 71, no. 7-9, pp , March [28] A. J. Goldstein, L. D. Harmon, and A. B. Lesk,Identification of human faces, Proceedings of the IEEE,vol. 59, no. 5, pp , [29] R. Brunelli and T. Poggio, Face recognition: Features versus templates, IEEE Transactions on Pattern Analysis and Machine Intelligence,vol. 15, no. 10, pp , [30] T. Kanade, Picture Processing by Computer Complex and Recognition of Human Faces, Ph.D. thesis, Kyoto University, 1973 [31] C. Tripathi,K.P. Singh,Face Recognition using Eigen Tensor based Linear Discriminant Analysis with SVM Classifier, M.Tech Thesis, Indian Institute of Information Technology Allahabad, [32] A. Samal and P. A.Iyengar,Automatic recognition and analysis of human faces and facial expressions: A survey, Pattern Recognition, vol. 25, no. 1, pp. 6577, [33] B. Scholkopf, A. Smola, and K. R. Muller, Nonlinear component analysis as a kernel eigenvalue problem, Neural Computation, MIT Press, vol. 10,no. 5, pp , [34] S. Mika, G. Ratsch, J. Weston, B. Scholkopf, and K.-R. Mller, Fisher discriminant analysis with kernels, Neural Networks for Signal Processing IX, Proceedings of the 1999 IEEE Signal Processing Society Workshop, pp. 4148, 1999.

A New Multi Fractal Dimension Method for Face Recognition with Fewer Features under Expression Variations

A New Multi Fractal Dimension Method for Face Recognition with Fewer Features under Expression Variations A New Multi Fractal Dimension Method for Face Recognition with Fewer Features under Expression Variations Maksud Ahamad Assistant Professor, Computer Science & Engineering Department, Ideal Institute of

More information

A Hierarchical Face Identification System Based on Facial Components

A Hierarchical Face Identification System Based on Facial Components A Hierarchical Face Identification System Based on Facial Components Mehrtash T. Harandi, Majid Nili Ahmadabadi, and Babak N. Araabi Control and Intelligent Processing Center of Excellence Department of

More information

An Integrated Face Recognition Algorithm Based on Wavelet Subspace

An Integrated Face Recognition Algorithm Based on Wavelet Subspace , pp.20-25 http://dx.doi.org/0.4257/astl.204.48.20 An Integrated Face Recognition Algorithm Based on Wavelet Subspace Wenhui Li, Ning Ma, Zhiyan Wang College of computer science and technology, Jilin University,

More information

The Analysis of Parameters t and k of LPP on Several Famous Face Databases

The Analysis of Parameters t and k of LPP on Several Famous Face Databases The Analysis of Parameters t and k of LPP on Several Famous Face Databases Sujing Wang, Na Zhang, Mingfang Sun, and Chunguang Zhou College of Computer Science and Technology, Jilin University, Changchun

More information

FACE RECOGNITION USING SUPPORT VECTOR MACHINES

FACE RECOGNITION USING SUPPORT VECTOR MACHINES FACE RECOGNITION USING SUPPORT VECTOR MACHINES Ashwin Swaminathan ashwins@umd.edu ENEE633: Statistical and Neural Pattern Recognition Instructor : Prof. Rama Chellappa Project 2, Part (b) 1. INTRODUCTION

More information

Fuzzy Bidirectional Weighted Sum for Face Recognition

Fuzzy Bidirectional Weighted Sum for Face Recognition Send Orders for Reprints to reprints@benthamscience.ae The Open Automation and Control Systems Journal, 2014, 6, 447-452 447 Fuzzy Bidirectional Weighted Sum for Face Recognition Open Access Pengli Lu

More information

Linear Discriminant Analysis in Ottoman Alphabet Character Recognition

Linear Discriminant Analysis in Ottoman Alphabet Character Recognition Linear Discriminant Analysis in Ottoman Alphabet Character Recognition ZEYNEB KURT, H. IREM TURKMEN, M. ELIF KARSLIGIL Department of Computer Engineering, Yildiz Technical University, 34349 Besiktas /

More information

Multidirectional 2DPCA Based Face Recognition System

Multidirectional 2DPCA Based Face Recognition System Multidirectional 2DPCA Based Face Recognition System Shilpi Soni 1, Raj Kumar Sahu 2 1 M.E. Scholar, Department of E&Tc Engg, CSIT, Durg 2 Associate Professor, Department of E&Tc Engg, CSIT, Durg Email:

More information

Image-Based Face Recognition using Global Features

Image-Based Face Recognition using Global Features Image-Based Face Recognition using Global Features Xiaoyin xu Research Centre for Integrated Microsystems Electrical and Computer Engineering University of Windsor Supervisors: Dr. Ahmadi May 13, 2005

More information

LOCAL APPEARANCE BASED FACE RECOGNITION USING DISCRETE COSINE TRANSFORM

LOCAL APPEARANCE BASED FACE RECOGNITION USING DISCRETE COSINE TRANSFORM LOCAL APPEARANCE BASED FACE RECOGNITION USING DISCRETE COSINE TRANSFORM Hazim Kemal Ekenel, Rainer Stiefelhagen Interactive Systems Labs, University of Karlsruhe Am Fasanengarten 5, 76131, Karlsruhe, Germany

More information

Dimension Reduction CS534

Dimension Reduction CS534 Dimension Reduction CS534 Why dimension reduction? High dimensionality large number of features E.g., documents represented by thousands of words, millions of bigrams Images represented by thousands of

More information

Linear Discriminant Analysis for 3D Face Recognition System

Linear Discriminant Analysis for 3D Face Recognition System Linear Discriminant Analysis for 3D Face Recognition System 3.1 Introduction Face recognition and verification have been at the top of the research agenda of the computer vision community in recent times.

More information

GENDER CLASSIFICATION USING SUPPORT VECTOR MACHINES

GENDER CLASSIFICATION USING SUPPORT VECTOR MACHINES GENDER CLASSIFICATION USING SUPPORT VECTOR MACHINES Ashwin Swaminathan ashwins@umd.edu ENEE633: Statistical and Neural Pattern Recognition Instructor : Prof. Rama Chellappa Project 2, Part (a) 1. INTRODUCTION

More information

Eigenfaces and Fisherfaces A comparison of face detection techniques. Abstract. Pradyumna Desale SCPD, NVIDIA

Eigenfaces and Fisherfaces A comparison of face detection techniques. Abstract. Pradyumna Desale SCPD, NVIDIA Eigenfaces and Fisherfaces A comparison of face detection techniques Pradyumna Desale SCPD, NVIDIA pdesale@nvidia.com Angelica Perez Stanford University pereza77@stanford.edu Abstract In this project we

More information

Face Recognition Using SIFT- PCA Feature Extraction and SVM Classifier

Face Recognition Using SIFT- PCA Feature Extraction and SVM Classifier IOSR Journal of VLSI and Signal Processing (IOSR-JVSP) Volume 5, Issue 2, Ver. II (Mar. - Apr. 2015), PP 31-35 e-issn: 2319 4200, p-issn No. : 2319 4197 www.iosrjournals.org Face Recognition Using SIFT-

More information

Face Recognition Based on LDA and Improved Pairwise-Constrained Multiple Metric Learning Method

Face Recognition Based on LDA and Improved Pairwise-Constrained Multiple Metric Learning Method Journal of Information Hiding and Multimedia Signal Processing c 2016 ISSN 2073-4212 Ubiquitous International Volume 7, Number 5, September 2016 Face Recognition ased on LDA and Improved Pairwise-Constrained

More information

Image Processing and Image Representations for Face Recognition

Image Processing and Image Representations for Face Recognition Image Processing and Image Representations for Face Recognition 1 Introduction Face recognition is an active area of research in image processing and pattern recognition. Since the general topic of face

More information

Face Detection and Recognition in an Image Sequence using Eigenedginess

Face Detection and Recognition in an Image Sequence using Eigenedginess Face Detection and Recognition in an Image Sequence using Eigenedginess B S Venkatesh, S Palanivel and B Yegnanarayana Department of Computer Science and Engineering. Indian Institute of Technology, Madras

More information

Class-dependent Feature Selection for Face Recognition

Class-dependent Feature Selection for Face Recognition Class-dependent Feature Selection for Face Recognition Zhou Nina and Lipo Wang School of Electrical and Electronic Engineering Nanyang Technological University Block S1, 50 Nanyang Avenue, Singapore 639798

More information

Mobile Face Recognization

Mobile Face Recognization Mobile Face Recognization CS4670 Final Project Cooper Bills and Jason Yosinski {csb88,jy495}@cornell.edu December 12, 2010 Abstract We created a mobile based system for detecting faces within a picture

More information

A GENERIC FACE REPRESENTATION APPROACH FOR LOCAL APPEARANCE BASED FACE VERIFICATION

A GENERIC FACE REPRESENTATION APPROACH FOR LOCAL APPEARANCE BASED FACE VERIFICATION A GENERIC FACE REPRESENTATION APPROACH FOR LOCAL APPEARANCE BASED FACE VERIFICATION Hazim Kemal Ekenel, Rainer Stiefelhagen Interactive Systems Labs, Universität Karlsruhe (TH) 76131 Karlsruhe, Germany

More information

CHAPTER 3 PRINCIPAL COMPONENT ANALYSIS AND FISHER LINEAR DISCRIMINANT ANALYSIS

CHAPTER 3 PRINCIPAL COMPONENT ANALYSIS AND FISHER LINEAR DISCRIMINANT ANALYSIS 38 CHAPTER 3 PRINCIPAL COMPONENT ANALYSIS AND FISHER LINEAR DISCRIMINANT ANALYSIS 3.1 PRINCIPAL COMPONENT ANALYSIS (PCA) 3.1.1 Introduction In the previous chapter, a brief literature review on conventional

More information

Directional Derivative and Feature Line Based Subspace Learning Algorithm for Classification

Directional Derivative and Feature Line Based Subspace Learning Algorithm for Classification Journal of Information Hiding and Multimedia Signal Processing c 206 ISSN 2073-422 Ubiquitous International Volume 7, Number 6, November 206 Directional Derivative and Feature Line Based Subspace Learning

More information

Learning to Recognize Faces in Realistic Conditions

Learning to Recognize Faces in Realistic Conditions 000 001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050

More information

Semi-Supervised PCA-based Face Recognition Using Self-Training

Semi-Supervised PCA-based Face Recognition Using Self-Training Semi-Supervised PCA-based Face Recognition Using Self-Training Fabio Roli and Gian Luca Marcialis Dept. of Electrical and Electronic Engineering, University of Cagliari Piazza d Armi, 09123 Cagliari, Italy

More information

Local Similarity based Linear Discriminant Analysis for Face Recognition with Single Sample per Person

Local Similarity based Linear Discriminant Analysis for Face Recognition with Single Sample per Person Local Similarity based Linear Discriminant Analysis for Face Recognition with Single Sample per Person Fan Liu 1, Ye Bi 1, Yan Cui 2, Zhenmin Tang 1 1 School of Computer Science and Engineering, Nanjing

More information

Diagonal Principal Component Analysis for Face Recognition

Diagonal Principal Component Analysis for Face Recognition Diagonal Principal Component nalysis for Face Recognition Daoqiang Zhang,2, Zhi-Hua Zhou * and Songcan Chen 2 National Laboratory for Novel Software echnology Nanjing University, Nanjing 20093, China 2

More information

Enhancing Performance of Face Recognition System Using Independent Component Analysis

Enhancing Performance of Face Recognition System Using Independent Component Analysis Enhancing Performance of Face Recognition System Using Independent Component Analysis Dipti Rane 1, Prof. Uday Bhave 2, and Asst Prof. Manimala Mahato 3 Student, Computer Science, Shah and Anchor Kuttchi

More information

IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 14, NO. 1, JANUARY Face Recognition Using LDA-Based Algorithms

IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 14, NO. 1, JANUARY Face Recognition Using LDA-Based Algorithms IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 14, NO. 1, JANUARY 2003 195 Brief Papers Face Recognition Using LDA-Based Algorithms Juwei Lu, Kostantinos N. Plataniotis, and Anastasios N. Venetsanopoulos Abstract

More information

Robust Face Recognition Using Enhanced Local Binary Pattern

Robust Face Recognition Using Enhanced Local Binary Pattern Bulletin of Electrical Engineering and Informatics Vol. 7, No. 1, March 2018, pp. 96~101 ISSN: 2302-9285, DOI: 10.11591/eei.v7i1.761 96 Robust Face Recognition Using Enhanced Local Binary Pattern Srinivasa

More information

On Modeling Variations for Face Authentication

On Modeling Variations for Face Authentication On Modeling Variations for Face Authentication Xiaoming Liu Tsuhan Chen B.V.K. Vijaya Kumar Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213 xiaoming@andrew.cmu.edu

More information

Face Recognition using Eigenfaces SMAI Course Project

Face Recognition using Eigenfaces SMAI Course Project Face Recognition using Eigenfaces SMAI Course Project Satarupa Guha IIIT Hyderabad 201307566 satarupa.guha@research.iiit.ac.in Ayushi Dalmia IIIT Hyderabad 201307565 ayushi.dalmia@research.iiit.ac.in Abstract

More information

Recognition of Non-symmetric Faces Using Principal Component Analysis

Recognition of Non-symmetric Faces Using Principal Component Analysis Recognition of Non-symmetric Faces Using Principal Component Analysis N. Krishnan Centre for Information Technology & Engineering Manonmaniam Sundaranar University, Tirunelveli-627012, India Krishnan17563@yahoo.com

More information

Dimensionality Reduction and Classification through PCA and LDA

Dimensionality Reduction and Classification through PCA and LDA International Journal of Computer Applications (09 8887) Dimensionality Reduction and Classification through and Telgaonkar Archana H. PG Student Department of CS and IT Dr. BAMU, Aurangabad Deshmukh Sachin

More information

School of Computer and Communication, Lanzhou University of Technology, Gansu, Lanzhou,730050,P.R. China

School of Computer and Communication, Lanzhou University of Technology, Gansu, Lanzhou,730050,P.R. China Send Orders for Reprints to reprints@benthamscienceae The Open Automation and Control Systems Journal, 2015, 7, 253-258 253 Open Access An Adaptive Neighborhood Choosing of the Local Sensitive Discriminant

More information

FACE RECOGNITION BASED ON GENDER USING A MODIFIED METHOD OF 2D-LINEAR DISCRIMINANT ANALYSIS

FACE RECOGNITION BASED ON GENDER USING A MODIFIED METHOD OF 2D-LINEAR DISCRIMINANT ANALYSIS FACE RECOGNITION BASED ON GENDER USING A MODIFIED METHOD OF 2D-LINEAR DISCRIMINANT ANALYSIS 1 Fitri Damayanti, 2 Wahyudi Setiawan, 3 Sri Herawati, 4 Aeri Rachmad 1,2,3,4 Faculty of Engineering, University

More information

Restricted Nearest Feature Line with Ellipse for Face Recognition

Restricted Nearest Feature Line with Ellipse for Face Recognition Journal of Information Hiding and Multimedia Signal Processing c 2012 ISSN 2073-4212 Ubiquitous International Volume 3, Number 3, July 2012 Restricted Nearest Feature Line with Ellipse for Face Recognition

More information

FACE RECOGNITION USING PCA AND EIGEN FACE APPROACH

FACE RECOGNITION USING PCA AND EIGEN FACE APPROACH FACE RECOGNITION USING PCA AND EIGEN FACE APPROACH K.Ravi M.Tech, Student, Vignan Bharathi Institute Of Technology, Ghatkesar,India. M.Kattaswamy M.Tech, Asst Prof, Vignan Bharathi Institute Of Technology,

More information

Heat Kernel Based Local Binary Pattern for Face Representation

Heat Kernel Based Local Binary Pattern for Face Representation JOURNAL OF LATEX CLASS FILES 1 Heat Kernel Based Local Binary Pattern for Face Representation Xi Li, Weiming Hu, Zhongfei Zhang, Hanzi Wang Abstract Face classification has recently become a very hot research

More information

International Journal of Digital Application & Contemporary research Website: (Volume 1, Issue 8, March 2013)

International Journal of Digital Application & Contemporary research Website:   (Volume 1, Issue 8, March 2013) Face Recognition using ICA for Biometric Security System Meenakshi A.D. Abstract An amount of current face recognition procedures use face representations originate by unsupervised statistical approaches.

More information

Face recognition using Singular Value Decomposition and Hidden Markov Models

Face recognition using Singular Value Decomposition and Hidden Markov Models Face recognition using Singular Value Decomposition and Hidden Markov Models PETYA DINKOVA 1, PETIA GEORGIEVA 2, MARIOFANNA MILANOVA 3 1 Technical University of Sofia, Bulgaria 2 DETI, University of Aveiro,

More information

A Survey on Feature Extraction Techniques for Palmprint Identification

A Survey on Feature Extraction Techniques for Palmprint Identification International Journal Of Computational Engineering Research (ijceronline.com) Vol. 03 Issue. 12 A Survey on Feature Extraction Techniques for Palmprint Identification Sincy John 1, Kumudha Raimond 2 1

More information

Applications Video Surveillance (On-line or off-line)

Applications Video Surveillance (On-line or off-line) Face Face Recognition: Dimensionality Reduction Biometrics CSE 190-a Lecture 12 CSE190a Fall 06 CSE190a Fall 06 Face Recognition Face is the most common biometric used by humans Applications range from

More information

Application of 2DPCA Based Techniques in DCT Domain for Face Recognition

Application of 2DPCA Based Techniques in DCT Domain for Face Recognition Application of 2DPCA Based Techniques in DCT Domain for Face Recognition essaoud Bengherabi, Lamia ezai, Farid Harizi, Abderraza Guessoum 2, and ohamed Cheriet 3 Centre de Développement des Technologies

More information

An Efficient Face Recognition using Discriminative Robust Local Binary Pattern and Gabor Filter with Single Sample per Class

An Efficient Face Recognition using Discriminative Robust Local Binary Pattern and Gabor Filter with Single Sample per Class An Efficient Face Recognition using Discriminative Robust Local Binary Pattern and Gabor Filter with Single Sample per Class D.R.MONISHA 1, A.USHA RUBY 2, J.GEORGE CHELLIN CHANDRAN 3 Department of Computer

More information

Multi-Modal Human Verification Using Face and Speech

Multi-Modal Human Verification Using Face and Speech 22 Multi-Modal Human Verification Using Face and Speech Changhan Park 1 and Joonki Paik 2 1 Advanced Technology R&D Center, Samsung Thales Co., Ltd., 2 Graduate School of Advanced Imaging Science, Multimedia,

More information

Robust Face Recognition via Sparse Representation Authors: John Wright, Allen Y. Yang, Arvind Ganesh, S. Shankar Sastry, and Yi Ma

Robust Face Recognition via Sparse Representation Authors: John Wright, Allen Y. Yang, Arvind Ganesh, S. Shankar Sastry, and Yi Ma Robust Face Recognition via Sparse Representation Authors: John Wright, Allen Y. Yang, Arvind Ganesh, S. Shankar Sastry, and Yi Ma Presented by Hu Han Jan. 30 2014 For CSE 902 by Prof. Anil K. Jain: Selected

More information

Technical Report. Title: Manifold learning and Random Projections for multi-view object recognition

Technical Report. Title: Manifold learning and Random Projections for multi-view object recognition Technical Report Title: Manifold learning and Random Projections for multi-view object recognition Authors: Grigorios Tsagkatakis 1 and Andreas Savakis 2 1 Center for Imaging Science, Rochester Institute

More information

The Novel Approach for 3D Face Recognition Using Simple Preprocessing Method

The Novel Approach for 3D Face Recognition Using Simple Preprocessing Method The Novel Approach for 3D Face Recognition Using Simple Preprocessing Method Parvin Aminnejad 1, Ahmad Ayatollahi 2, Siamak Aminnejad 3, Reihaneh Asghari Abstract In this work, we presented a novel approach

More information

Facial Expression Recognition using Principal Component Analysis with Singular Value Decomposition

Facial Expression Recognition using Principal Component Analysis with Singular Value Decomposition ISSN: 2321-7782 (Online) Volume 1, Issue 6, November 2013 International Journal of Advance Research in Computer Science and Management Studies Research Paper Available online at: www.ijarcsms.com Facial

More information

MPSO- PCA Based Facial Recognition

MPSO- PCA Based Facial Recognition International Journal of Innovative Research in Electronics and Communications (IJIREC) Volume 1, Issue 1, April 2014, pp.27-32 www.arcjournal.org MPSO- PCA Based Facial Recognition Pushparaj Pa pushprajpal@gmail.com

More information

A new Face Database and Evaluation of Face Recognition Techniques

A new Face Database and Evaluation of Face Recognition Techniques A new Face Database and Evaluation of Face Recognition Techniques D. ALEXIADIS 1, V. SYRRIS 2, A. PAPASTERGIOU 3, A. HATZIGAIDAS 4, L. MARIUTA Technological Educational Institute of Thessaloniki Department

More information

SINGLE-SAMPLE-PER-PERSON-BASED FACE RECOGNITION USING FAST DISCRIMINATIVE MULTI-MANIFOLD ANALYSIS

SINGLE-SAMPLE-PER-PERSON-BASED FACE RECOGNITION USING FAST DISCRIMINATIVE MULTI-MANIFOLD ANALYSIS SINGLE-SAMPLE-PER-PERSON-BASED FACE RECOGNITION USING FAST DISCRIMINATIVE MULTI-MANIFOLD ANALYSIS Hsin-Hung Liu 1, Shih-Chung Hsu 1, and Chung-Lin Huang 1,2 1. Department of Electrical Engineering, National

More information

Face Recognition for Different Facial Expressions Using Principal Component analysis

Face Recognition for Different Facial Expressions Using Principal Component analysis Face Recognition for Different Facial Expressions Using Principal Component analysis ASHISH SHRIVASTAVA *, SHEETESH SAD # # Department of Electronics & Communications, CIIT, Indore Dewas Bypass Road, Arandiya

More information

Misalignment-Robust Face Recognition

Misalignment-Robust Face Recognition Misalignment-Robust Face Recognition Huan Wang 1 Shuicheng Yan 2 Thomas Huang 3 Jianzhuang Liu 1 Xiaoou Tang 1,4 1 IE, Chinese University 2 ECE, National University 3 ECE, University of Illinois 4 Microsoft

More information

Research Article A Multifactor Extension of Linear Discriminant Analysis for Face Recognition under Varying Pose and Illumination

Research Article A Multifactor Extension of Linear Discriminant Analysis for Face Recognition under Varying Pose and Illumination Hindawi Publishing Corporation EURASIP Journal on Advances in Signal Processing Volume, Article ID, pages doi:.// Research Article A Multifactor Extension of Linear Discriminant Analysis for Face Recognition

More information

Face Recognition using Tensor Analysis. Prahlad R. Enuganti

Face Recognition using Tensor Analysis. Prahlad R. Enuganti Face Recognition using Tensor Analysis Prahlad R. Enuganti The University of Texas at Austin Literature Survey EE381K 14 Multidimensional Digital Signal Processing March 25, 2005 Submitted to Prof. Brian

More information

Robust Face Recognition via Sparse Representation

Robust Face Recognition via Sparse Representation Robust Face Recognition via Sparse Representation Panqu Wang Department of Electrical and Computer Engineering University of California, San Diego La Jolla, CA 92092 pawang@ucsd.edu Can Xu Department of

More information

Face recognition based on improved BP neural network

Face recognition based on improved BP neural network Face recognition based on improved BP neural network Gaili Yue, Lei Lu a, College of Electrical and Control Engineering, Xi an University of Science and Technology, Xi an 710043, China Abstract. In order

More information

PCA and KPCA algorithms for Face Recognition A Survey

PCA and KPCA algorithms for Face Recognition A Survey PCA and KPCA algorithms for Face Recognition A Survey Surabhi M. Dhokai 1, Vaishali B.Vala 2,Vatsal H. Shah 3 1 Department of Information Technology, BVM Engineering College, surabhidhokai@gmail.com 2

More information

Face Recognition System Using PCA, LDA & Jacobi Method

Face Recognition System Using PCA, LDA & Jacobi Method Available onlinewww.ejaet.com European Journal of Advances in Engineering and Technology, 2017,4 (5): 326-331 Research Article ISSN: 2394-658X Face Recognition System Using PCA, LDA & Jacobi Method Neel

More information

Enhanced (PC) 2 A for Face Recognition with One Training Image per Person

Enhanced (PC) 2 A for Face Recognition with One Training Image per Person Enhanced (PC) A for Face Recognition with One Training Image per Person Songcan Chen, *, Daoqiang Zhang Zhi-Hua Zhou Department of Computer Science Engineering Nanjing University of Aeronautics Astronautics,

More information

Facial Feature Extraction by Kernel Independent Component Analysis

Facial Feature Extraction by Kernel Independent Component Analysis Facial Feature Extraction by Kernel Independent Component Analysis T. Martiriggiano, M. Leo, P.Spagnolo, T. D Orazio Istituto di Studi sui Sistemi Intelligenti per l Automazione - C.N.R. Via Amendola 1/D-I,

More information

Face detection and recognition. Many slides adapted from K. Grauman and D. Lowe

Face detection and recognition. Many slides adapted from K. Grauman and D. Lowe Face detection and recognition Many slides adapted from K. Grauman and D. Lowe Face detection and recognition Detection Recognition Sally History Early face recognition systems: based on features and distances

More information

An Effective Approach in Face Recognition using Image Processing Concepts

An Effective Approach in Face Recognition using Image Processing Concepts An Effective Approach in Face Recognition using Image Processing Concepts K. Ganapathi Babu 1, M.A.Rama Prasad 2 1 Pursuing M.Tech in CSE at VLITS,Vadlamudi Guntur Dist., A.P., India 2 Asst.Prof, Department

More information

Performance Evaluation of PCA and LDA for Face Recognition

Performance Evaluation of PCA and LDA for Face Recognition Performance Evaluation of PCA and LDA for Face Recognition S. K. Hese, M. R. Banwaskar Department of Electronics & Telecommunication, MGM s College of Engineering Nanded Near Airport, Nanded, Maharashtra,

More information

Dr. K. Nagabhushan Raju Professor, Dept. of Instrumentation Sri Krishnadevaraya University, Anantapuramu, Andhra Pradesh, India

Dr. K. Nagabhushan Raju Professor, Dept. of Instrumentation Sri Krishnadevaraya University, Anantapuramu, Andhra Pradesh, India Volume 6, Issue 10, October 2016 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Design and

More information

A Study on Different Challenges in Facial Recognition Methods

A Study on Different Challenges in Facial Recognition Methods Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 4, Issue. 6, June 2015, pg.521

More information

Two-View Face Recognition Using Bayesian Fusion

Two-View Face Recognition Using Bayesian Fusion Proceedings of the 2009 IEEE International Conference on Systems, Man, and Cybernetics San Antonio, TX, USA - October 2009 Two-View Face Recognition Using Bayesian Fusion Grace Shin-Yee Tsai Department

More information

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 8. Face recognition attendance system based on PCA approach

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 8. Face recognition attendance system based on PCA approach Computer Aided Drafting, Design and Manufacturing Volume 6, Number, June 016, Page 8 CADDM Face recognition attendance system based on PCA approach Li Yanling 1,, Chen Yisong, Wang Guoping 1. Department

More information

An Enhanced Face Recognition System based on Rotated Two Dimensional Principal Components

An Enhanced Face Recognition System based on Rotated Two Dimensional Principal Components An Enhanced Face Recognition System based on Two Dimensional Principal Components Garima, Sujit Tiwari Abstract Face has been one of the widely used modality from very beginning of biometrics recognition

More information

Study and Comparison of Different Face Recognition Algorithms

Study and Comparison of Different Face Recognition Algorithms , pp-05-09 Study and Comparison of Different Face Recognition Algorithms 1 1 2 3 4 Vaibhav Pathak and D.N. Patwari, P.B. Khanale, N.M. Tamboli and Vishal M. Pathak 1 Shri Shivaji College, Parbhani 2 D.S.M.

More information

APPLICATION OF LOCAL BINARY PATTERN AND PRINCIPAL COMPONENT ANALYSIS FOR FACE RECOGNITION

APPLICATION OF LOCAL BINARY PATTERN AND PRINCIPAL COMPONENT ANALYSIS FOR FACE RECOGNITION APPLICATION OF LOCAL BINARY PATTERN AND PRINCIPAL COMPONENT ANALYSIS FOR FACE RECOGNITION 1 CHETAN BALLUR, 2 SHYLAJA S S P.E.S.I.T, Bangalore Email: chetanballur7@gmail.com, shylaja.sharath@pes.edu Abstract

More information

Image Based Feature Extraction Technique For Multiple Face Detection and Recognition in Color Images

Image Based Feature Extraction Technique For Multiple Face Detection and Recognition in Color Images Image Based Feature Extraction Technique For Multiple Face Detection and Recognition in Color Images 1 Anusha Nandigam, 2 A.N. Lakshmipathi 1 Dept. of CSE, Sir C R Reddy College of Engineering, Eluru,

More information

Adaptive Sparse Kernel Principal Component Analysis for Computation and Store Space Constrained-based Feature Extraction

Adaptive Sparse Kernel Principal Component Analysis for Computation and Store Space Constrained-based Feature Extraction Journal of Information Hiding and Multimedia Signal Processing c 2015 ISSN 2073-4212 Ubiquitous International Volume 6, Number 4, July 2015 Adaptive Sparse Kernel Principal Component Analysis for Computation

More information

Extended Isomap for Pattern Classification

Extended Isomap for Pattern Classification From: AAAI- Proceedings. Copyright, AAAI (www.aaai.org). All rights reserved. Extended for Pattern Classification Ming-Hsuan Yang Honda Fundamental Research Labs Mountain View, CA 944 myang@hra.com Abstract

More information

Recognition: Face Recognition. Linda Shapiro EE/CSE 576

Recognition: Face Recognition. Linda Shapiro EE/CSE 576 Recognition: Face Recognition Linda Shapiro EE/CSE 576 1 Face recognition: once you ve detected and cropped a face, try to recognize it Detection Recognition Sally 2 Face recognition: overview Typical

More information

Face Biometrics Based on Principal Component Analysis and Linear Discriminant Analysis

Face Biometrics Based on Principal Component Analysis and Linear Discriminant Analysis Journal of Computer Science 6 (7): 693-699, 2010 ISSN 1549-3636 2010 Science Publications Face Biometrics Based on Principal Component Analysis and Linear Discriminant Analysis Lih-Heng Chan, Sh-Hussain

More information

Face Recognition using Principle Component Analysis, Eigenface and Neural Network

Face Recognition using Principle Component Analysis, Eigenface and Neural Network Face Recognition using Principle Component Analysis, Eigenface and Neural Network Mayank Agarwal Student Member IEEE Noida,India mayank.agarwal@ieee.org Nikunj Jain Student Noida,India nikunj262@gmail.com

More information

Laplacian MinMax Discriminant Projection and its Applications

Laplacian MinMax Discriminant Projection and its Applications Laplacian MinMax Discriminant Projection and its Applications Zhonglong Zheng and Xueping Chang Department of Computer Science, Zhejiang Normal University, Jinhua, China Email: zhonglong@sjtu.org Jie Yang

More information

[Gaikwad *, 5(11): November 2018] ISSN DOI /zenodo Impact Factor

[Gaikwad *, 5(11): November 2018] ISSN DOI /zenodo Impact Factor GLOBAL JOURNAL OF ENGINEERING SCIENCE AND RESEARCHES LBP AND PCA BASED ON FACE RECOGNITION SYSTEM Ashok T. Gaikwad Institute of Management Studies and Information Technology, Aurangabad, (M.S), India ABSTRACT

More information

Haresh D. Chande #, Zankhana H. Shah *

Haresh D. Chande #, Zankhana H. Shah * Illumination Invariant Face Recognition System Haresh D. Chande #, Zankhana H. Shah * # Computer Engineering Department, Birla Vishvakarma Mahavidyalaya, Gujarat Technological University, India * Information

More information

Hybrid Face Recognition and Classification System for Real Time Environment

Hybrid Face Recognition and Classification System for Real Time Environment Hybrid Face Recognition and Classification System for Real Time Environment Dr.Matheel E. Abdulmunem Department of Computer Science University of Technology, Baghdad, Iraq. Fatima B. Ibrahim Department

More information

Independent components of face images: A represent at ion for face recognit ion

Independent components of face images: A represent at ion for face recognit ion 4th Joint Symposium on Neural Computation Proceedings 1997 Independent components of face images: A represent at ion for face recognit ion Marian Stewart Bartlett Terrence J. Sejnowski University of California

More information

A Real Time Facial Expression Classification System Using Local Binary Patterns

A Real Time Facial Expression Classification System Using Local Binary Patterns A Real Time Facial Expression Classification System Using Local Binary Patterns S L Happy, Anjith George, and Aurobinda Routray Department of Electrical Engineering, IIT Kharagpur, India Abstract Facial

More information

Class-Information-Incorporated Principal Component Analysis

Class-Information-Incorporated Principal Component Analysis Class-Information-Incorporated Principal Component Analysis Songcan Chen * Tingkai Sun Dept. of Computer Science & Engineering, Nanjing University of Aeronautics & Astronautics, Nanjing, 210016, China

More information

Facial Feature Extraction Based On FPD and GLCM Algorithms

Facial Feature Extraction Based On FPD and GLCM Algorithms Facial Feature Extraction Based On FPD and GLCM Algorithms Dr. S. Vijayarani 1, S. Priyatharsini 2 Assistant Professor, Department of Computer Science, School of Computer Science and Engineering, Bharathiar

More information

FACE RECOGNITION BASED ON LOCAL DERIVATIVE TETRA PATTERN

FACE RECOGNITION BASED ON LOCAL DERIVATIVE TETRA PATTERN ISSN: 976-92 (ONLINE) ICTACT JOURNAL ON IMAGE AND VIDEO PROCESSING, FEBRUARY 27, VOLUME: 7, ISSUE: 3 FACE RECOGNITION BASED ON LOCAL DERIVATIVE TETRA PATTERN A. Geetha, M. Mohamed Sathik 2 and Y. Jacob

More information

PARALLEL GABOR PCA WITH FUSION OF SVM SCORES FOR FACE VERIFICATION

PARALLEL GABOR PCA WITH FUSION OF SVM SCORES FOR FACE VERIFICATION PARALLEL GABOR WITH FUSION OF SVM SCORES FOR FACE VERIFICATION Ángel Serrano, Cristina Conde, Isaac Martín de Diego, Enrique Cabello 1 Face Recognition & Artificial Vision Group, Universidad Rey Juan Carlos

More information

Face Recognition for Mobile Devices

Face Recognition for Mobile Devices Face Recognition for Mobile Devices Aditya Pabbaraju (adisrinu@umich.edu), Srujankumar Puchakayala (psrujan@umich.edu) INTRODUCTION Face recognition is an application used for identifying a person from

More information

Face Detection by Fine Tuning the Gabor Filter Parameter

Face Detection by Fine Tuning the Gabor Filter Parameter Suraj Praash Sahu et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol (6), 011, 719-74 Face Detection by Fine Tuning the Gabor Filter Parameter Suraj Praash Sahu,

More information

HUMAN TRACKING SYSTEM

HUMAN TRACKING SYSTEM HUMAN TRACKING SYSTEM Kavita Vilas Wagh* *PG Student, Electronics & Telecommunication Department, Vivekanand Institute of Technology, Mumbai, India waghkav@gmail.com Dr. R.K. Kulkarni** **Professor, Electronics

More information

Tensor Sparse PCA and Face Recognition: A Novel Approach

Tensor Sparse PCA and Face Recognition: A Novel Approach Tensor Sparse PCA and Face Recognition: A Novel Approach Loc Tran Laboratoire CHArt EA4004 EPHE-PSL University, France tran0398@umn.edu Linh Tran Ho Chi Minh University of Technology, Vietnam linhtran.ut@gmail.com

More information

FACE RECOGNITION FROM A SINGLE SAMPLE USING RLOG FILTER AND MANIFOLD ANALYSIS

FACE RECOGNITION FROM A SINGLE SAMPLE USING RLOG FILTER AND MANIFOLD ANALYSIS FACE RECOGNITION FROM A SINGLE SAMPLE USING RLOG FILTER AND MANIFOLD ANALYSIS Jaya Susan Edith. S 1 and A.Usha Ruby 2 1 Department of Computer Science and Engineering,CSI College of Engineering, 2 Research

More information

A Laplacian Based Novel Approach to Efficient Text Localization in Grayscale Images

A Laplacian Based Novel Approach to Efficient Text Localization in Grayscale Images A Laplacian Based Novel Approach to Efficient Text Localization in Grayscale Images Karthik Ram K.V & Mahantesh K Department of Electronics and Communication Engineering, SJB Institute of Technology, Bangalore,

More information

Comparison of Different Face Recognition Algorithms

Comparison of Different Face Recognition Algorithms Comparison of Different Face Recognition Algorithms Pavan Pratap Chauhan 1, Vishal Kumar Lath 2 and Mr. Praveen Rai 3 1,2,3 Computer Science and Engineering, IIMT College of Engineering(Greater Noida),

More information

Face and Facial Expression Detection Using Viola-Jones and PCA Algorithm

Face and Facial Expression Detection Using Viola-Jones and PCA Algorithm Face and Facial Expression Detection Using Viola-Jones and PCA Algorithm MandaVema Reddy M.Tech (Computer Science) Mailmv999@gmail.com Abstract Facial expression is a prominent posture beneath the skin

More information

Decorrelated Local Binary Pattern for Robust Face Recognition

Decorrelated Local Binary Pattern for Robust Face Recognition International Journal of Advanced Biotechnology and Research (IJBR) ISSN 0976-2612, Online ISSN 2278 599X, Vol-7, Special Issue-Number5-July, 2016, pp1283-1291 http://www.bipublication.com Research Article

More information

Face Recognition using Tensor Analysis. Prahlad R. Enuganti

Face Recognition using Tensor Analysis. Prahlad R. Enuganti Face Recognition using Tensor Analysis Prahlad R. Enuganti The University of Texas at Austin Final Report EE381K 14 Multidimensional Digital Signal Processing May 16, 2005 Submitted to Prof. Brian Evans

More information

A Direct Evolutionary Feature Extraction Algorithm for Classifying High Dimensional Data

A Direct Evolutionary Feature Extraction Algorithm for Classifying High Dimensional Data A Direct Evolutionary Feature Extraction Algorithm for Classifying High Dimensional Data Qijun Zhao and David Zhang Department of Computing The Hong Kong Polytechnic University Hung Hom, Kowloon, Hong

More information