FACE DETECTION USING PRINCIPAL COMPONENT ANALYSIS

Transcription

1 International Journal of Computer Engineering & Technology (IJCET) Volume 7, Issue 3, May-June 2016, pp , Article ID: IJCET_07_03_016 Available online at Journal Impact Factor (2016): (Calculated by GISI) ISSN Print: and ISSN Online: IAEME Publication FACE DETECTION USING PRINCIPAL COMPONENT ANALYSIS Parteek Kumar M.Tech. CSE PPIMT Hisar, Haryana, India Praveen Sehgal HOD, CSE PPIMT Hisar, Haryana, India ABSTRACT This paper describes for a robust face recognition system using skin segmentation technique. This paper addresses the problem of detecting faces in color images in the presence of various lighting conditions. In this paper the face is preprocessed using histogram equalization to avoid illumination problems and then is detected using skin segmentation method. The principal component analysis using neural network is used to recognize the extracted facial features. Key words: Histogram Equalization, Principal Component Analysis Cite this Article: Parteek Kumar and Praveen Sehgal, Face Detection Using Principal Component Analysis, International Journal of Computer Engineering and Technology, 7(3), 2016, pp INTRODUCTION A facial recognition system is an application that uses image processing and is used for identification and verification of a person in an image or a video frame. It is not possible for human to capture all the faces in his mind and recognize them. Face detection and recognition has a number of applications nowadays. Some of them are listed below: 1. Virtual reality 2. Database recovery 3. Multimedia 4. Computer entertainment 5. Information security 6. Biometric applications editor@iaeme.com

2 Face Detection Using Principal Component Analysis 7. Home video surveillance system. In this paper we will use face recognition system for both identification and verification using principle component analysis. Advantages and Disadvantages of PCA The key advantages of PCA are listed below: 1. Recognition with PCA method is simple and efficient as compared to other approaches. 2. Data compression is achieved. 3. No knowledge of geometry of face is required [5]. 4. The key disadvantages of PCA are: 5. The covariance matrix is difficult to solve accurately and hence is time consuming.[5] 6. The size and location of each face image must remain similar. [7] 7. The method is very sensitive to scale. 2. LITERATURE SURVEY Face recognition is one of those challenging problems and till now, there is no technique that provides a robust solution to all situations. Several algorithms and techniques for face recognition have been developed in the past by researchers. Some of them are discussed briefly in this section. The problem of automatic face recognition involves three key steps/subtasks: 1. Detection of faces 2. Feature extraction 3. Identification and/or verification. Sometimes, different subtasks can be carried out simultaneously. For example, the facial features (eyes, nose, mouth) extraction step can be performed with the face detection. A popular and robust face detection algorithm uses an object detector developed at MIT by Viola and Jones [1]. The detector uses a cascade of boosted classifiers working with Haar-like features to decide whether a region of an image is a face. On web blogspot [2] present a face recognition system that attempts to recognize faces using skin segmentation technique. Yanjiang Wang et al. [3] proposed a fast face detection method in color images under complex background. The method firstly calculates similar pixel color image with the person's skin color clustering and region segmentation, and then use the face features of the wavelet decomposition analysis for each candidate area, if the detection is similar to the characteristics of the regional distribution of the face with one of the predefined model, the regional representative of the face. In [4], Aamer S.S.Mohamed presented an approach that relies on skin based color, while features can be extracted from two dimensional Discrete Cosine transform and neural networks which used to detect faces by using skin color from DCT coefficient of Cb and Cr features vectors. 3. METHODOLOGY The previous section illustrated different techniques and methods of face detection and recognition. Each category of method performs well in certain criteria and also has drawbacks as well. Systems with robustness and certain level of accuracy are still far away. Keeping in view our case study the following architecture is proposed for the detection and recognition system editor@iaeme.com

3 Parteek Kumar and Praveen Sehgal The image is to be preprocessed by applying histogram equalization to avoid illumination problem. As there are three steps in face recognition system, first the face will be detected (if there is a face in the image or video) by skin segmentation technique. A range is specified for the chrominance component of an image. Based on this specified range, the system is able to differentiate between skin color and the background. This gives face on the image. For second step i.e. facial features extraction, the following steps will be performed: 1. Acquire image 2. Apply transform i.e. Haar Transform 3. Threshold image 4. Extract details of image 5. Scanning 6. Extract features Then after extracting the features, the face has to be recognized that is the third step. For face recognition we will be using a holistic matching approach i.e. PCA using neural method. To perform PCA several steps will be performed: Figure 1 Steps to perform PCA The eigen object recognizer class performs all of this and then feeds the transposed data as a training set into a Neural Network. When it is passed to recognize it performs PCA and compares the generated Eigen values and Eigenvectors to the ones from the training set.the Neural Network then produces a match if one has been found or a negative match if no match is found. 4. EXPERIMENTAL ANALYSIS First of all we have performed face detection using skin segmentation technique. To avoid illumination problem, we preprocessed our image using histogram equalization editor@iaeme.com

4 Face Detection Using Principal Component Analysis Figure 2 Original Image The next step is to detect the skin regions in an image. But because of certain broken blobs they are identified as separate blobs and this makes it difficult to detect faces with accuracy. We would fill the gaps with white spaces so as to make it a solid white blob. MATLAB has an inbuilt function for the same known as imfill. Before performing imfill we need to convert our grayscale image into binary image. At last a red bounding box has to be put around the detected face. Figure 3 Face Detected Image 5. CONCLUSION This paper presents an efficient approach for face detection of image which handled illumination issues. For preprocessing, histogram equalization is used and for face detection skin segmentation technique is used in order to detect faces in an image. In future it can be used for face recognition with the help of Principal Component Analysis. REFERENCES [1] Paul Viola and Michael Jones, Robust real-time object detection, In International Journal of Computer Vision, [2] [3] Yanjiang Wang and Baozong Yuan, A Fast Human Face Detection Method From Color Images Under Complex Background [J], Acta Electronica Sinica, 2002, 30, editor@iaeme.com

5 Parteek Kumar and Praveen Sehgal [4] Aamer S.S. Mohamed, Ying Weng, Stan S Ipson, Jianmin Jiang, Face Detection Based on Skin Color in Image by Neural Networks, IEEE, 2007 [5] P. J. Phillips, P. J. Flynn, T. Scruggs, K. W. Bowyer, J. Chang, K. Hoffman, J. Marques, J. Min and W. Worek, Overview of the Face Recognition Grand Challenge, in Computer vision and pattern recognition, CVPR IEEE Computer Society Conference on, 2005, pp [6] D. Srinivasulu Asadi, Ch. DV Subba Rao and V. Saik-rishna A Comparative Study of Face Recognition with Principal Component Analysis and Cross- Correlation Technique, International Journal of Computer Applications Vol. 10, [7] S.K. Singh, D. S. Chauhan, M. Vatsa, R. Singh, A Robust Skin Color Based Face Detection Algorithm, Tamkang Journal of Science and Engineering, 6(4), pp (2003) [8] Gunjan Dashore and Dr. V.Cyril Raj, An Efficient Method For Face Recognition Using Principal Component Analysis(PCA), IJATER, 2(2), March 2012 [9] Taranpreet Singh Ruprah, Face Recognition Based on PCA Algorithm with Neural Network, International Journal of Computer Science & Informatics (IJCSI), Vol.- II, Issue-1, 2, December 2013 [10] Mattew Turk and Alex Pentland, Eigenfaces for Recognition, SPIE Vol.1192 IRCVV in (i989), pp:23-32 [11] Nazish, Face recognition using neural networks. Proc. IEEE INMIC 2001, pp: [12] T. Chen, W. Yin, X.-S. Zhou, D. Comaniciu, T. S. Huang, Total Variation Models for Variable Lighting Face Recognition and Uneven Background Correction, IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(9), 2006, pp [13] Rein-Lien Hsu, Mohamed Abdel-Mottaleb and Anil K. Jain, Face Detection in Color Images, IEEE Transactions on Pattern Analysis and Machine [14] Ming-Hsuan Yang, David J. Kriegman and Na-rendra Ahuja, Detecting Faces in Images, IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(1), pp ,January [15] Song, J., Chi, Z., and Liu, J. (2006). A robust eye detection method using combined Binary edge and intensity information. Pattern Recognition, 39(6): [Song et al.,2006] [16] Zhao, W., Chellappa, R., Phillips, P., and Rosenfeld, A. (2003). Face recognition: A Literature survey. ACM, Computing Surveys, 35(4): [Zhaoet al., 2003] [17] E. Bagherian, R. Wirza, N.I. Udzir, Extract of Facial Feature Point, IJCSNS International Journal of Computer Science and Network Security, 9 (1), January [18] V.Vezhnevets, V.Sazonov A. Andreeva, A Survey on Pixel- Based Skin Color Detection Techniques, Graphics and Media Laboratory, Moscow State University, Moscow, Russia. [19] P. Viola and M. Jones. Robust real-time face detection. International Journal of Computer Vision, 57(2): , May [20] Jyoti Verma, Vineet Richariya, Face Detection And Recognition Model Based On Skin Colour And Edge Information For Frontal Face Images, International Journal of Computer Engineering and Technology, 3(3), 2012, pp [21] A.Hemlata and Mahesh Motwani, Single Frontal Face Detection by Finding Dark Pixel Group and Comparing Xy-Value of Facial Features, International Journal of Computer Engineering and Technology, 4(3), 2012, pp editor@iaeme.com