Bollini Lokesh, Abhishek Nallamothu, Mr.S.Planiappan

 
ABSTRACT
Day by day technology is changing and way of securing and automation is also trending. Facial recognition (or face recognition) is a type of biometric software application that can identify a specific individual in a digital image by analyzing and comparing patterns. Facial recognition systems are commonly used for security purposes but are increasingly being used in a variety of other automation applications. In real time, face recognition algorithms deal with large data base. Execution of these face recognition algorithms take high computational power and time on large database. Our objective is to improve speed of face recognition on large data base by using PCA algorithm. The goal of our proposing PCA algorithm is to reduce the dimensionality of the data by mapping the data into a lower dimensionality subspace while retaining as much as possible of the variation present in the original dataset. We formally prove this algorithm on ORL face data base with best precision.
Keywords: PCA: Principle Component Analysis, MATLAB: Matrix Laboratory, ORL: Olivetti Research Laboratory
INTRODUCTION
Facial recognition (or face recognition) is a type of biometric software application that can identify a specific individual in a digital image by analyzing and comparing patterns. Facial recognition systems are commonly used for security purposes but are increasingly being used in a variety of other applications. The Kinect motion gaming system, for example, uses facial recognition to differentiate among players. Currently, a lot of facial recognition development is focused on smartphone applications. Smartphone facial recognition capacities include image tagging and other social networking integration purposes as well as personalized marketing. A research team at Carnegie Mellon has developed a proof-of-concept iPhone app that can take a picture of an individual and — within seconds — return the individual’s name, date of birth and social security number. Facebook uses facial recognition software to help automate user tagging in photographs. Here’s how facial recognition works in Facebook: Each time an individual is tagged in a photograph, the software application stores information about that person’s facial characteristics. When enough data has been collected about a person to identify them, the system uses that information to identify the same face in different photographs, and will subsequently suggest tagging those pictures with that person’s name. Facial recognition software also enhances marketing personalization. For example, billboards have been developed with integrated software that identifies the gender, ethnicity and approximate age of passersby to deliver targeted advertising.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

The main aim of this project is to improve the computational speed of face recognition by using PCA algorithm. This can be done by reducing the dimensionality of images, while doing computations on images in data base. We propose a PCA algorithm with reduced dimensionality in calculations, and we formally prove this algorithm on ORL face data base of ten different images of each of 40 distinct subjects with best precision.
RELATED WORK
The proposed face recognition system by using PCA algorithm overcomes certain limitations of the existing face recognition system. It is based on reduction of dimensionality and extracting the dominating features of a set of human faces stored in the database and performing mathematical operations on the values corresponding to them. Hence when a new image is fed into the system for recognition then it will reduce dimensionality of new image and extract the main features to compute and find the distance between the input image and the stored images. Thus, some variations in the new face image to be recognized can be tolerated. When the new image of a person differs from the images of that person stored in the database, the system will be able to recognize the new face and identify who the person is. The proposed system is better mainly due to the use of facial features rather than the entire face. Its advantages are in terms of:

Recognition accuracy and better discriminatory power Computational cost because of reduction in dimensionality and removing of noise from data set
Concentrating on main features require less processing to train the PCA.
Because of the use of dominant features and hence can be used as an effective means of authentication

Principle Component Analysis
Principal Components Analysis (PCA) was invented by Karl Pearson in 1901 and is now used in many fields of science. It is a way of identifying patterns in data, and expressing the data in such a way as to highlight their similarities and differences. Since patterns in data can be hard to find in data of high dimension, where the luxury of graphical representation is not available, PCA is a powerful tool for analyzing data. The other main advantage of PCA is that once you have found these patterns in the data, and you compress the data, ie. by reducing the number of dimensions, without much loss of information.
The main aim of this project is to improve the computational speed of face recognition by using PCA algorithm. This can be done by reducing the dimensionality of images, while doing computations on images in data base. We propose a PCA algorithm with reduced dimensionality in calculations, and we formally prove this algorithm on ORL face data base of ten different images of each of 40 distinct subjects with best precision.
Actually issues arise once performing arts face recognition in a very high-dimensional area (curse of dimensionality). therefore we have a tendency to area unit managing spatiality issues in face recognition performance. important enhancements will be achieved by 1st mapping the information into a lower-dimensional sub-space. Mapping knowledge|of knowledge|of information} into lower dimensional data is feasible by PCA formula. In PCA formula, 1st it’ll convert all face pictures (N X N pixel) in knowledge base into face vector (N2X1 dimensionality).It hundreds of these face vectors into one matrix (N2X M matrix; here M=number of face pictures in database).It cipher average face vector(N2X1 dimensionality) by doing mean on all face vectors. It calculate normalized face vectors matrix (N2X M dimensionality) by subtracting average face vector from every face vector. It calculate variance matrix to search out out best Eigen|chemist}|chemist} vectors (best Eigen vectors represent best Eigen faces).It calculate signature of image and place it in( M X S dimensionality; here S=number of signatures). PCA converts input image (image for face recognition) into face vector, then it converts into normalized face vector and thereby it verify weight vector of input image. Finally it compare weight vectors and thereby it verify the person.
Face recognition bioscience is that the science of programming a laptop to acknowledge a personality’s face. once someone is listed during a face recognition system, a video camera takes a series of snapshots of the face and so represents it by a singular holistic code. once somebody has their face verified by the pc, it captures their current look and compares it with the facial codes already hold on within the system. The faces match, the person receives authorization; otherwise, the person won’t be known. the prevailing face recognition system identifies solely static face pictures that just about specifically match with one among the photographs hold on within the information. once this image captured nearly specifically matches with one among the photographs hold on then the person is known and granted access.
once this image of someone is significantly totally different, say, in terms of facial features from the photographs of that person that area unit already hold on within the information the system doesn’t acknowledge the person and thence access are denied.
The existing or ancient face recognition system has some limitations which may be overcome by adopting new ways of face recognition:

The existing system cannot tolerate variations within the new face image. It needs the new image to be nearly specifically matching with one among the photographs within the information which can otherwise end in denial of access for the individual.
The performance level of the prevailing system isn’t considerable.

CONCLUSION
The PCA method is an unsupervised technique of learning that is mostly suitable for databases that contain images with no class labels. PCA improve speed of face recognition by mapping higher dimensionality of face image into lower dimensionality. PCA provides best precision in face recognition process. In future we are planning to implement automation in security and automation in attendance by using this algorithm. We will try to get more efficiency and precision by combining this algorithm with other algorithms. We are planning to implement this algorithm for recognizing multi faces by combining this algorithm with other face recognition algorithms.
REFERENCES:
[1] A.S Syed navaz, T. Dhevi sri & Pratap mazumdar “Face recognition using principle component analysis and neural networks” International Journal of Computer Networking, Wireless and Mobile Communications (IJCNWMC) ISSN 2250-1568 Vol. 3, Issue 1, Mar 2013, 245-256
[2] Lindsay I Smith “A tutorial on Principal Components Analysis”February 26, 2002
[3] Sasan Karamizadeh, Shahidan M. Abdullah, Azizah A. Manaf, Mazdak Zamani, Alireza Hooman “ An Overview of Principal Component Analysis” Journal of Signal and Information Processing 2013, 4, 173-175
[4] Toshiyuki Sakai, M. Nagao, Takeo Kanade, “Computer analysis and classification of photographs of human face,” First USA Japan Computer Conference, 1972
[5] Yuille, A. L., Cohen, D. S., and Hallinan, P. W., “Feature extraction from faces using deformable templates”, Proc. of CVPR, (1989).