[IJET-V1I1P4] Author :Abdurrauf A. Ibrahim, Murtala, A. Garba

Please download to get full document.

View again

of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report
Category:

Documents

Published:

Views: 6 | Pages: 4

Extension: PDF | Download: 0

Share
Related documents
Description
The goal of this paper is to present a critical survey of existing literatures on human face detection and recognition over the last 4-5 years. An application for automatic face detection and tracking in video streams from surveillance cameras in public or commercial places is discussed in this paper. Prototype is designed to work with web cameras for the face detection and tracking system based on Visual 2010 C# and Open CV. This system can be used for security purpose to record the visitor face as well as to detect and track the face. Keywords:- Face Detection, Face Recognition, Open CV, Face Tracking, Video Streams.
Tags
Transcript
  International Journal of Engineering and Techniques – Volume 1 Issue 1, 2015   ISSN: 2395-1303 http://www.ijetjournal.org  Page 1 Face Detection and Recognition in Video Juhi Raut, Snehal Patil 1 , Shraddha Gawade, Prof. Mamta Meena ( Guide ) 2   B.E. Computer Engineering (pursuing),Atharva College of Engineering, Mumbai Malad (w), India. ---------------------------------------------------------- ************************************ ----------------------------------------------------   Abstract: The goal of this paper is to present a critical survey of existing literatures on human face detection and recognition over the last 4-5 years. An application for automatic face detection and tracking in video streams from surveillance cameras in public or commercial places is discussed in this paper. Prototype is designed to work with web cameras for the face detection and tracking system based on Visual 2010 C# and Open CV. This system can be used for security purpose to record the visitor face as well as to detect and track the face. Keywords:- Face Detection, Face Recognition, Open CV, Face Tracking, Video Streams.  ---------------------------------------------------------- ************************************ ----------------------------------------------------   I.   INTRODUCTION Human have a remarkable ability of identifying faces in a variety of poses. It is highly desirable that this ability be replicated in computers and can be utilized at the basic levels. Face detection and recognition is an application of biometrics. Face recognition is becoming an active research area in several disciplines such as image processing, pattern recognition, computer vision, neural networks, psychology and physiology. It is a dedicated process and an application of the general object recognition process. Automatic face recognition is an attractive biometric approach, since it focuses on the same identifier to distinguish one person from another that is their faces. One of its main goal is to understanding complex human visual system and the knowledge of how humans represent faces in order to discriminate different identities with high accuracy. Face detection and recognition from video is an application that uses new method for detecting and recognizing faces from video frames which provided from video cameras. The best result obtained by using Principal Component Analysis. In this approach the overall face detection, feature extraction and face recognition is carried out in a single step. II.   FACE DETECTION   Face detection [1] is necessary to know whether an image contains a face or not. Automatic detection of image is the first step in most atomic vision system. Generally, automatic face detection [6] and recognition systems are comprised of three steps as shown in fig 1. Image Fig1. Basic flow of Face Detection and Recognition System It is an effective computer of a complete system and allows for both demonstration and testing in a real environment as identifying the sub-region of the image containing a face will significantly reduce the subsequent processing and allow a more specific model to be applied to the recognition task. Face detection is to locate a face in a given image and to separate it from the remaining scene. Several approaches have been proposed to fulfill the task  .  A.    Elliptical Structure The elliptical structure method locates the head outline by the edge finder and then fits an ellipse to mark the boundary between the head region and the background. However, this method is applicable only to frontal views, the detection of non-frontal views needs to be investigated.   RESEARCH ARTICLE OPEN ACCESS Face Detection Feature Extraction Face RecognitiInput Result  International Journal of ISSN: 2395-1303 htt:  B.   Face Space In face space approach, images of faces d radically when projected into the face projections of non face images appear q This basic idea is used to detect the presen a scene. At every location in the image, distance between the local sub images and The detection uses a cascade of boost working with Haar-like features[1] to deci region   of an image is a face. Haar–likes fe input to the basic classifier. The featu particular classifier is a specified by its s within the region of interest and the scale. III.   FACE   RECOGNITION Face Recognition generally involves tw Detection [5], where an image is searche face, then image processing cleans up the for easier recognition. Fig 2. Block Diagram of Face Detection and Recog Face Recognition [5], where that processed face is compared to a databa faces, to decide who that person is. The Face Recognition is Face detection from an The Open CV library makes it fairly ea frontal face in an image using its Haar-Detector. The Block diagram of a fac system is as shown in fig 2 [1]. Face dete are example of general class of patter systems require similar components to normalize the face, extract a set of featur these to a database of stored examples [recognition systems perform which typic rectangular bounding box around the face Engineering and Techniques – Volume 1 Issue 1,  //www.ietournal.or  o not change space, while ite different. ce of faces in calculate the ace space. ed classifiers de whether a atures are the e used in a ape, position stages: Face to find any facial image ition System etected and se of known first stage in Image. y to detect a ascade Face recognition tion systems recognition locate and s and match 5]. All faces ally places a r faces in the images. This can be achieved rob We can use 2D feature extract Eigen faces that operate on all t face detected recognition. This better extract out the required fa with pose. There are five gener recognition system. The phases ar  1.   Capture image 2.   Detect face in image 3.   Feature extraction 4.   Compare with database 5.   Recognize face IV.   PROPOSED   SYSTEM The system is to build that wi highlight the detected face in vid be able to identify and track the The system should take live in should be able to detect faces, re and compare it with the databas track of people. It should also co in video. The application is bas and Open CV. The different modules and t application are shown in above ta TABLE I. F ACE D ETEC TABLE II. F EATURE E XTR TABLE III. F ACE R ECOG Model Name Face Detectio Input Given Real Time Vid Output Given Faces in Frame Procedure Steps 1. System tak from camera 2. It detects h returns its co-o 3. After the cperforms the l from images. Model Name Face Detectio Input Given Set of detected Output Given Feature vector Procedure Steps 1. The system t 2. It Spatial Fe and return the f 2015   Page 2 stly and in real time . ion method [4] that is e image pixels in the   allows the systems to e features and to deal l phases [4] in face : ll recognize faces and eo. The system should person with his name. put from camera and cognize them in video . System should keep unt the people present d on Visual 2010 C# eir working used in les I, II, III. TION M ODULE   ACTION M ODULE   ITION M ODULE   module o Stream es continues video stream uman face from frame and dinates -ordinates are obtained, it gical operation to get faces module Faces. akes set of detected faces atures using the Eigen faces eature Vector.  International Journal of ISSN: 2395-1303 htt: V.   PRINCIPAL   COMPONENT   A PCA [3] is a dimensionality reducti based on extracting the desired number component of the multidimensional da principal component is the linear combi original dimension that has the maximum v th principal component is the linear com the highest variance, subject to being orth first principal component. The basic vectors constructed by PCA dimension as the input face in images; the “Eigenfaces”. PCA is an information theor coding and decoding face image may giv the information content of face image, emsignificant local and global features. We w the relevant   information in a face image, efficiently as possible and compare one f with a database of models encoded similarl   VI.   EIGEN   FACES Eigen face approach [3] is one of appearance-based face recognition m method utilizes the idea of the princip analysis and decomposes face images into characteristic feature images called eigenf in Fig 3 [7]. There are a variety of appro representation, which can be classifie categories: template-based, feature- appearance-based.   Model Name Face Detection module Input Given Captured faces Output Given Matching Person faces Procedure Steps 1. The System then takes in and confirms the face co-ord 2. If the Captured face matc in Video then they are track Engineering and Techniques – Volume 1 Issue 1,  //www.ietournal.or  ALYSIS n technique of principal a. The first ation of the ariance; the n bination with gonal to n-1 had the same were named approach of insight into phasizing the ant to extract ncode it as a ace encoding . the earliest thods. This l component a small set of ces as shown ches for face into three ased, and Fig 3. Eigen Fac  A.   Template-Matching The simplest template-matchirepresent a whole face using a si 2-D array of intensity, which is u the original face image. The a matching [3] is the simplicity; h large memory requirement and in  B.    Appearance-Based In appearance-based [2] approac project onto a linear subspace of l subspace is first constructed by analysis on a set of training ima its eigenvectors. Now, the conce extended to Eigen features, such mouth etc. for the Detection [3] o C.   Feature-Based In feature-based [2] approaches, g as position and width of eye eyebrow's thickness and arch invariant moments, are extracte Feature-based [3] approaches requirement and a higher re template-based approach. captured faces inates. hes with faces d in Video. 2015   Page 3 s ng [2] approaches gle template, that is, a sually an edge map of vantage of template-wever, it suffers from fficient matching.   h the face images are w dimensions. Such a principal component es, with eigenfaces as ts of eigenfaces were as Eigen eyes, Eigen facial features . eometric features, such s, nose, and mouth, s, face breadth, or to represent a face. ave smaller memory ognition speed than  International Journal of Engineering and Techniques – Volume 1 Issue 1, 2015   ISSN: 2395-1303 http://www.ijetjournal.org  Page 4 VII.   OPEN   CV   Open CV (Open Source Computer Vision Library) is a library of programming functions mainly aimed at real time computer vision. It is free for use under the open source BSD license. The library is cross-platform. It focuses mainly in real-time image processing. If the library finds Intel’s Integrated Performance Primitives on the system , it will use these proprietary optimized routines to accelerate itself. The library was srcinally written in C and this C interface makes Open CV portable to some specific platform such as digital signal processors. Wrappers for languages such as C#, Python, Rubyand Java (using Java CV) have been developed to encourage adoption by a wider audience . However, since version 2.0, Open CV includes both its traditional C interface as well as a new C++ interface. This new interface seeks to reduce the number of lines of code necessary to code up vision functionality as well as reduce common programming errors such as memory leaks that can arise when using Open CV in C. Most of the new developments and algorithms in Open CV are now developed in the C++ interface. Unfortunately, it is much more difficult to provide wrapper in a other language to C++ code as opposed to C code; therefore the other language wrappers are generally lacking some of the newer Open CV 2.0 features. Emgu CV is a cross platform .Net wrapper to the Intel Open CV image processing library. Allowing Open CV functions to be called from .NET compatible languages such as C#, VB, VC++, Python etc. Emgu CV has two layers of wrapper: ã   Layer 1: The basic layer contains function, structure and enumeration mappings which directly reflect those in Open CV. ã   Layer 2: The second layer contains classes that mix in advantages from the .NET world. The CvInvoke class (Emgu.CV.CvInvoke) provides a way to directly invoke Open CV within .NET languages. Each method in this class corresponds to a function in Open CV of the same name. Emgu CV also borrows some   existing structures in .NET to represents structures in Open CV.   VIII.   ADVANTAGES 1.   It enables real time detection of person’s in video. 2.   It provides advanced query to detect person in video. 3.   It provides a multiple face detection and recognition in a video. 4.   Processing time is comparatively fast.   Acknowledgment: We especially thank our internal project guide Prof. Mamta Meena for their guidance, encouragement, cooperation and suggestions given at progressing stages of project. Our effort would be incomplete without the mention of computer departments Project Heads. Finally, we would like to thank our Principal Prof. Shrikant Kallurkar and H.O.D Prof. Mahendra Patil and all teaching, non- teaching staff of the college and friends for their moral support rendered during the course of the project work and for their direct and indirect involvement in the completion of our project work  . References: 1)   Faizan Ahmad, Aaima Najam and Zeeshan Ahmed,” Image Based FaceDetection and Recognition: State Of Art”, IJCSI International Jounrnal Of Computer Science Issues, Vol.9, Issue 6, No 1, November 2012. 2)   Aparna Behara, M.V.Raghunadh, “Real Time Face Recognition system for Time amd Attendance Applications”, International Journal of Electrical , Electronics and Data Communication, ISSN:2320-2084, Volume-1, Issue-4. 3)   Ming-Hsuan Yang, Member, IEEE, David J.Kriegman, Senior Member, IEEE and Narendra Ahuja, Fellow, IEEE, “ Detecting Face In Images: A Servey”, IEEE Transaction On Pattern Analysis and Machine Intelligence, Vol.24, No-1, January 2002. 4)   Mohammad A. Ali, Abdelfatah Aref Tamimi and Omaima N. A. AL-Allaf, “ Integrated System For Monitoring And Recognizing Students During Class Session”, The International Journal of Multimedia and Its Applications (IJMA) Vol. 5, No.6, December 2013. 5)   J.Suneetha, “ A Survey On Video-based Face Recognition Approaches”, International Journal of Application or Innovation in Engineering And Management (IJAIEM), Volume 3, Issue 2, February 2014. 6)   Wikipedia,http://en.wikipedia.org/wiki/Face_detection. 7)   “http://docs.opencv.org/trunk/_images/eigenface_reconstruction_opencv1.png”.  
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks