International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395 -0056
Volume: 04 Issue: 03 | Mar -2017
p-ISSN: 2395-0072
www.irjet.net
Smart Doorbell System based on Face Recognition Jaychand Upadhyay1, Parkar Rida2, Sunidhi Gupta3, Noman Siddique4 1Asst.
Professor, Dept. of IT Engineering, Xavier Institute of Engineering, Maharashtra, India 2Dept. of IT Engineering, Xavier Institute of Engineering, Maharashtra, India 3Dept. of IT Engineering, Xavier Institute of Engineering, Maharashtra, India 4Dept. of IT Engineering, Xavier Institute of Engineering, Maharashtra, India
---------------------------------------------------------------------***--------------------------------------------------------------------1. INTRODUCTION Abstract - In recent years appreciable progress has been made in the field of face recognition. Face recognition system is an intelligent application, which can recognize or confirm a man from advanced sources, for example, (computerized picture or a video stream). Through the work of computer engineering engineers, PCs can now beat people in many face recognition assignments, especially those in which substantial databases of human faces must be looked. The easy approach to perceive any face is by making an analogy with the facial components from the input picture (tested image) in a facial database. A system with the ability to distinguish and recognize faces has numerous potential applications including crowd and airport surveillance, private security and enhanced human-computer interaction. This project aims to supersede highly priced image processing boards by putting to use Raspberry pi board with ARMv7 Cortex-A7 as the core within Opencv library. This project is mainly based on image processing by porting the Opencv library to the Raspberry Pi board. The ultimate crucial errand in face recognition is how to find the most identical between the tested and training faces.
During the past 3 decades, intensive analysis has been conducted on automatically recognizing the identity of people supported their facial pictures. In spite of the existence of multiple technologies like fingerprint and iris recognition, the external body part remains one in every of the foremost widespread cues for identity recognition in biometry. Face recognition possesses the non-intrusive nature and is usually effective without the participant’s cooperation or knowledge. It makes a decent compromise between performance dependableness and social acceptance and well balances security and privacy. Alternative biometric ways don't possess these benefits. Maybe, fingerprint recognition technique re-quire collaborate to join forces in creating cooperate physical contact with the device surface [1]. Similarly, iris recognition strategies need cooperation in putting their eyes properly relative to the camera. Face identification represents one in every of the foremost used styles of biometry [1]. The face recognition embedded systems are good enough to be employed in totally different applications like terrorist’s identification, security systems and identity verification access. After all it's enforced in several public and even dedicated areas. Due to the well-developed technologies linked to engineering, we will get satisfying results of face identification. The extracted details from faces are analyzed and compared with the already existing similar face operated details within the database. For example in monitoring systems the detection of an anonymous face more than once leads to saving this face traits in the database for further identification. This strategy is very useful in detecting criminals and thieves.
In this paper, how to recognize a face is introduced; for evaluating the proposed system, the author has used two analysis algorithms which are Eigen face and Independent Component Analysis (ICA). The local dataset utilized as a part of this paper is pre-processed using statistical standard techniques. Pre-processing software, Face Identification Evaluation System Version 5.0 under Unix Shell scripts, was written via ANSII C code, which is provided by the Colorado State University (CSU). Independent Component Analysis algorithm (ICA) is written using Matlab R2012b for face recognition implementation. The system is based on the criteria of low power consumption, resources optimization, and enhanced operation speed. This paper reviews the related work in the field of home automation systems and presents the system design, software algorithm and implementation.
2. RELATED WORK This project uses the Eigen faces algorithm using opencv library to perform face recognition. The script can capture a picture that is born-again to gray scale image, then apply the Eigen face approach and sites a face then crop the image in N² dimensions. To decrease the number of images, face pictures are re generated into two-dimensional array with eight bit intensity values, with the accurate trait extraction the algorithm calculate the common face image (in grey-scale mode for additional data reduction) and deduct the ensuing vector from every Eigen face vector to finally acquire the
Key Words: Camera, Doorbell, Eigen Faces, Face Recognition, Independent Component Analysis (ICA), Raspberry Pi, Security.
© 2017, IRJET
|
Impact Factor value: 5.181
|
ISO 9001:2008 Certified Journal
| Page 2840