NOVAL FACE RECOGNITION DOOR LOCKING USING AI
issue 1

NOVAL FACE RECOGNITION DOOR LOCKING USING AI

Payal A. Sakhare;                                                                 

M-Tech Student                                                                       

Rahul Dhuture; 

Guide 

Nagma Shaikh  

Co-Guide

Abstract—

Home security is one of the significant issues in day to day life and people always try to make life easier at the same time. Most of the doorways are constrained by peoples by the use of keys, security cards, secret phrase are the examples for opening the doorway. Among other things, human face recognition is one of the best techniques which are used for user authentication. Face recognition is a complex multidimensional structure and needs great computation methods for acknowledgment. The paper consist of three subsystems: specific face identification, face acknowledgment and programmed doorway get to control. This system based on AI, to make the door only accessible when your face is recognized by the recognition algorithms and then only you are allowed in by the house. When any person comes in front of the door, if face is registered it recognizes face by the recognition algorithms and the door gets unlock, if the face is not registered then it will raise an alarm. In case the face recognition part corrupts for that we added password function for the entrance.

Index Terms— Face Recognition, SIFT, SVM, Door Locking

I. INTRODUCTION

Humans as individuals, have unique characteristics and distinctive. These characteristics may be accustomed recognize or identify persons this is called as biometric recognition. Automatic face recognition is a type of biometric software application that may identify a particular individual in an exceedingly digital image by analyzing and comparing patterns. Face recognition is employed for security purpose. In today’s networked world, the necessity to keep up the safety of information or physical property is becoming both increasingly important and increasingly difficult. From time to time we hear about the crimes of mastercard fraud, computer break inn’s by hackers, or security breaches in an exceedingly company, building. Biometric access controls are automated methods of verifying or recognizing the identity of a living person on the premise of some physiological characteristics, like fingerprints or facial expression Facial reputation is a type of biometric software application which will identify a particular man or woman in an exceedingly digital image with the help of reading and evaluating patterns. Face recognition is employed for safety purpose. In latest networked international, the necessity to carry the safety of records or physical belongings is turning into both increasingly crucial and more and more difficult. In maximum of these crimes, the criminals have been taking advantage of a essential flaw inside  the conventional get admission to manipulate systems: the systems do not access without keys, passwords, identity cards, and PIN numbers. None of those approaches are really defining us. Rather, they simply are  manner to authenticate us. It goes without pronouncing that if someone steals, duplicates, or acquires those identification means, he or she can be able to get entry to our records or our personal belongings any time they need. This generation is based in a field called “biometrics”. Biometric get right of entry to controls are automated strategies of verifying or spotting the identity of a living character on the premise of some physiological traits, together with fingerprints or facial features. In today’s world of connectivity and clever devices there is an urgent need to adjust our current day to day items and make them smart. To change and modernize any item we need to eliminate its existing drawbacks and add greater functionality. The foremost drawbacks in a common door lock is that everybody can open a conventional door lock by means of duplicating or stealing the key and its truly impossible if we need our friends and family to go into our house, without being virtually present over there. So, to definitely convert this ordinary door lock into a smart lock, which could open the door each time we flip up in the front of the gate the database of photos are saved in database of friends .If the person is available in front of door then camera capture the photo and in shape with the reference picture. If the face matches then the door is open. So an era has come where devices can have interaction with its customers and at the equal time make certain of their protection and hold improvising themselves.

II. LITERATURE SURVEY

Door lock security systems are categorized based totally on era used 1) GSM primarily based, 2) smart card primarily based, 3) Password based, 4) Biometric based totally, 5) RFID based totally, 6) Door telephone primarily based, 7) Bluetooth primarily based, eight) Social networking sites based, 9) OTP primarily based, 10) motion detector based, eleven) VB based, 12) mixed gadget 2.1 Password based systems. Except fingerprint recognition the vein detector and iris scanner gives best and accurate result so, in the financial institution security gadget, microcontroller continuously observes the Vein Detector and Iris Scanner via keypad authenticated codes at some stage in night the wi-fi movement detector may be energetic, if any variant occurs in its output, it’ll be sensed by way of the controller and alert sounds can be given by it. These days, the short based most important factor analysis method is proposed wherein the amendment of major factor evaluation approach for the face reputation and face detection process is carried out. The image is captured by the net digital camera and it gets matched with the image stored inside the database. For detecting obstacles, the system requires numerous sensors. With the assist of GSM module, sends SMS to a respective number. Now a days a new version is created for protection of door without problems managed like far flung manipulate operations by means of a GSM hand set acts because the transmitter and the alternative GSM smartphone set with the DTMF associated with the motor attached to door with the use of DTMF decoder, a stepper motor and microcontroller unit. these days humans need to be relaxed even though they’re away from domestic. whilst the owner isn’t at his domestic, security of domestic and vital matters is the large trouble in the front of all two frameworks have been created which relies upon on GSM based totally generation. For detection of the gate-crashes, it takes place by means of taking pictures photo via internet digital camera. When peoples aren’t at their houses, the device sends notification in terms of SMS to the crisis variety. A novel administrator based device may login without any stretch to the device. The visitors report and listen their recorded messages and additionally robotically lock the door using cell verbal exchange technology.

III. PROBLEM FORMULATION

In the Artificial Intelligence, the auto authorization becomes a challenging and interesting topic. To realize particular person’s authorization, the face recognition is the best technology which attracted many researchers’ attention. The face recognition is explained as a machine learning process. The high dimensional feature of subject is received then the identity of subject must be provided. Every human faces have a particular angle and shape that requires complex calculations in order to recognize it. Individuals are distinguished by their faces, with which they are being identified.

IV. PROPOSED MATHOLOGY

Figure: 1 Block diagram of proposed system

Face Recognition is the process of determining whether or not a face is present in a picture unlike face recognition which distinguishes unique human faces, face detection simplest suggests whether or no longer a face is found in a picture. The group of feature based totally face popularity strategies is that the Scale Invariant Feature Transform (SIFT) proposed by Lowe. The key points that are notable and stable detects by the SIFT method for images in several resolutions and uses scale and rotation invariant descriptors to represent the key-points.

Feature extraction:- By using SIFT and PCA, feature extraction can done. SIFT could be a scale invariant feature transform which is employed for Invariant to scale change, Invariant to rotation change, Invariant to illumination change ,Robust to addition of noise, Robust to substantial range of transformation, Robust to 3D view point, Highly distinctive for discrimination. Principle component analysis (PCA) could be a multivariate technique that analyzes a face data within which observation are described by several inter-correlated dependent variables. The goal of PCA is to cut back the dimensionality of the information while retaining the maximum amount as possible of the variation present within the dataset. Face recognition:-By using support vector machine classifier face will be recognized. SVM is capable of fast and accurate classification of a high number of faces as a result Door Lock is open once the face is recognized. The serial interface is with computer; the information is send serially to microcontroller which decrypts the information and opens the door lock.

Figure: 2 Images from face database

V. CONCLUSION

An automatic face recognition and door lock open by use of SIFT algorithm is discussed and studied. The system comforts the process for door lock and the demand of smart home is rapidly increases in the today’s life. This system is a part of home automation for smart home. The system can be further enhanced to control the various appliances and also it can be added much more features like retina detection and fingerprint based door open.

REFERENCES:

  1. Momoko Hada, Ryoko Yamada and Shigeru Akamatsu, “How does the transformation of an avatar face giving a favorable impression affect human recognition of the face?” IEEE 2018
  2. Jalendu Dhamija, Tanupriya Choudhury, IEEE 2017 International Conference on Computational Intelligence and Networks
  3. Asmaa I. Ismail; Hanaa S. Ali, IEEE Member; Fathi A. Farag , “  Efficient Enhancement and Matching for Iris Recognition using SURF” 2015 IEEE
  4. Nitin K. Mahadeo, Andrew P. Papli´nski, Sid Ray , “Optimization of Iris Codes for Improved Recognition” 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.
  5. R. Rizal Isnanto , “Iris Recognition Analysis Using Biorthogonal Wavelets Tranform for Feature Extraction” 2014 1st International Conference on Information Technology, Computer and Electrical Engineering (ICITACEE).
  6. Gautam Narang, Soumya Singh Arjun Narang, “ Robust Face Recognition Method Based on SIFT Features Using Levenberg-Marquardt Backpropagation Neural Networks”, 978-14799-2764-7/13/$31.00 ©2013 IEEE
  7. Patrik Kamencay, Martin Breznan, Dominik Jelsovka, and Martina Zachariasova, “Improved Face Recognition Method based on segmentation Algorithm using SIFT-PCA”, 978-1-4673-1118-2/12/$31.00 ©2012 IEEE.

Related posts

Automated Disease Detection And Analysis Of Grape Leaves

admin

Defence Mechnisms By Using Decoy Technology

admin

Analysis of Normal and Epileptic EEG Signal with Filtering Method

admin

Leave a Comment