SOCIAL MEDIA AUTHENTICATION USING FACIAL RECOGNITION
SOCIAL MEDIA AUTHENTICATION USING FACIAL RECOGNITION

Area / Specialisation: OpenCV with Python

Project Type: Facial Recognition

BATCH No : 06
REGISTER NUMBER NAME OF STUDENT NUMBER OF
MEETINGS WITH
GUIDE
17L31A0513 A.RAJEE PRIYA 02
17L31A0525 M.DINESH 01
17L31A0535 V.SOWMYA 03
17L31A0533 K.SUNIL 01

GUIDE NAME: Mrs DR.E.LAXMI LYDIA

DEPARTMENT OF CSE Date:

12/24/2018 1
Motivation/Origin

The Motivation behind this project:

 Face Recognition system are going to be used more and more in the future for
security reasons because they provide better performance over other security
systems.

 By using Face Recognition one’s verification and identification can be

improved.

 Face recognition systems are used extensively for surveillance, security,

authentication and access control systems.

 The proposed method for the authentication is highly efficient with respect to
authenticity of user. By using this method, only the authenticated user would
be able to get access to the account.

DEPARTMENT OF CSE Date

12/24/2018 2
 Abstract :
Social Media Webapp Authentication using Facial Recognition is a substantiate
mechanism for users to access their social media webapp accounts with the
face recognition along with typing passwords . If the password was forgotten
by the user then he/she should request for OTP through mail, which is a great
deal of time consuming chore in emergency situations. In that event, face
recognition turn out to be handy while accessing one’s account. Besides, Face
Recognition is used for security purpose. It avoids unauthorized access to social
media webapp accounts and also curtails the security risks. We proposed a
memory efficient way to register multiple angle and changing illumination face
image data to overcome the problem of Picture Attacks by hackers or
unauthorized users. Face Recognition can secure the personal data from the
attackers. The project uses python flask framework, opencv . It makes
information confidential.

DEPARTMENT OF CSE Date

12/24/2018 3
Problem Statement :

Different security and privacy models and techniques for authentication are used
over the social media web application. These techniques have limitations. In order
to minimize the security risk of authentication and validation , a methodology
called face detection and facial recognition is discussed. It provides the facility to
improve security and to avoid unauthorized access of social media webapp
accounts.

DEPARTMENT OF CSE Date

12/24/2018 4
 Existing System :

It is very effective and secure way of authentication. This

technique is being widely used all over the world for
testament and also proposed for attestation over social media
platforms.

1. Two step Authentication:

Two step Authentication is an authentication method that
requires the user to provide two or more verification factors to
gain access to a resource such as an application, online
account.

DEPARTMENT OF CSE Date

12/24/2018 5
2. Graphical Password
An image can be used as password rather than letters, digits or
special characters; user selects an image as a password at the
registration time and then uses that image as a password for
the authentication over the social media.

3. Signature Based Technique

Here, user inputs the signature in digital form using any
optical device and it is checked on the server.These signatures
are stored on the server in image form and later they are used
as an authentication parameter over the social media sites for
checking the validity of the user.

DEPARTMENT OF CSE Date

12/24/2018 6
Proposed System :
 In social media webapp, authentication should be fast and in reliable fashion
to ensure that the user who is accessing one’s account is appropriate user or
not.

 Face detection technique has multiple applications among various

organizations. This provides a secure way to log in to social media webapp.
This face recognition technique needs a camera for a device to log in to
the social media.

 All the smart devices have a built in camera in the current technology world.
After authentication of the credentials like username and password provided
by user. The system requires user’s authorization via face detection by
the camera on his or her device.

DEPARTMENT OF CSE Date

12/24/2018 7
 After successful detection, the user can login into the account. Else, it will send
a message connoting unsuccessful login to the mail. Coming to the face
detection system, it provides feasibility to login by using efficient
implementation. While taking authorization, the system requires different
angles of users face. Only the original user can provide his or her information
correctly.

 The need of this proposed system is that every device should have a built
in camera.

DEPARTMENT OF CSE Date

12/24/2018 8
Framework Used:

Open CV

Open CV is a cross-platform library using which we can develop

real-time computer vision applications. It mainly focuses on
image processing, video capture and analysis including features like
face detection and object detection.

Open CV is used for all sorts of image and video analysis,

like facial recognition and detection ,license plate reading,
photo editing, advanced robotic vision and optical character recognition

DEPARTMENT OF CSE Date

12/24/2018 9
Flowchart:

Fig : Login Process

DEPARTMENT OF CSE Date

12/24/2018 10
 IMPLEMENTATION

recognizer.py
Import face_recognition
import numpy as np
import cv2
import os
def Recognizer():
video = cv2.VideoCapture(0)
known_face_encodings = []
known_face_names = []
base_dir = os.path.dirname(os.path.abspath(__file__))
image_dir = os.path.join(base_dir, "static")
image_dir = os.path.join(image_dir, "profile_pics")
names = []
for root,dirs,files in os.walk(image_dir):
for file in files:
if file.endswith('jpg') or file.endswith('png'):
path = os.path.join(root, file)
img = face_recognition.load_image_file(path)
# label = file[:len(file)-4]
label = file
if label == 'default.jpg':
pass
else:
img_encoding = face_recognition.face_encodings(img)[0]
known_face_names.append(label)
26/12/2018
12/24/2018 11
 known_face_names.append(label)
known_face_encodings.append(img_encoding)
face_locations = []
face_encodings = []
while True:
check, frame = video.read()
small_frame = cv2.resize(frame, (0,0), fx=0.5, fy= 0.5)
rgb_small_frame = small_frame[:,:,::-1]
face_locations = face_recognition.face_locations(rgb_small_frame)
face_encodings = face_recognition.face_encodings(rgb_small_frame, face_locations)
face_names = []
for face_encoding in face_encodings:
matches = face_recognition.compare_faces(known_face_encodings,
np.array(face_encoding), tolerance = 0.5)
face_distances =
face_recognition.face_distance(known_face_encodings,face_encoding)
best_match_index = np.argmin(face_distances)
if matches[best_match_index]:
name = known_face_names[best_match_index]
face_names.append(name)
if name not in names:
names.append(name)

26/12/2018
12/24/2018 12
for (top,right,bottom,left), name in zip(face_locations, face_names):
top*=2
right*=2
bottom*=2
left*=2
 
cv2.rectangle(frame, (left,top),(right,bottom), (0,255,0), 2)
cv2.imshow("Face Recognition Panel",frame)
if cv2.waitKey(5000):
break
video.release()
cv2.destroyAllWindows()
 

26/12/2018
12/24/2018 13
RESULTS

Home page

Home page

Registration page

26/12/2018 14
12/24/2018
 Login page

Authenticating via webcam

26/12/2018
12/24/2018 15
 Succesful authentication
after login

New post

26/12/2018
12/24/2018 16
 Post is created

Update & Delete options

26/12/2018
12/24/2018 17
 Updated post

Pop up for delete post

26/12/2018
12/24/2018 18
Post is deleted

26/12/2018
12/24/2018 19
Structured Abstract
 PURPOSE: The purpose of the project is to secure an individual social media accounts
 OBJECTIVE: The objective of this project is to reduce unauthorized access
 PRACTICAL IMPLICATIONS:
For example, If this project is implemented then huge number of applications can be
developed based on this.
 Show how much % they completed in the form of a table.

DEPARTMENT OF CSE
100 0

100 0
100 0
100 0

100 0
0 0
0 0

Date
12/24/2018 20
Mapping of PO’s

PROGRAM SPECIFIC OUTCOMES

PSO1 PSO2 PSO3
CO1 2 3 2
CO2 2 2 2
CO3 1 3 2

Map with your project and mention the corresponding outcome. Here I given you as a reference

DEPARTMENT OF CSE Date

12/24/2018 21
Literature Review

IEEE Reference Paper-1

Enhancing the Authentication Mechanism of Social Media Websites
using Face Detection.

This paper includes different models used for the authentication of social
media account. Importance of security in authentication is described in paper.
Different security and privacy models and techniques for authentication are
used over the social media. These techniques have limitations. In order to
minimize the security risk of authentication and validation, a methodology
called face detection and facial recognition is discussed in this paper. It
provides the facility to improve security and to avoid unauthorized access of
social media accounts.

Reference link : https://ieeeexplore.iee.org/document/8981692

12/24/2018 22
Time Line

Stages Time required(for

example given below)
1. Problem identification and definition
06/03/2021-08/03/2021

2.  Collection of the related articles and

requirements  12/03/2021-14/03/2021

3.  Contact guide and industrial people for

designing  15/03/2021-21/03/2021

4.   Building a model for proposed work 

-

5.   Implementation, Testing, Results and

Analysis -

6.   Thesis, preparation and submission.

-

DEPARTMENT OF CSE Date

12/24/2018 23