Sie sind auf Seite 1von 21

Face Recognition by using OpenCV

By
T. Siva Nageswara Rao[N140191]
E3-CSE-01
Abstract :-
• The main aim of Face recognition project is to
Identifying a person with an image whereas the
person is unknown person. And this report describes
the face detection and recognition.
• It reports the technologies available in the Open-
Computer-Vision (OpenCV) library and methodology
to implement them by using python for face
detection.
• Here, I build this working model using LBPH
algorithm and Haar – Cascades and for face
recognition Eigenfaces and Fisher faces are also
used if we need. Such models are taken input images
and then understand and analyze the image. And give
desired results when we are given unknown person
image.
Introduction

• What is Face Recognition?

• When you look at your friend walking down the


street or picture of him , you recognize that he is
your friend.

• Here main thing is that, face of your friend. So, you


can easily recognize him by looking at his face. Well ,
this is what you doing in face recognition.
Introduction(contd.)

•But how does face recognition works?


• It is quite simple and intuitive. When you meet someone
for the first time in your life you don’t recognize him,
right? When he talks or seeing you. look at his face, eyes
,nose ,mouth ,color and all. That is your mind learning
for the face recognition of that person face data.

• While talking, he tells his name ,so your mind will learn
that collected data belong to his name.

• Next time when you see same person you can easily
recognize him. This is how face recognition works.
Requirements
Prerequisites:
• Need to know about python and it’s Libraries.
• Pandas , Numpy.
System:
• 4GB RAM with 500GB Hard Disk.
Technologies:
• Python
• OpenCV , Numpy , and Python ML Libraries
Tools:
• Spyder (Anaconda tool)
Basis for the Project:-

• The whole project based on the face recognition


system, And how It will work and How it will storing
data ,these things are mainly focus in this project. And
How the faces can be recognized.

• And It can detect the face of the person that, we meet


at the first time. And then , If it is first time ,it will store
the features in vector list for the next time purpose.
Basis for the Project(Cntd.):-

• If the person is already met . Then, It will compare the


features with those , who are present in the vectors
list. So, it will easily identify the person’s name.

• It is there in items list, So the person can be identified


by the features.

• Finally ,it can recognize the person’s name based on


features that present in vector list.
Problem Definition:-
• Existing System:
• There is already existed systems like Principal
Component Analysis (PCA) and Linear
Discriminant Analysis (LDA). PCA is to reduce the
large dimensionality of the data space to the
smaller intrinsic dimensionality of feature face.

• The jobs of PCA are prediction, redundancy


removal, feature extraction, data compression
etc. PCA can do the face recognition but not
efficient as our method.
Problem Definition:-
• Proposed System:
• So, our intention is to reduce the time to
recognize the unknown person. Face recognition
by using OpenCV is efficiently do the recognition
by using the algorithms LBPH and Haar-
Cascades.
• And it reaches users expectations and easier way
to recognize a person. And I am planning to
design face recognizer and that is available to all
users.
Flow chart Representation :
Procedure:-
Face Detection: Look at the picture and find a face in it. We need
to give a some random pictures in this step. It will detect faces.

Data Gathering: Extract unique characteristics of face, that it can


use to differentiate him/her from another person ,like eyes mouth
, nose…. Etc.

Data Comparison: Despite variations in light or expression ,it will


compare those unique features to all the features of the all the
people you know.

Face Recognition: It will determine the face of the person that, we


already given as the testing data.
Face Detection:-
• In this step we are using Haarcascade frontal face to
classify the different faces in the image. And after
that, founded faces.
• It returns the position of the detected faces as
Rect(x,y,w,h).
Data Gathering:-
• In this step we are Gathering the different type of
images from the testing data. Those are new
images(New person Image) , that we want to train the
images by using the training module.

• Here we are using for each person one directory to


train the persons faces from the image.

• If we have person face , then easily we can detect the


face of the person for the next time. If that person met
in future.
Data Gathering(Contd.):-
• And we are using labels to the detected faces.
Data Comparison:-
• In this step we are taking unknown person images as a
input images.

• Then , we are comparing each and every image with


already trained data in last step. So after comparison ,
we need to find the correct person of the input image.

• In this Data comparison, we are using LBPH Face


recognizer to recognize the unknown person face in
trained data.
Data Comparison(Contd.):-
• We need to train the LBPH face recognizer algorithm.
Then we predicting the faces in the prediction module

• And this LBPH Face recognizer is trained with faces


and labels.
• After that it can easily recognize the face of the person
Face Recognition:-
• In this step we are using all the above steps. That we
are taking one test image and that will be taken as
prediction module input.
Final Output:-
Advantages:-
• We can easily Identify the unknown person.
• By using this, we can easily Identify the customers that
are frequently buy items.
• By using this, we make live attendance for class

Disadvantages:-
• Face Recognition requires high-quality cameras and
advanced software to ensure accuracy and speed.
• Changes in appearance i.e, including facial hair and
camera angle can also cause issues.
Conclusion:-
• Here, that face recognizer needs to be good in
recognize the unknown person’s face, actually It has to
be. This will be used in different real time examples
like, checking person’s face while entering into the
metro trains, checking the criminal’s images easily.

• And it can be easily identifying the person who is theft


in ATM’s. In Future it will be used in major projects like
above ones.

Das könnte Ihnen auch gefallen