Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
44 views9 pages

Project Report 2

The document describes a face detection system developed by Vineet Thakur for their Bachelor's degree. The system was trained on a database of 7,488 images of the developer's face. It uses OpenCV and MATLAB to detect faces in videos and images by tracking facial features and comparing to the training database. The system can currently only recognize the developer's face due to the small training database and lack of computational power. The document outlines the steps taken to implement the face detection system and compares it to Amazon Rekognition, noting the commercial system has significantly more capabilities.

Uploaded by

Vineet Thakur
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
44 views9 pages

Project Report 2

The document describes a face detection system developed by Vineet Thakur for their Bachelor's degree. The system was trained on a database of 7,488 images of the developer's face. It uses OpenCV and MATLAB to detect faces in videos and images by tracking facial features and comparing to the training database. The system can currently only recognize the developer's face due to the small training database and lack of computational power. The document outlines the steps taken to implement the face detection system and compares it to Amazon Rekognition, noting the commercial system has significantly more capabilities.

Uploaded by

Vineet Thakur
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

FACE DETECTION SYSTEM

Submitted in partial fulfillment of the requirements for the award of degree of


BACHELOR OF ENGINEERING
IN
COMPUTER SCIENCE & ENGINEERING

Submitted to: Submitted by:


Er. KULVINDER SINGH​ ​VINEET THAKUR
18BCS2109

​DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING


Chandigarh University, Gharuan
August 2020
Implementation (100%)
The final result of the project is that it detects face from videos.
The code shown in the below image is the part of the system that is going to detect the faces and
track them.
I have used myself as the model or you can say for the sample video I have used my face.
The database I have created a database of my images. Total no of images in my database is
7,488.

The above images have been used to train my model. The only thing my software can recognise
is my face due to lack of computation power my machine can perform and it takes a lot of time
to train a model and also the database of individuals.
The below image was taken from the video that my model is going to detect the face and
recognize it.

My model is going to first track the face using track points.


The Hue channel data will check the contrast of the image.

The images that are attached below was taken while my model was tracking and detecting my
face at the same time.
The same can done with an image. If we insert an image then it will compare it with the stored
information and will detect the face.
Output validation and comparison

The main component of our software is that it is going to do the image processing. It is going to
capture the image then take it and then process it. After processing it then it will check it with the
database with whom it resembles. If it is a match then it will show the details of the person
otherwise it will say not found.

To do the image processing we are going to use a software MATLAB.


We are going to use it to do image processing because it has inbuilt tools for it, we can also
include other toolbox as per requirement.
Below I’m going to give the basic explanation of the steps that I’m going to perform do the
image processing.
We are going to use Computer vision system toolbox(image below)
These are the basics of the face detection we are going to use. The step by step algorithm is
going to be used to detect the face.
The Hue contrast can be seen below which I have used in the code to make the picture more
visible.
Sometimes the image is not very clear then we are going to some options to clear the image by
improving contrast like below
We already know that Amazon have a facial recognition system i.e

My system and the Amazon Rekognition have the difference that can’t be measured at all. The
Amazon is so good at developing their system. They make you pay less as you process more
images or videos you use to process.

You might also like