Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Neural Network System for Face Detection
Transcript of Neural Network System for Face Detection
MATLAB (Matrix Laboratory)
is a high-level programming language solving high-performance problems with complex numeric calculations and visualization.
First Stage :
searching for potential face regions:
1. Skin color segmentation in YCbCr color space.
2. Label all four-connected components and find the center of each block.
3. Find any three centers of three different blocks forming an isosceles triangle.
4. Clip the potential face regions.
Feed the result into neural network system, and obtain the output (face or non-face).
Senior project -2013
Sarah AlEssa 210043560
Asrar AlMoqbil 210023024
Areej AlQhtani 210038067
Huda AlSulmi 210039589
Fatimah AlBraidy 210042627
Dr. Gamal M. Behery
Face detection system is automated system that takes an image and extract the face out of it.
face detection is an important prior step for face recognition system,
which is widely used as in security systems, face verification systems, telecommunication, video surveillance.
Aim of this project
is to construct an algorithm that demonstrates all the stages we need to go through in order to build an efficient face detection system.
In order to proceed in face detection algorithm, we need to cover 3 base knowledge areas:
Gray level image
Image segmentation is a process of partitioning the image into regions or segments that makes it easier to read its details or analyzing them.
Color Image processing
Skin Color Segmentation
In YCbCr color space, the color is
displayed as three numerical values that are put together to produce the resulted color, these components are:
Y luminance component( brightness)
Cb is blue minus luma (B - Y)
Cr is red minus luma (R – Y)
YCbCr color space
MATLAB and digital image processing
Image Processing Toolbox.
MATLAB and Neural Networks
Neural Network Toolbox.
Feed Forward Neural Network
Back-Propagation Neural Network
A Model of a Single Neuron
Feed-Forward neural network
Hidden Nodes and Layers
Back Propagation Architecture
This algorithm describes the back propagation in neural network with one hidden layer.
The algorithm includes three steps:
Computing the error
Update the weights
First Stage :
1. Skin color segmentation in YCbCr color space:
2. Connected component labeling:
3. Forming an isosceles triangle:
4. Clip the potential face regions:
Second Stage :
Build the neural network system
After completing the stages:
1. searching for potential face regions
2. face verification
the face in the image should be detected, and the face region should be marcked using square shape around it -for example.
Thank you for your attention.
Fatimah AlBraidy Sarah AlEssa Asrar AlMoqbil Areej AlQhtani Huda AlSulmi