Mini Project On

Gender and Age Detection using

OpenCV

Under the guidance of: By:

Dr. S. Zahoor UI Huq K Swarna Kanth Reddy -555
S Yaswanth Reddy – 5C3
A Sai Kiran – 5B8
 CONTENTS

• Introduction
• Project Overview
• Existing and Proposed Systems
• Method/Techniques involved in the project
• Software and Hardware Requirements
• Applications
• Output Screen
• Conclusion
 INTRODUCTION

 Increasing amount of applications, on gender and age detection.

 Performance of existing methods on real-world images.

 deep-convolutional neural networks (CNN), a significant increase in

performance can be obtained on these tasks.
 PROJECT OVERVIEW

 We used UTKFace dataset for age and gender classification.

 Trained CNN (Convolutional Neural Networks) using UTKFace

dataset for age and gender prediction.
 Created a python application using OpenCV deep learning module
to perform real time age and gender detection.
 Encountered with accuracy of 87.6
 UTKFACE DATASET

• 20k+ face images with aligned and cropped faces

• labelled by age, gender, and ethnicity
• Labels are formated like [age]_[gender]_[race]_[date&time].jpg
• [age] is an integer from 0 to 116, indicating the age
• [gender] is either 0 (male) or 1 (female)
 EXISTING SYSTEM

• Age/Gender estimation are based on calculating ratios

between different measurements of facial features .
• Once facial features (e.g. eyes, nose, mouth, chin, etc.) are
localized and their sizes and distances measured, ratios
between them are calculated and used for classifying the face
into different age categories
PROPOSED SYSTEM

PROPOSED SYSTEM
CONVOLUTIONAL NEURAL NETWORKS

• These networks have 3 types of layers: Input layer, hidden layer and
output layer.
• In these networks, data moves from the input layer through the
hidden nodes and to the output nodes.
• The convolutional, pooling and ReLU layers - learnable features
extractors,
• fully connected layers - machine learning classifier.
CONVOLUTION NEURAL NETWORK WITH HIDDEN LAYERS
 NETWORK ARCHITECTURE

The network comprises of only three convolutional layers and

two fully-connected layers with a small number of neurons.
Images are first rescaled to 256 X 256 and a crop of 227 X 227
is fed to the network.
 Each convolutional layer, followed by a rectified linear
operator (ReLU), a max pooling layer taking the maximal value
of 3 X 3 regions with two-pixel strides and a local response
normalization layer.
 Last fully connected layer maps to the final classes for age or
gender.
NETWORK ARCHITECTURE
NETWORK ARCHITECTURE
S/W & H/W REQUIREMENTS
Software Requirements Hardware Requirements

• Sypder (Anaconda • i3 or higher Processor

Navigator) based Computer
• Packages- Keras, Numpy, • 4GB - RAM
Sklearn, OpenCV,
MatplotLib
 APPLICATIONS

Application Areas - Better ad targeting for customers, better

content recommendation system, security surveillance, and
other fields as well.
Social media websites like Facebook could use the information
about the age and gender of the people to better infer the
context of the image.
In human machine interaction.
OUTPUT SCREEN
 CONCLUSION

• CNN can be used to provide improved age and gender

classification results.
• Both real time and static face detection was carried out.
Taking example from the related problem of face recognition
we explore how well deep CNN perform on these tasks using
Internet data.
• One can also try to use a classification model instead of
regression for Age Prediction by classifying into set of ranges
of age.
THANK YOU