Scribd
Upload
6 views

Age Prediction and Gender Detection Using CNN

No

Uploaded by

hastidholakiya3390
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
6 views

Age Prediction and Gender Detection Using CNN

No

Uploaded by

hastidholakiya3390
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Gender Detection

&
Age Prediction
Team Leader Name : Ayesha Arab
Team Member Name : Hasti Dholakiya, Kirti Upadhyay,
Nishith Raval.
Gender Detection and Age Prediction

Outline:
• Introduction to the Project
• Data Collection and Preprocessing
• Model Training using Keras and TensorFlow
• System Diagram
• Tools and Technologies
• Results
• UseCases of Gender Detection and Age Prediction
Introduction
• This presentation will cover the functionality
of age prediction and gender detection using
Convolutional Neural Networks (CNNs). It will
also showcase the system diagram and flow
of training a model with accuracy and
results.

• In the frontend, we used HTML, CSS, and


JavaScript, while in the backend, we used
Machine Learning and CNNs.
Introduction

Convolutional Neural Input and output layers,


Networks (CNNs) for 02 multiple hidden layers
image recognition and
processing

01 03
Also used for Natural Regularized multilayer
Language Processing perceptron
(NLP)
04
Data Collection and Preprocessing
• Data Compilation:
Merging data from multiple files into one DataFrame, selecting
essential columns for age, gender, and facial attributes

• Data Refinement:
Cleaning data by mapping ages and genders to numerical labels,
dropping irrelevant records, and splitting the dataset for training
and testing.

• Image Processing:
Loading, resizing, and converting images to arrays for gender
and age model training and testing datasets.
Model Training using Keras &
TensorFlow
Model Training:
• The model is compiled using the Adam optimizer, sparse
categorical cross-entropy loss, and accuracy as the evaluation
metric.

• Early stopping is implemented with a patience of 3 epochs to


prevent overfitting. The training process involves 50 epochs with
a batch size of 32.

• Once trained, the model is saved as


age_gender_model_final.h5'. Subsequently, its performance is
evaluated on the test dataset.
System Diagram
.
Tools and Technologies
.

Front End Back End Python

TensorFlow and Keras: Used


for constructing and
training neural network
models

HTML OpenCV (cv2): Involved in


image processing tasks
CSS
JAVASCRIP Pandas and NumPy:
Employed for data
T manipulation and numerical
computation
Face Recognition Library:
Contributes to the
preprocessing stage
through face detection in
images.
Training and Results

Training process of the


CNN model
01
02 Measuring accuracy
and performance

Evaluation of age 03
prediction and gender
detection results
Results:
Results:
Results:
UseCases of Gender Detection & Age
Prediction
UseCases where this system will be applicable in various
sectors such as:
• Health Care
• Retail Marketing
• Online Advertisement
• Security and Law Enforcement
• Criminal Investigation
• Education
• Movie Recommendation

You might also like