project face detection

16
CSE 2100 Software Development Project 1 Project Title Face Detection

Upload: abu-saleh-musa

Post on 05-Dec-2014

2.235 views

Category:

Education


3 download

DESCRIPTION

 

TRANSCRIPT

Page 1: Project Face Detection

CSE 2100Software Development Project 1

Project Title

Face Detection

Page 2: Project Face Detection

Supervisor

Mr. Md. Aminul Haque Akhand Assistant Professor

Department of Computer Science and EngineeringKhulna University of Engineering and Technology

Credit: Abu Saleh Md. Musa (0907013) Sanjoy Dutta (0907008)

Page 3: Project Face Detection

Objectives

• The objective of our project is to design software that can detect human faces from an image.

Page 4: Project Face Detection

What is Face Detection

• Face detection is a computer technology that determines the locations and sizes of human faces in arbitrary (digital) images. It detects facial features and ignores anything else, such as buildings, trees and bodies.

Page 5: Project Face Detection

Why we chose Face Detection Project?

• Compatible with Modern Era.• Not common in JAVA.• Basic programme for Recognition(Recognition

is not possible without Detection).• Security Maintenance and Media

Empowering.• Needed for visual applications in Robotics.

Page 6: Project Face Detection

Examples of Implementation 1 Picasa Photo Viewer

(people panel)

Page 7: Project Face Detection

Examples of Implementation 2

• Facebook Tagging

Page 8: Project Face Detection

Procedure at a Glance• Read an image from disk (.JPG, etc.)• Convert it into a jjil.core.Image• Generally we’ll have an RGB image (colored image) and

so need to convert it to 8-bit grayscale, which is what the Gray8DetectHaarMultiScale class requires.

• Load facial properties to the class form Haar profile for detecting faces.

• Apply Gray8DetectHaarMultiScale to our 8-bit grey image.

• Retrieve result from Gray8DetectHaarMultiScale.

Page 9: Project Face Detection

Step by Step Analysis

As part of preprocessing we ensured certain things to make our software functional:

• The input is a colored image• There are multiple faces with frontal view and upright orientation• The size of faces within the image should approximately be the

same• Little deviation in brightness for all the faces within the image • Faces have to be greater than a certain size in the image so that

facial features can be detected.• Standard dimension is not more than 1600 X 1200 px.

Step 1

Page 10: Project Face Detection

Step 2

• Convert image to jjil.core.Image Where jjil means Jon's Java Imaging

Library.

Page 11: Project Face Detection

Main Image R/G/B Image 8 bit grayscale

Step 3

Page 12: Project Face Detection

Step 4

Applying Gray8DetectHaarMultiScale to our 8-bit grey image.

Page 13: Project Face Detection

Final Output

Page 14: Project Face Detection

Limititions

• It can’t detect faces without frontal view and upright orientation

Page 15: Project Face Detection

• Design an intelligence system that can analyse objects.

• Make them enable to see and feel like us. • Remove all its limitations and eager to develop this

software. • Enable them to suggest us to make the best use of

objects.• Empower media and security services.

Future Plan

Page 16: Project Face Detection

Conclusion

Thanks all