talking glasses [final]

40

Upload: kariem-muhammed

Post on 29-Jul-2015

415 views

Category:

Technology


0 download

TRANSCRIPT

• Problem Definitiono Theoretical Areas of Problemo Project Domaino Specific Problem definition

• Motivations• Project Objective• Survey

o Related workso Asking Experts

• Project Architectureo Software Architecture

• Challenges• Time plan• Tools • References

Computer Vision

Machine Learning

Object Classification

Stereo Vision

The primary motivation for our work is helping blind people

Interested

In Computer Vision

Interested

In Machine Learning

Using stereo vision techniques and narration to develop an object recognition system that help blind to independently navigate the surrounding.

Developing object classification system that may be useful for a large scale of applications.

Surveillance systems

Medical systems

Improve segmentation accuracy using the depth information.

“Object Categorization by Learned Universal Visual Dictionary” by Microsoft

Research

• “This field of research is very interesting, but it needs a lot of work to reach to the goals.” Prof. Dr. Mostafa Gadal-Haqq

• “I think your project is somewhat ambitious for a senior project.” Prof. David G. Stork

Classification

Classification

Feature Extraction

Post Processing

Data Acquisition

(Stereo Camera)

Depth Estimation

Segmentation

Narrator(Text To Speech

Engine)

• Complex scene• Segmentation problems [supposed to be enhanced with

stereo vision]• Recognition reliability

– Prevent misclassifications

• Invariance with respect to– Scale– Illumination– Rotation

• Background clutter– Especially natural images don’t show only the object, but also

contain background information

• Partial occlusion [supposed to be enhanced by selecting discriminative features to describe objects]

Complex Scene

Segmentation

Partial Occlusion

Languages• C++Software• Visual Studio 2010 Professional Edition• OpenCV library

Hardware• Stereo Camera

• Books– Pattern Classification (2nd Edition) by Richard

O. Duda,Peter E. Hart,David G. Stork.– Digital Image Processing 3rd Ed by Gonzalez

and Woods 

• Papers