You are on page 1of 19

A

PRESENTATION OF PROJECT SYNOPSIS OF B.E. 7TH SEMESTER

Under the guidance of : Asst.Prof.Anish Lazrus

Presented by1

S Vijaylaxmi Shripritha Gupta Jagrati Shrivastava


5/2/2012 5:01 PM

OUTLINE OF PRESENTATION
Objective Motivation Challenges Hardware and software requirements Introduction Proposed methodology Flow diagram Input digit image Pre process Feature extraction Multi feature classification Expected outcome Future scope References

5/2/2012 5:01 PM

5/2/2012 5:01 PM

OBJECTIVE

This is an digit recognition application. We draw a digit in the picture box and the module recognizes the digit and displays the result.

MOTIVATION
Digit recognition module finds itself applied in the following areas:
o

5/2/2012 5:01 PM

ZIP code recognition.

Lottery ticket recognition.


Cards used for e-payment.

CHALLENGES

5/2/2012 5:01 PM

Every individual has his/her unique handwriting. No imaging system has 100% recognition accuracy.

Actual performance dependent on environment.

HARDWARE & SOFTWARE REQUIREMENTS


HARDWARE:

5/2/2012 5:01 PM

256 MB RAM. Windows 2000/XP. Intel N atom processor.

SOFTWARE:
o o

VB.NET. Database used: notepad.


6

5/2/2012 5:01 PM

INTRODUCTION
This is an digit recognition application. First we draw a digit in the picture box. Then the image processing begins and recognize the digit and returns the result. Database includes handwritten digits which are collected from many people. These digit images converted to binary type before added to the database.

PAPER-01:
5/2/2012 5:01 PM

Rafael M. O. Cruz, George D. C. Cavalcanti and Tsang Ing RenCenter of Informatics, Federal University of Pernambuco Recife, Brazil

It is herein proposed a handwritten digit recognition system which uses multiple feature extraction methods and classifier ensemble. The combination of the feature extraction methods is motivated by the observation that different feature extraction algorithms have a better discriminative power for some types of digits.

PROPOSED METHODOLOGY

5/2/2012 5:01 PM

Some of the misclassified digits are ambiguous either by segmentation problems, peculiar writing style or distortions. A strategy to reject this ambiguous digits and improvements to recognize the misclassified digits that can be easily recognized by humans are being studied. Our approach is based on the following block diagram.

PROPOSED FLOW DIAGRAM OF OUR


WORK
Input digit image Feature extraction
5/2/2012 5:01 PM

Pre process

Multi feature classification

10

INPUT DIGIT IMAGE


5/2/2012 5:01 PM

We have used a Bitmap function which allows the user to input a handwritten digit.

11

PRE-PROCESS
5/2/2012 5:01 PM

Digit is written with black digital ink. It searches from left to right for black pixels starting from left top corner of the area specified for writing. A trace of a black pixel is the indication of the presence of a character. The continuous absence of black pixels black could be a gap between two rows. To make sure whether it is a gap, algorithm searches from left to right against every white pixel, if there is no trace of white pixel for the entire row, the gap is confirmed.

12

oTHINNING
5/2/2012 5:01 PM

The digit images are thinned to a thickness of one pixel.

This can lead undesired side-effects ,such as spurs.

(a) threshold original

(b) thinned (with spurs)

(c) Pruned (by 5 pixels)


13

FEATURE EXTRACTION
5/2/2012 5:01 PM

In the feature extraction module, two feature sets are extracted from each digit image: 1. Numbers of ending point, bifurcation, cross and their location, 2. Number of closed curve.
Number of Ending point Possible digits 0 1 2 3 4 >=5

08

69

123 457

234 5

reject ed
14

MULTI FEATURE CLASSIFICATION


5/2/2012 5:01 PM

In this phase multi-features such as orientation information, likeness degree are considered to recognize the digit. The digit which has the highest percentage of similarity is taken as the final recognized digit.

15

5/2/2012 5:01 PM

EXPECTED OUTCOMES

The main goal is purely educational one, a moderate recognition rate of 95% is expected. We will try to recognize handwritten digits of different people.

16

5/2/2012 5:01 PM

FUTURE SCOPE

Future work will focus on better selection of multifeatures and improve the system performance. Multiple digits can be recognized simultaneously.

17

5/2/2012 5:01 PM

REFERENCES

Lottery Digit Recognition Based on Multi-featuresDecong Yu, Lihong Ma, Member, IEEE and Hanqing Lu, Senior Member, IEEE. On-line Handwritten Character RecognitionMuhammad Faisal Zafar, Dzulkifli Mohamad, and Razib M. Othman Handwritten Digit Recognition Using Multipl Feature Extraction Techniques and Classifier EnsembleRafael M. O. Cruz, George D. C. Cavalcanti and Tsang Ing Ren Center of Informatics, Federal University of Pernambuco

18

5/2/2012 5:01 PM

Thank You
Any Suggestion please

19

You might also like