Ppt final

SIGN RECOGNITION System FOR DEAF AND DUMB

PEOPLE

GUIDED BY PRESENTED BY

Ms.I.KANAGA SANGEETHA M.E

ASSISTANT PROFESSOR M.RINCY JULIET

DEPARTMENT OF ECE (98010106052)

A.SUBACHITHIRAI MEENAL

(98010106068)

M.SURIYA

(98010106072)

Department of ECE

OBJECTIVE

To develop a system for automatic translation of static gestures used by dumb and deaf persons into audible speech understandable by a common person.

INTRODUCTION

• Sign recognition system act as a interpreter between common people and socially aided people.

• It captures the gesture shown by the deaf and dumb people and convert immediately into known language.

LITERATURE SURVEY 1

Title Sign language recognition system

Author Er.Aditi Kalsh, Dr.N.S.Garewal

Years June 2013

JournalInternational Journal of EngineeringResearch

ExplanationCanny edge and peak detection is used for feature extraction.

LITERATURE SURVEY 2

Title Real Time Indian Sign Language Recognition System to aid Deaf and Dumb people

Author Rajam, P. Subha and Dr G Bala krishnan

Years June 2011

Journal 13th International Journal Communication Technology(ICCT)

Explanation Cam shift and HSV approach for recognizing gesture

LITERATURE SURVEY 3

Title SignPro-An Application Suite for Deaf and Dumb

Author Ashish Sethi, Hemanth, Kuldeep kumar

YearsMay 2012

JournalInternational Journal of Computer Science Engineering Technology

ExplanationSkin segmentation and region growth approach is used for image recognition

Tools RequiredSOFTWARE MATLAB

HARDWARE Laptop with external web camera

serial board communication kit

BLOCK DIAGRAMImage acquisition

Harris corner detection

Harris Corner Detection

• In the harries corner detection we calculate the cim , r, c.

• Cim - binary image marking corners. • r - Row coordinates of corner points. • c - Column coordinates of corner points

Histogram

• Histograms can be used to provide useful image statistics Information • It plots the number of pixels for each

value• Histogram we are using the num-bins

function

SIFT

K-NN Algorithm• This approach is used to match the key points

calculated from the capture image using the harries and sift.

• Now the key points matched with the database image.

• IDX = knn search searches the reference data set to find the k-nearest neighbours of each query point represented by each row.

Text to Speech• Serial board communication kit is used to

convert Text to Speech.• This Kit contains PIC Microcontroller,

Selector, Voice recorder, Amplifier, Speaker, Step down transformer.

Captured image of sign ‘A’

Captured image of sign ‘B’

Captured image of sign ‘C’

Database images in the folder A ,D

LETTER A

LETTER D

database images in the folder B,C

• LETTER B

Text Result

THANK YOU