Ppt final
•Download as PPTX, PDF•
3 likes•1,306 views
This document describes a sign recognition system to translate sign language gestures from deaf and dumb people into audible speech. The system uses Harris corner detection and SIFT to extract features from captured images of gestures. K-nearest neighbors algorithm is used to match features to a database of images and identify the sign. A serial board communication kit then converts the text result to speech using a text-to-speech module and speaker. Examples show the system correctly identifying the signs for letters A, B, and C from captured images.
Report
Share
Ppt final
- 1. 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
- 2. 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.
- 3. 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.
- 4. LITERATURE SURVEY 1 Title Sign language recognition system Author Er.Aditi Kalsh, Dr.N.S.Garewal Years June 2013 Journal International Journal of Engineering Research Explanation Canny edge and peak detection is used for feature extraction.
- 5. 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
- 6. LITERATURE SURVEY 3 Title SignPro-An Application Suite for Deaf and Dumb Author Ashish Sethi, Hemanth, Kuldeep kumar Years May 2012 Journal International Journal of Computer Science Engineering Technology Explanation Skin segmentation and region growth approach is used for image recognition
- 7. Tools Required SOFTWARE MATLAB HARDWARE Laptop with external web camera serial board communication kit
- 8. BLOCK DIAGRAM Image acquisition Harris corner detection Histogram SIFT K-NN Algorithm Text to speech
- 9. 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
- 10. 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
- 11. SIFT Orientation assignment Key point localization Scale space Extrema detection
- 12. 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.
- 13. 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.
- 14. Captured image of sign ‘A’
- 15. Captured image of sign ‘B’
- 16. Captured image of sign ‘C’
- 17. Database images in the folder A ,D LETTER A LETTER D
- 18. database images in the folder B,C • LETTER B
- 19. Text Result
- 20. THANK YOU