High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Kinect Sensor Based Gesture Recognition for Dumb People with Voice Extraction

Author(s):

Sapana Hanamantu Jindam , SVERI College of Engineering, Pandharpur.; Babita Bira Kolekar, SVERI College of Engineering, Pandharpur; Archana Ajinath Pawar, SVERI College of Engineering, Pandharpur; Lalita A. Palange, SVERI College of Engineering, Pandharpur; Prof. Arshad Qureshi

Keywords:

Image Processing, Kinect Sensor, Skeletal Image

Abstract

Nowadays, we mostly hear the word new Technology. This word advancing the old technologies along with some extra features. This growing towards new discoveries and inventions in the field of science and technology but unfortunately, there are very fewer innovations for physically challenged people. Who face difficulties in communicating with normal people. As Disable people use sign language as their Prime medium for communication. Mostly, sign languages are not understood by the common people. Research says that many studies have been done to eliminate such kind of communication barrier. And so, many research works have been proposed to remove this communication gap between normal people and physically impaired people. So, In this Paper, we are introducing one idea which will remove the complexity in earlier proposed methods such as, Microcontrollers, Aurdino, flex sensor based hand gloves. We are advancing this process in our study by using a Kinect sensor. Kinect is a motion sensing input devices produced by Microsoft. Many applications out of which motion sensing and 3-D image processing is the main requirement for our work to detect the signs, gestures, and movements of hands or bodies. Our work starts from capturing an image of the body to convert into the skeletal image and from image processing to feature extraction of the detected image hence getting an output along with voice. The experimental results of our Presented algorithm are also very good with an accuracy of 90%.

Other Details

Paper ID: IJSRDV7I20278
Published in: Volume : 7, Issue : 2
Publication Date: 01/05/2019
Page(s): 317-320

Article Preview

Download Article