Interact With Computer System Using Real-Time Static Hand Gesture Recognition

Abstract

Hand gestures are powerful approach of communication among humans and the sign language is the most effective and natural communication for the peoples. In this work, the hand gesture system is developed. The setup of the proposed system needs fixed position webcam with 20 mega pixel resolution. It captures snapshot using RGB colour space from static distance. This work is divided in different stages, which are image pre-processing, the region extraction after that feature extraction & last feature matching. First stage converts captured input image into binary image using gray threshold, followed by morphological operations. At Second stage, blob extracts hand region and crop is applied for getting extracted region and then “Sobel” edge detection is applied on the above extracted region. Third stage gives feature vectors as area of edge, which will be compared with feature vectors of a dataset by using Euclidian distance in the further stage. Minimum Euclidian distance gives accurate recognition of perfect matching gesture for performing corresponding activity. This paper have experiments for static hand gestures related to all 26 alphabets. The Training dataset having number of samples of each symbol in different shapes, different positions and conditions of environment. It can reliably recognize single hand gestures in real time. It can achieve upto 82.88% of recognition rate in static background with least Euclidian distance.

Authors and Affiliations

Ajit R. Waingankar, Sanket S. Gurav, Abhishek A. Naik, Rohit S. Gangan

Keywords

Related Articles

FEA analysis of notched adherent joint between metal and composite object using ANSYS CZM technique

Traditionally in engineering, structural joining has been synonymous to riveting, bolting and other purely mechanical fastening together with welding or soldering in the case of metallic construction materials. Up until...

16 BIT UNSIGNED MULTIPLIER USING PROPOSED CSLA

This project deals with the comparison of the VLSI design of the conventional carry select adder (CSLA) based 16-bit unsigned integer multiplier and the VLSI design of the proposed carry select adder (CSLA) based 16-bit...

Review of Developments in Pin-On-Disc Tribometer for Environment Control

The design of earlier pin-on-disc tribometer mainly focuses on normal pressure and environmental conditions and aim to measure the effect of load and speed on material performance. The effect of gas pressure and environ...

Fuzzy Logic Controller Based Speed Analysis and Control of DC Shunt Motor

In this paper, various speed control techniques for a dc shunt motor namely conventional PID controller and fuzzy logic controller, using a PID controller and Fuzzy Logic Controller have been discussed and a comparative...

Temperature Sensor Using SCADA

Automatic temperature control is first step in applications such as industries. The main aim of temperature sensor using SCADA is to sense the temperature and perform the required performance using temperature sensors....

Download PDF file
  • EP ID EP22058
  • DOI -
  • Views 158
  • Downloads 5

How To Cite

Ajit R. Waingankar, Sanket S. Gurav, Abhishek A. Naik, Rohit S. Gangan (2016). Interact With Computer System Using Real-Time Static Hand Gesture Recognition. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 4(4), -. https://europub.co.uk/articles/-A-22058