Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/221732
Title: Framework of cost effective hand gesture recognition system
Researcher: B P Pradeep Kumar
Guide(s): Manjunatha M B
Keywords: Engineering and Technology, Hand Gesture, Image processing, Cost effectiveness
University: Jain University
Completed Date: 13/10/2018
Abstract: A Human to Human interaction can be achieved through speech, writing, expressions and gestures. But to interact with a computer a standard interface is generally used. The advancement in the technology has allowed thoughts to improvise the interface for interaction between the human and computer. Human Computer Interaction (HCI) is a research area which defines innovative interfaces for human computer interaction. The recent works in this area have focused on the use of human gestures for computer interaction, especially Hand gestures. The Hand Gestures allows interaction in a virtual reality environment and also can provide a convenient way for physically disabled, such as blind people to interact with a computer. There are languages for defining the meaning of a hand Gesture pattern such as American standard language, British standard language, Japanese standard language etc. A Hand Gesture Recognition (HGR) system allows a user to interact in a sign language with the computer. Several HGR systems have come into existence few of those using Hardware sensors such as accelerometers and gyroscope but majority of the systems developed use visual or imaging technique. newlineThe research focus here lies in the development of a visual Hand Gesture Recognition (HGR) system to establish an effective interface for computer interaction. An Extensive insight into the works related to existing visual HGR systems revealed that the HGR system performance is greatly affected by lighting variations, image quality derived from camera and background clutters. Apart from this the existing ASL HGR systems suffer from resolution inconsistencies, complex feature extraction techniques and the classification techniques used could not cope with the complex scenarios of recognition. To provide an effective solution to the above problems this research emphasizes on three main objectives; enhancing the newlinecaptured image resolution for accurate gesture detection, developing a new model for defining the features to be extracted and utilizing a classification technique which increases the recognition performance. A robust visual HGR system is proposed which is shown as framework in different sections of the thesis. The thesis presents the framework for Resolution-Enhancement, Modelling of American sign language (ASL) recognition, Hybrid framework for American Sign Language recognition and finally a Framework for performance analysis. newline
Pagination: 160 p.
URI: http://hdl.handle.net/10603/221732
Appears in Departments:Department of Electronics Engineering

Files in This Item:
File Description SizeFormat 
10_chapter 7.pdfAttached File535.33 kBAdobe PDFView/Open
01_title.pdf21.89 kBAdobe PDFView/Open
02_certificate.pdf90.01 kBAdobe PDFView/Open
03_contents.pdf449.51 kBAdobe PDFView/Open
04_chapter 1.pdf445.62 kBAdobe PDFView/Open
05_chapter 2.pdf358.44 kBAdobe PDFView/Open
06_chapter 3.pdf914.08 kBAdobe PDFView/Open
07_ chapter 4.pdf514.15 kBAdobe PDFView/Open
08_chapter 5.pdf433.27 kBAdobe PDFView/Open
09_chapter 6.pdf816.96 kBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.