An American Sign Language Recognition System using Bounding Box and Palm FEATURES Extraction Techniques
S. Shivashankara1, S. Srinath2
1S. Shivashankara, Research Scholar, Department of Computer Science and Engineering, Sri Jayachamarajendra College of Engineering, Mysuru (Karnataka), India.
2S. Srinath, Department of Computer Science and Engineering, Sri Jayachamarajendra College of Engineering, Mysuru (Karnataka), India.
Manuscript received on 17 December 2018 | Revised Manuscript received on 28 December 2018 | Manuscript Published on 09 January 2019 | PP: 492-505 | Volume-7 Issue-4S November 2018 | Retrieval Number: E2077017519/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The sign language is absolutely an ocular interaction linguistic over and done with its built-in grammar, be nothing like basically from that of spoken languages. This research paper presents, an inventive context, whose key aim is to achieve the transmutation of 24 static gestures of American Sign Language alphabets into human or machine identifiable manuscript of English language. The gestures sets considered for cognition and recognition process are purely invariant to location, Background, Background color, illumination, angle, distance, time, and also camera resolution in nature. The gesture recognition process is carried out after clear segmentation and preprocessing stages. As an outcome, this paper yields an average recognition rate of 98.21%, which is an outstanding accuracy comparing to state of art techniques.
Keywords: American Sign Language, Bounding Box Technique, Canny Edge Detector, CIE Color Model, Gesture Recognition.
Scope of the Article: Pattern Recognition