Please use this identifier to cite or link to this item:
http://repository.kln.ac.lk/handle/123456789/25407
Title: | A Comprehensive Review on Vision-based Sign Language Detection and Recognition |
Authors: | Weerasinghe, R. L. Ganegoda, G. U. |
Keywords: | ANN, computer vision, gesture recognition, image processing, sign language |
Issue Date: | 2022 |
Publisher: | Department of Industrial Management, Faculty of Science, University of Kelaniya Sri Lanka |
Citation: | Weerasinghe R. L.; Ganegoda G. U. (2022), A Comprehensive Review on Vision-based Sign Language Detection and Recognition, International Research Conference on Smart Computing and Systems Engineering (SCSE 2022), Department of Industrial Management, Faculty of Science, University of Kelaniya Sri Lanka. 88-95. |
Abstract: | Deaf or hard-hearing people's primary mode of communication is sign language. Communication between hard hearing and hearing people is greatly aided by sign language recognition technologies. With the advent of technology, many approaches proposed for sign language recognition. Among them, vision-based approaches are more convenient than sensor-based approaches. Vision-based approaches are involved five different stages where various techniques and algorithms are utilized in various approaches. The accuracy of the recognition is based on the techniques used and the quality of the input. Background invariance and lighting conditions highly affect the accuracy of the result. Simply by increasing the quality of the input, each and every method can approach a high accuracy rate. This paper provides a comprehensive introduction and comparison of the existing vision-based sign language detection and recognition approaches. |
URI: | http://repository.kln.ac.lk/handle/123456789/25407 |
Appears in Collections: | Smart Computing and Systems Engineering - 2022 (SCSE 2022) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
SCSE 2022 14.pdf | 13.76 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.