文章基本信息

标题：Real-Time Static Devnagri Sign Language Translation using Histogram
本地全文：下载
作者：Jayshree R. Pansare ; Kirti S. Rampurkar ; Pritam L. Mahamane 等
期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2013
卷号：2
期号：4
页码：1455-1459
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Sign language is nowadays widely used in hearing impaired people as communication media. It has different applications in many domains like HCI (Human Computer Interaction), Robot Control, Security, Gaming, Computer Vision, etc. Devnagri Sign Language Translation System using histogram matching algorithm is proposed in this paper; for recognizing Devnagri Sign Language (DSL) alphabets, the steps of algorithm are Image capturing, Image Pre-processing, Hand region extraction, Feature extraction and histogram matching. Image is captured in RGB color space using 8 Mega Pixel I-ball web camera mounted on top of the laptop. In image pre-processing morphological operation like blurring, noise removing is done. In region extraction stage, hand region is extracted and then edge detection is done using canny edge operator. The third stage is feature extraction; in this the histogram of cropped image is taken. The histogram obtained in this stage is compared with the histogram of image in the training dataset and similarity factor is calculated. Perfect match is obtained for highest similarity factor. In this paper hand gestures for Devnagri Sign Language (DSL) which includes 13 vowels ("swars") and 33 consonants ("Vyanjan") are taken. The result mainly depends upon illumination conditions and background texture and by considering all these possibilities our hand gesture recognition system gives up to 87.82% accuracy.
关键词：DSL; Edge oriented histogram; Region ; extraction; Swars; Vyanjan