Joshi, Anshal

Profile Picture

Filters

20171
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Publication
    American Sign Language translation using edge detection and cross co-relation
    (2017-05) Joshi, Anshal; Arzuaga, Emmanuel; College of Engineering; Rivera Vega, Pedro I.; Sierra, Heidy; Department of Electrical and Computer Engineering; Alers Valentín, Hilton
    According to the World Health Organization (WHO), there are approximately 360 million people worldwide that have disabling hearing loss and 70 million that are mute. Developing communication advancements is very complex and its been a challenge for many years. Currently, American Sign Language, which is expressed through the hands and face and perceived through the eyes, is the standard language of communication for the Deaf community. However, the development of better communication mechanisms for the hearing impaired is still a big challenge. Our main objective is to implement an automated translation system which can translate the American Sign Language to English text using common computing environments such as a computer and a generic webcam. In this investigation, a real-time approach for hand gesture recognition system is presented. Two di erent approaches are used to translate English letters and words. In the method to recognize letters, rst, the hand gesture is extracted from the main image by the image segmentation, morphological operation and edge detection technique and then processed to feature extraction stage. And for the words, a video sequence is captured then divided into frames and process them for the frame selection stage. In frame selection stage, frames are sampled and selected for feature extraction and then the gesture is extracted from all of the frames by the same using the same technique as image segmentation, morphological operation, edge detection technique and combined by Montage. In feature extraction stage the Cross-correlation coe cient is applied on the gesture to recognize it. In the result part, the proposed approach is applied on American Sign Language (ASL) database and we are able to achieve 92 - 94% accuracy in translation.