International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 07 | July 2023
p-ISSN: 2395-0072
www.irjet.net
Real Time Sign Language Translation Using Tensor Flow Object Detection John Pius Thayiparampil1, Kiran k2, T Binto Binu3, Sreejith Viju4 BTECH UG Students ,Department of Computer Science and Engineering ,TOMS college of Engineering
1-4
APJ Abdul Kalam Technological University, Kerala ,India ---------------------------------------------------------------------***--------------------------------------------------------------------TensorFlow Object identification, a potent framework for Abstract - Deaf and mute people frequently use sign
object identification tasks, will be used in this project to create a sign language translator. We can close the communication gap between the hearing-impaired community and the rest of society by utilizing the capabilities of deep learning and image processing. For people who have hearing impairments, sign language is an essential form of communication. The majority of people, however, lack the abilities needed to comprehend and interpret sign language. Effective communication and inclusion for the deaf and hard of hearing are severely hampered as a result. To solve this problem, we suggest creating a sign language translator that makes use of TensorFlow Object Detection, a cutting-edge tool for instantly identifying and classifying items.
language to exchange information. In India, there are 63 million people who have severe hearing loss. At least 76–89 .Due to their limitations, people with speech and hearing impairments frequently have trouble communicating. Translation systems are required to help bridge the gap in under- standing that exists between the hearing and speechimpaired groups and the general population. So introducing a real-time translates sign language is one solution to this issue. It will be able to translate all of the varied gestures by gathering and analyzing motion data. Sign language can be read by humans in every- day culture. Using TensorFlow object identification, a novel real-time sign language translation has been developed that can translate sign language based on hand movements and is understandable to both average individuals and those with major hearing and speech impairments. The Software will detect gesticulation.
1.2 GESTURE RECOGNITION Gesture recognition acts as a translator between computers and people and enables computers to understand human behaviors. This would enable natural humancomputer interaction without putting any extra mechanical components in direct contact with users. The sign language used by the deaf and dumb community is gesture-based. When it was difficult to transmit audio or when typing and writing were difficult but there was still the possibility of vision, this group relied on sign language for communication.
Key Words: Real Time translation, Sign Language, Tensor Flow , Object Detection , Gesture detection, Hearing and speech Impairment.
1.INTRODUCTION Motion detection tracks shifts in an object's position with relation to its surroundings, and vice versa. We can identify moving objects in front of the camera with the aid of this motion detection program. This software can be used to carry out the following activities, among others: [1] Take assist for education purpose for deaf and muted students and teachers; [2] helps in meeting and conference; [3] helps to get effective treatment in medical sector ; [4] for more effective and more involving communication. A very good way to lessen the communication gap. This project report's main goal is to give a thorough description of the Sign Language Translator system that was created as part of Real time sign language translator. Through the use of this technique, people who are deaf or hard of hearing and non-sign language users can communicate more effectively. This initiative aims to improve inclusiveness for the deaf population and allow effective communication by utilizing cutting-edge technology and artificial intelligence. Computer vision and machine learning have made considerable strides in recent years, enabling improvements in a number of areas, including the translation and recognition of sign languages.
© 2023, IRJET
|
Impact Factor value: 8.226
Fig -1: Gesture Recognition
|
ISO 9001:2008 Certified Journal
|
Page 907