International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024
p-ISSN: 2395-0072
www.irjet.net
Sign Language Recognition using Machine Learning 1Raj Botre, 2Yogiraj Pradeshi,3Prathamesh Kalunge, 4Shubham Gajare, 5 S.S.Bhosale 1,2,3,4, UG Students, Department of Computer Science And Engineering, SVERI’s College of Engineering, Pandharpur,
Maharashtra, India
5, Assistant Professor, Department of Computer Science And Engineering, SVERI’s College of Engineering, Pandharpur,
Maharashtra, India -------------------------------------------------------------------------***--------------------------------------------------------------------Abstract— Sign language recognition systems bridge the communication gap between hearing and deaf individuals by translating hand gestures into textual or spoken language. This project proposes an advanced machine learningbased framework for accurate and real-time sign language recognition [4]. The system captures gestures through a camera, preprocesses the input using image processing techniques, and extracts key features like hand orientation, position, and movement. These features are fed into a deep learning model, such as a Convolutional Neural Network (CNN), for gesture classification. For dynamic gestures, a Long Short-Term Memory (LSTM) network processes temporal dependencies, enabling seamless recognition of continuous signs and sentences. A comprehensive dataset of sign language gestures is used for training, ensuring high accuracy across diverse signs and variations. The system is designed to map recognized gestures to corresponding text or speech outputs, providing real-time translation. Customizable modules ensure compatibility with regional sign languages, enhancing its inclusivity. Interactive feedback mechanisms notify users if gestures are unclear, improving usability and reliability. The solution is optimized for low-cost hardware and mobile devices, ensuring affordability and portability. Its accessible design includes multilingual support and compatibility with assistive technologies, making it suitable for diverse user groups. The proposed system demonstrates potential to revolutionize communication for the hearing-impaired community, fostering inclusivity and enabling equal opportunities in education, employment, and social interaction. This project paves the way for practical and scalable solutions in the field of assistive technology using advanced machine learning.[6] Keywords: Sign Language, Deaf People, Machine Learning.
I.
INTRODUCTION
Sign language is a vital mode of communication for millions of people worldwide, particularly for those who are hearingimpaired or speech-disabled. It employs a combination of hand gestures, facial expressions, and body movements to convey messages effectively. However, the lack of proficiency in sign language among the general population creates a communication barrier, making it challenging for the hearing-impaired community to fully integrate into mainstream social, educational, and professional environments. [1] To address this issue, the development of a reliable, real-time sign language recognition system has become an area of growing interest in the field of assistive technologies.
advancements in machine learning (ML) and computer vision have paved the way for more efficient, softwarebased solutions that leverage deep learning to accurately interpret sign language gestures from video input. These technologies have the potential to bridge the communication gap by translating sign language into text or speech, making interaction more inclusive and accessible Sign language is a vital means of communication for millions of individuals with hearing impairments worldwide. However, there is often a communication barrier between sign language users and those who do not understand sign language. To bridge this gap, the development of automated systems for sign language recognition is crucial.[7] Traditional methods of translating sign language are labor-intensive and require manual intervention, limiting their scalability and realtime applicability. Recent advances in machine learning (ML) have opened up new possibilities for automating sign language detection with higher accuracy and efficiency.
Traditional sign language recognition systems relied on hardware-intensive solutions, such as gloves embedded with sensors, to detect hand movements. While effective to some extent, these approaches were often expensive, cumbersome, and limited in scalability. Recent
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 455