Vision Based Deep Learning Approach for Dynamic Indian Sign Language Recognition in Healthcare

Uchil, Aditya P.; Jha, Smriti; Sudha, B G

doi:10.1007/978-3-030-37218-7_43

Cited by 7 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Uchil et al [90] suggested a novel dimension for recognizing the dynamic gesture by OpenPose and the hybrid approach with CNN. They eliminated the use of cameras and sensors by providing a way in Colab to identify the dynamic sign.…”

Section: A Hybrid Models Using Deep Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Detail Study of Sign Language Communication for Deaf-Mute to Normal Person

Pareshbhai Mansangbhai,

B. Vaghela

2024

IJCDS

View full text Add to dashboard Cite

Even though sign language recognition systems are crucial, there isn't a well-structured literature study to help Deaf-Mute. Approximately 160 research articles were found and reviewed in this survey of the sign recognition framework. From this pool, around 100 research articles were chosen, analyzed, and categorized. Each of these articles was grouped according to several criteria, such as datasets, kind of sign language, data collection methods, preprocessing, segmentation, and learning strategy. Notably, the largest study on sign language recognition utilized cameras and focused on steady, separated, and one-handed signs. Researchers are receiving increasing attention these days for creating cost-effective sign language recognition. According to the review, the main reason why commercialized sign language hasn't yet been developed as the current methods for identifying sign language either require expensive equipment or take too much computational time. The objective of this article is to provide an overview of recent developments in the field of computer vision, enabling other researchers to gain insights and potentially develop more efficient sign language systems.

show abstract

Section: A Hybrid Models Using Deep Learning Methodsmentioning

confidence: 99%

“…With the help of ISLRTC YouTube and RKMVERIthe campus recognized a few healthcare-related keywords and prepared the dataset. They claimed 85% accuracy with their hybrid approach of CNN-RNN [90]. RNN (Recurrent Neural Network) provides more memory with the limitation of computational complexity.…”

Section: A Hybrid Models Using Deep Learning Methodsmentioning

confidence: 99%

A Detail Study of Sign Language Communication for Deaf-Mute to Normal Person

Pareshbhai Mansangbhai,

B. Vaghela

2024

IJCDS

View full text Add to dashboard Cite

show abstract

“…Uchil et al [47] proposed designing a system for deaf people to maintain their health. They recorded videos of 20 words in Hindi Sign Language and translated them using a CNN.…”

Section: Using Convolutional Neural Network To Translate Gesturesmentioning

confidence: 99%

A Novel Assistive Glove to Convert Arabic Sign Language into Speech

Alzubaidi

Otoom

Rwaq

2023

ACM Trans. Asian Low-Resour. Lang. Inf. Process.

View full text Add to dashboard Cite

People with speech disorders often communicate through special gestures, and sign language gestures. However, other people around them might not understand the meaning of those gestures. The research described in this paper is aimed at providing an assistive device to help those people communicate with others, by translating their gestures into a spoken voice that others can understand. The proposed device includes an electronic glove that is worn on the hand. It employs an MPU6050 accelerometer/gyro with 6 degrees of freedom to continuously monitor hand orientation and movement, plus a potentiometer for each finger, to monitor changes in finger posture. The signals from the MPU6050 and the potentiometers are routed to an Arduino board, where they are processed to determine the meaning of each gesture, which is then voiced using the audio streams stored in an SD memory card. The audio output drives a speaker, allowing the listener to understand the meaning of each gesture. We built a database with the help of 10 deaf people who cannot speak. We asked them to wear the glove while performing a set of 40 Arabic sign language words, and recorded the resulting data stream from the glove. That data was then used to train seven different learning algorithms. The results showed that the Decision Tree learning algorithm achieved the highest accuracy of 98%. A usability study was then, conducted to determine the usefulness of the assistive device in real-time.

show abstract

“…Keskin et al [26] and Vidalón et al [27] proposed Hidden Markov Model (HMM) based recognition system for medical terms of Turkish and Brazilian sign language respectively. Uchil et al [28] represented an RNN based approach to recognize healthcarerelated words of the Indian sign language dictionary. Starner et al [29] developed an HMM-based system for recognizing sentence-level ASL.…”

Section: Sn Computer Sciencementioning

confidence: 99%

Design and Development of a Humanoid Robot for Sign Language Interpretation

et al. 2021

View full text Add to dashboard Cite

Purpose With an indisputable complexity of communication for hearing and speaking impaired people, most sign language recognition systems utilize virtual reality or onscreen robots. This paper presents the design and development of a special and low-cost humanoid robot that can perform as a sign language interpreter. To the best of our knowledge, this is the first endeavor to fabricate a humanoid robot for Bangla sign language (BdSL) and Medical signs interpretation. Methods Considering the plethora of design criteria and balancing between rigidity and flexibility 3D models of the robotics parts are designed and 3D printed ensuring cost efficiency. With the help of modern fabrication technology, the robot is developed and assembled with proper actuators and circuitry. An image dataset is built comprising 950 images for BdSL recognition and made publicly available. We utilized the Recurrent neural network (RNN) and Convolutional neural network (CNN) for deep learning model establishment and feature extraction from video and image data. Results The developed humanoid robot has 43 Degrees of freedom (DoF) which includes two 15 DoF hands. It can imitate 16 BdSL alphabets in sign language, can capture a video or image input in real-time from the user, and recognize 10 medical signs and 38 alphabets of BdSL. The learning model for video-based medical sign recognition achieved 87.5% test accuracy. Image-based Bangla sign language recognition achieved an overall test accuracy of 98.19% in our dataset and 93.8% in another available dataset. Conclusion Compared to the state-of-the-art robotic systems for sign language interpretation, our approach has achieved higher kinematic characteristics, remarkable results in sign recognition, and impressive competency in sign imitation; all at almost 10 times lower cost than the state-of-the-art systems. The results are evidence that our approach is efficient and suitable in helping hearing and speaking impaired people. Moreover, this work initiates a research scope that can be further extended for creating equal opportunities for the hearing and speaking impaired community.

show abstract

Vision Based Deep Learning Approach for Dynamic Indian Sign Language Recognition in Healthcare

Cited by 7 publications

References 10 publications

A Detail Study of Sign Language Communication for Deaf-Mute to Normal Person

A Detail Study of Sign Language Communication for Deaf-Mute to Normal Person

A Novel Assistive Glove to Convert Arabic Sign Language into Speech

Design and Development of a Humanoid Robot for Sign Language Interpretation

Contact Info

Product

Resources

About