# Image-based Indian Sign Language Recognition: A Practical Review using Deep Neural Networks

**Authors:** Mallikharjuna Rao K, Harleen Kaur, Sanjam Kaur Bedi, M A Lekhana  
**arXiv:** [2304.14710](https://arxiv.org/abs/2304.14710)  
**Date:** 2023-04-28  
**Summary page:** https://www.summarizepaper.com/en/paper/2304.14710v1/

## Key points

- People with vocal and hearing disabilities face communication difficulties
- Sign language is often used to overcome these difficulties
- Most people in the general population do not understand sign language, creating a barrier
- This issue is particularly evident in places like banks, airports, and supermarkets
- A sign language recognition (SLR) system is necessary to address this problem
- Previous research has focused on American Sign Language (ASL)
- There is a need for similar systems that cater to the needs of the deaf and hard-of-hearing community in India
- The study proposes an Indian Sign Language (ISL) recognition system
- The objective of the model is to develop a real-time word-level sign language recognition system that can translate sign language into text
- Computer vision techniques such as grayscale conversion, dilatation, and masking are used for image processing
- Images captured through a web camera are fed into a convolutional neural network (CNN) for recognition
- The best model achieves an accuracy rate of 99%
- This research contributes to bridging the communication gap faced by individuals with hearing disabilities in India
- The study highlights the importance of SLR systems in facilitating communication for individuals with vocal and hearing disabilities
- The practical solution developed can be implemented in various settings to improve accessibility and inclusivity for the deaf and hard-of-hearing community in India.

## Summary

People with vocal and hearing disabilities face communication difficulties, which they often overcome by using sign language. However, this creates a barrier as most people in the general population do not understand sign language. This problem is particularly evident in places like banks, airports, and supermarkets. To address this issue, a sign language recognition (SLR) system is necessary. Previous research has focused on American Sign Language (ASL), but there is a need for similar systems that cater to the needs of the deaf and hard-of-hearing community in India. In this study titled "Image-based Indian Sign Language Recognition: A Practical Review using Deep Neural Networks," authors Mallikharjuna Rao K, Harleen Kaur, Sanjam Kaur Bedi, and M A Lekhana propose an Indian Sign Language (ISL) recognition system. The main objective of their model is to develop a real-time word-level sign language recognition system that can translate sign language into text. The researchers utilize computer vision techniques such as grayscale conversion, dilatation, and masking to process images captured through a web camera. These images are then fed into a convolutional neural network (CNN) for recognition. The authors report that their best model achieves an impressive accuracy rate of 99%. This research contributes to bridging the communication gap faced by individuals with hearing disabilities in India by providing an effective ISL-based sign language recognition system. Overall, this study highlights the importance of SLR systems in facilitating communication for individuals with vocal and hearing disabilities. By focusing on ISL specifically for the Indian context, the researchers have developed a practical solution that can be implemented in various settings to improve accessibility and inclusivity for the deaf and hard-of-hearing community in India.