Sign Language recognition using Machine learning

DOI

https://www.doi.org/10.59256/ijire.20260702019

Published In

Volume 7, Issue 2 (2026)

Published Online

March-April 2026

Pages

145-151

Article Type

Original Article

Check for updates

Sign language recognition is very important for establishing good communication between hearing-impaired people and machines. This project is intended to classify the use of hand gestures for communication with robots, using the machine learning techniques. Using Media Pipe for precise hand detection and key point detection, a powerful dataset of sign language gestures is ready for model training. To increase the performance of classification, specific learning methods, AdaBoost and Gradient Boosting are implemented. A combination of multiple weak learners is employed through these techniques to create a strong predictive model, that can separate complex patterns of gestures with better accuracy and generalization. OpenCV makes it easy to grab video in real time and process the data, enabling the system to interpret gestures in real time and feed back to the user. This integrated approach adds to the robot's capacity to comprehend and react to human gestures, driving more intuitive and inclusive human-robot interactions. The use of ensemble-based learning in gesture recognition is a scalable, reliable solution for the translation of sign language into machine-understandable commands, driving advances in accessibility and automation in robotic communication.