Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 11 Issue: 04 | Apr 2024

p-ISSN: 2395-0072

www.irjet.net

KANNADA SIGN LANGUAGE RECOGNITION USING HAND GESTURES Mr. Hemanth C 1, Ms. Jyothi M2, Mr. L Adarsh3, Ms. Shashikala H M4, Ms. Thejaswini N5 1 Assistant Professor, Dept. of Computer Science and Engineering, Maharaja Institute of Technology,

Thandavapura, Karnataka, India

2-5Students, Dept of Computer Science and Engineering, Maharaja Institute of Technology,

Thandavapura, Karnataka, India ---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Sign language serves as a crucial mode of communication for individuals with hearing impairments, facilitating their interaction and integration into society. Recognizing the gestures and movements of Learning sign language is difficult because of the complexity of sign language. In this paper, we will introduce a new method of Kannada sign language recognition using machine learning. The proposed system is employs a comprehensive dataset of Kannada sign language gestures, capturing a diverse range of hand configurations and movements. Preprocessing techniques are applied to enhance the quality and consistency data, including standardization and feature extraction. Machine learning algorithms, including CNNs, are used to train and classification tasks. CNNs are utilized for spatial feature extraction from image-based data. The system achieves high classification accuracy rates, thus showing promise for practical deployment in real-world applications. Furthermore, the system's adaptability and scalability make it suitable for accommodating additional gestures and improving performance over time. This research makes strides in enhancing accessibility and inclusivity for people with hearing impairments by offering an effective and dependable method. It employs machine learning techniques to aid in the recognition of Kannada sign language.

enabling the creation of robust systems with high accuracy in recognizing KSL gestures. TensorFlow, an open-source library, is notable for its flexibility and scalability in implementing deep learning architectures, particularly convolutional neural networks (CNNs), essential for understanding sequential and spatial patterns inherent in KSL. 1.1 PROBLEM DEFINITION Traditional Kannada sign language recognition relies on hand recognition, aiding communication between hearing and deaf individuals. Our project innovatively combines sign and behavioral signals, improving communication clarity and security. Using multiple instance learning algorithms, we attain higher accuracy in understanding signs, facilitating effective communication with deaf and mute individuals, conveying letters, words, or intentions. Our software recognizes hand gestures, computing parameters to interpret human communication accurately, determining the individual's communication state through static gestures analysis. 1.2 OBJECTIVE The main objective of our project is to ensure the communication experience as complete a possible for both hearing and deaf people. The work presented in Indian Regional language, Kannada, the system to develop system’s for automatic translation and static gestures of alphabets in Kannada sign language. Signs of the deaf individual can be recognized and translated in Kannada language for the benefit of deaf & dumb people.

Key Words: Sign Language, CNN, Gesture, Machine Learning, Hand, Movements

1. INTRODUCTION The introduction outlines the imperative for systems capable of recognizing Kannada Sign Language (KSL) gestures, highlighting the existing gap in communication between KSL users and non-signers. KSL serves as a crucial mode of communication for the Deaf community in Karnataka, India, reflecting the region's linguistic and cultural heritage. However, developing accurate systems for recognizing KSL gestures poses a significant technological challenge. Integrating machine learning and deep learning, particularly through frameworks like TensorFlow and Media Pipe, offers a promising solution to bridge this communication gap. The complexity of KSL lies in its diverse gestures, involving variations in hand shapes, movements, and expressions, requiring advanced computational methods for real-time interpretation. Machine learning, combined with deep learning models, has emerged as a transformative approach,

Impact Factor value: 8.226

1.3 SCOPE The Communication forms a very important and basic aspect of our lives. Whatever we do, whatever we say, somehow does reflect some of our communication, though it want be directly. To understand the very fundamental behavior of a human, we will analyze this communication through some hand gesture, also called, the affect data. Data can be sign, image etc. Using this communicational data for recognizing the gesture also forms an interdisciplinary field, called Affective Computing. This paper summarizes the previous works done in gesture recognition based on various sign models and computational approaches.

ISO 9001:2008 Certified Journal

Page 1273