| Autor: |
Rangdale, Sonali, Sarkarkar, Prasanna, Kadam, Shubham, Tegyalwar, Himanshu, Waghmare, Chetan, Shinde, Shreyas |
| Předmět: |
|
| Zdroj: |
Grenze International Journal of Engineering & Technology (GIJET); Jan2024, Vol. 10 Issue 1, Part 2, p1931-1937, 7p |
| Abstrakt: |
The objective of paper is to study a deep learning strategy for recognizing Indian Sign Language (ISL) using convolutional neural networks CNN. Sign language recognition has become a crucial tool for enhancing communication and accessibility for the deaf in India. ISL is a visual-gestural language used by the deaf population in India, and it is a visual-gestural language used by the deaf community in India. Gesture based communication Application is a characteristic language that involves various methods for articulation for correspondence in day-to-day existence. This paper presents a Programmed interpretation framework for token of manual letters in order in English communication via gestures. It manages pictures of uncovered hands, which permits the client to cooperate with the framework in a characteristic manner. In proposed system a deep learning model using CNN with ISL (Indian Sign Language) Dataset is used. The model consists of preprocessing the dataset to extract features from the image, then train a CNN to recognize the signs and gestures with audio. Experimental results show that the proposed approach is capable of achieving accuracy in recognizing ISL signs and gestures with accuracy of 0.868, recall 0.856, precision 0.863, and F-score is 0.884 respectively. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
