American Sign Language Alphabet Recognition by Extracting Feature from Hand Pose Estimation
| Autor: | Azmain Yakin Srizon, Md. Al Mehedi Hasan, Jungpil Shin, Akitaka Matsuoka |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
American Sign Language
Computer science ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION TP1-1185 Sign language Biochemistry Article Analytical Chemistry distance-based features Fingers angle-based features Sign Language Feature (machine learning) Humans media-pipe support vector machine Electrical and Electronic Engineering Instrumentation Pose finger spelling a dataset business.industry Character (computing) massey dataset Chemical technology Pattern recognition Recognition Psychology Hand Atomic and Molecular Physics and Optics language.human_language United States Support vector machine light gradient boosting machine language american sign language recognition Artificial intelligence Gradient boosting business Algorithms Sign (mathematics) |
| Zdroj: | Sensors (Basel, Switzerland) Sensors Volume 21 Issue 17 Sensors, Vol 21, Iss 5856, p 5856 (2021) |
| ISSN: | 1424-8220 |
| Popis: | Sign language is designed to assist the deaf and hard of hearing community to convey messages and connect with society. Sign language recognition has been an important domain of research for a long time. Previously, sensor-based approaches have obtained higher accuracy than vision-based approaches. Due to the cost-effectiveness of vision-based approaches, researchers have been conducted here also despite the accuracy drop. The purpose of this research is to recognize American sign characters using hand images obtained from a web camera. In this work, the media-pipe hands algorithm was used for estimating hand joints from RGB images of hands obtained from a web camera and two types of features were generated from the estimated coordinates of the joints obtained for classification: one is the distances between the joint points and the other one is the angles between vectors and 3D axes. The classifiers utilized to classify the characters were support vector machine (SVM) and light gradient boosting machine (GBM). Three character datasets were used for recognition: the ASL Alphabet dataset, the Massey dataset, and the finger spelling A dataset. The results obtained were 99.39% for the Massey dataset, 87.60% for the ASL Alphabet dataset, and 98.45% for Finger Spelling A dataset. The proposed design for automatic American sign language recognition is cost-effective, computationally inexpensive, does not require any special sensors or devices, and has outperformed previous studies. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|