Computer Vision-Based Smart Artificial Hand for Upper Limb Amputee

Anish Sambhare, Ajay Lohar, Rushank Suryawanshi, Narendra Bhagat

2025

Abstract

This study introduces a novel approach for classifying grips in prosthetic hands to facilitate object manipulation using Convolutional Neural Networks (CNN). The experimental findings indicate that this method outperforms traditional learning models. Initially, three machine learning algorithms were assessed: Decision Tree, SVM, and Random Forest, to classify various objects such as bottles, cell phones, and cups. The classification accuracies achieved were 76%, 90%, and 95%, respectively. Traditional machine learning preprocessing techniques proved to be quite complex, making CNNs a more attractive option due to their ability to perform feature extraction and classification without the need for extensive preprocessing. The CNN developed in this study achieved a training accuracy of 99% and a testing accuracy of 97.5%, surpassing contemporary models like YOLO v3 and Faster R-CNN. The integration of data-augmented training and dropout regularization enhances the model’s robustness and generalizability. This allows the prosthetic hand to achieve precise grip control in a cost-effective and sensor-free manner, making the system a dependable choice for real-time applications, thereby improving accessibility and functionality in prosthetic design.

Paper Citation

in Harvard Style

Sambhare A., Lohar A., Suryawanshi R. and Bhagat N. (2025). Computer Vision-Based Smart Artificial Hand for Upper Limb Amputee. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 413-419. DOI: 10.5220/0013619200004664

in Bibtex Style

@conference{incoft25,
author={Anish Sambhare and Ajay Lohar and Rushank Suryawanshi and Narendra Bhagat},
title={Computer Vision-Based Smart Artificial Hand for Upper Limb Amputee},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT},
year={2025},
pages={413-419},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013619200004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT
TI - Computer Vision-Based Smart Artificial Hand for Upper Limb Amputee
SN - 978-989-758-763-4
AU - Sambhare A.
AU - Lohar A.
AU - Suryawanshi R.
AU - Bhagat N.
PY - 2025
SP - 413
EP - 419
DO - 10.5220/0013619200004664
PB - SciTePress