Improving the Accuracy of Identifying Real-Time Indian Twins Using CNN Compared with Random Forest

Vallipi Dasaratha, J. Sheela

2023

Abstract

The objective of this study is to achieve real-time identification and analysis of Indian twins using the random forest algorithm, while also comparing its performance with the Convolutional Neural Network (CNN) algorithm in terms of accuracy. Materials and Methods: For the purpose of face recognition of twins with face and ID recognition, the random forest algorithm is chosen over the Convolutional Neural Network (CNN). The study involves two groups, namely Group 1 and Group 2, with an overall sample size of 1430 and 20 sample iterations for each group. Results and Discussion: The comparison and classification of real-time Indian twins are conducted using the Random Forest algorithm and the performance is measured using the CNN algorithm. The achieved accuracy rates are 52.3965% for Random Forest and 64.305% for CNN. By comparing the accuracy of both algorithms, independent sample tests reveal a statistically significant difference with a p-value of 0.001 (p<0.05), confirming the significance of the hypothesis through an independent sample t-test. Conclusion: This study evaluated the effectiveness of two image processing algorithms, namely Random Forest and CNN. The results indicate that Random Forest achieves an accuracy of 52.3965%, outperforming CNN which achieved an accuracy of 64.3050%. This suggests that for identification using ID recognition, Random Forest provides superior performance compared to CNN.

Paper Citation

in Harvard Style

Dasaratha V. and Sheela J. (2023). Improving the Accuracy of Identifying Real-Time Indian Twins Using CNN Compared with Random Forest . In Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT; ISBN 978-989-758-661-3, SciTePress, pages 487-493. DOI: 10.5220/0012520100003739

in Bibtex Style

@conference{ai4iot23,
author={Vallipi Dasaratha and J. Sheela},
title={Improving the Accuracy of Identifying Real-Time Indian Twins Using CNN Compared with Random Forest },
booktitle={Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT},
year={2023},
pages={487-493},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012520100003739},
isbn={978-989-758-661-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT
TI - Improving the Accuracy of Identifying Real-Time Indian Twins Using CNN Compared with Random Forest
SN - 978-989-758-661-3
AU - Dasaratha V.
AU - Sheela J.
PY - 2023
SP - 487
EP - 493
DO - 10.5220/0012520100003739
PB - SciTePress