Page 460 - Kaleidoscope Academic Conference Proceedings 2024

2024 ITU Kaleidoscope Academic Conference

CONCLUSIONS& FUTURE WORK: Process. Comput. Intell. E-Age, vol. 4, pp.
This research examines the impact of data 2137–2141, 2002, doi:
augmentation in deep learning to improve 10.1109/ICONIP.2002.1199054.
human-computer interactions by predicting the [7] G. R. Naik, D. K. Kumar, H. Weghorn, and
correct hand motions of humans. After doing M. Palaniswami, “Subtle hand gesture
this study, we can conclude that CNN is a data- identification for HCI using temporal
driven methodology and that data augmentation Decorrelation source separation BSS of
has a significant impact on deep learning. As a surface EMG,” Proc. - Digit. Image Comput.
result, this proposed objective can be used to Tech. Appl. 9th Bienn. Conf. Aust. Pattern
forecast more accurate results in terms of better Recognit. Soc. DICTA 2007, pp. 30–37,
communication between humans and machines. 2007, doi: 10.1109/DICTA.2007.4426772.
Convolutional Neural Networks (CNN), which
have an accuracy of 99.76%, produce the best [8] H. Heo, E. C. Lee, K. R. Park, C. J. Kim,
outcomes. Random forest findings outperform and M. Whang, “A realistic game system
other standard algorithms, barring CNN, with an using multi-modal user interfaces,” IEEE
accuracy of 84.43%. Deaf and mute people will Trans. Consum. Electron., vol. 56, no. 3, pp.
also become accustomed to that in their 1364–1372, 2010, doi:
everyday conversation, and these data can also 10.1109/TCE.2010.5606271.
be utilized for researchers working in the same [9] P. Trigueiros, F. Ribeiro, and L. P. Reis, “A
field. In the Future for the balancing dataset, comparison of machine learning algorithms
upsampling techniques can be used. The applied to hand gesture recognition,” Iber.
imbalance dataset can be transformed into the Conf. Inf. Syst. Technol. Cist., no. January,
balance dataset via smoteup sampling. Machines 2012.
can operate on samples equally when all classes [10] Z. Chen, J. Kim, J. Liang, J. Zhang, and Y.
have the same number of values. Since we are Yuan, “Real-Time Hand Gesture
now attaining accuracy close to 99%, we are not Recognition Using Finger Segmentation,”
using this strategy. But in the future, we can use vol. 2014, 2014.
this approach to obtain greater precision [11] G. Li et al., “Hand gesture recognition based
Conflicts of Interest: The authors declare no on convolution neural network,” Cluster
conflict of interest. Comput., vol. 22, pp. 2719–2729, 2017, doi:
REFERENCES: 10.1007/s10586-017-1435-x.
[1] A. Mindess, Reading Between the Signs: [12] S. R. Kalbhor and A. M. Deshpande, “Digit
Intercultural Communication for Sign Recognition Using Machine Learning and
Language Interpreters. John Murray Press, Convolutional Neural Network,” Proc. 2nd
2014. [Online]. Available: Int. Conf. Trends Electron. Informatics,
https://books.google.co.in/books?id=9vd8D ICOEI 2018, pp. 604–609, 2018, doi:
AAAQBAJ 10.1109/ICOEI.2018.8553954.
[2] T. Kumaravel, T. Pradeep, V. J. Mouliesh, [13] Y. Z. and L. Wang, “Yiwen Zhao apllication
and P. Ragul, “Hand Gesture Recognition of CNN in sign llanguage,” 2018.
By Deep Convolutional Neural Network,” [14] M. Z. Islam, M. S. Hossain, R. Ul Islam, and
vol. 12, no. 7, pp. 1751–1761, 2021. K. Andersson, “Static hand gesture
[3] A. D. Wilson and A. F. Bobick, “Learning recognition using convolutional neural
visual behavior for gesture analysis,” Proc. network with data augmentation,” 2019 Jt.
IEEE Int. Conf. Comput. Vis., pp. 229–234, 8th Int. Conf. Informatics, Electron. Vision,
1995, doi: 10.1109/iscv.1995.477006. ICIEV 2019 3rd Int. Conf. Imaging, Vis.
[4] S. S. Pravardhan, C. Vamsi, T. B. Kumar, Pattern Recognition, icIVPR 2019 with Int.
M. Saad, U. Rahman, and C. V. Kumar, Conf. Act. Behav. Comput. ABC 2019, pp.
“Hand Gesture Recognition for Mute People 324–329, 2019, doi:
Using Machine Learning,” vol. 11, no. 06, 10.1109/ICIEV.2019.8858563.
pp. 469–473, 2022. [15] R. F. Pinto, C. D. B. Borges, A. M. A.
[5] P. Thwe and M. Thè, “Hand Gesture Almeida, and I. C. Paula, “Static Hand
Detection and Recognition System: A Gesture Recognition Based on
Critical Review,” Int. J. Comput., vol. 32, Convolutional Neural Networks,” J. Electr.
pp. 64–72, 2019. Comput. Eng., vol. 2019, 2019, doi:
[6] S. G. Wysoski, M. V. Lamar, S. Kuroyanagi, 10.1155/2019/4167890.
and A. Iwata, “A rotation invariant approach [16] J. P. Sahoo, S. Ari, and S. K. Patra, “Hand
on static-gesture recognition using boundary gesture recognition using PCA based deep
histograms and neural networks,” ICONIP CNN reduced features and SVM classifier,”
2002 - Proc. 9th Int. Conf. Neural Inf. Proc. - 2019 IEEE Int. Symp. Smart

– 416 –