Attendance face detection on mobile device using particle swarm optimization and linear discriminant analysis
Keywords:
Artificial intelligence, Face detection, Mobile deviceAbstract
Artificial intelligence-based facial recognition has been utilized extensively in automation systems, however processing facial recognition properly is necessary to achieve high accuracy with reasonable computing time. On the model and mobile application side, there were two different kinds of experiments run in this study. Feature selection in modeling is done by the use of the Linear Discriminant Analysis method, K-Nearest Neighbor classification, and Particle Swarm Optimization. The ORL database, which contains 255 facial photographs from various persons, was the source of the dataset for this study. The K-Nearest Neighbor method model combined with Particle Swarm Optimization with Confusion Matrix results in the largest accuracy improvement in the model test, producing values of accuracy 97.77%, precision 97.22%, recall 97.61%, and k-fold 97.3, The average value has increased by three points compared to earlier testing without Particle Swarm Optimization. It results in an inference time of 4 to 5 seconds on Android 7 test devices and a half-second inference time on Android 11 test devices when tested on actual devices to measure the response speed of objects identified by mobile devices for facial recognition.
References
[1] A. Pradipta and S. A. Yogyakarta, “Implementasi Metode Caesar Chiper Alphabet Majemuk Dalam Kriptografi Untuk Pengamanan Informasi,” ijns.org Indonesian Journal on Networking and Security, vol. 5, p. 3, 2016.
[2] A. Kurnia, A., and M. A. I. Pakereng, “Perancangan Teknik Kriptografi Block Cipher Berbasis Pola Batik Ceplok Yogyakarta,” vol. 8, no. 1, 2019.
[3] M. Nurtanzis Sutoyo, “Kombinasi Algoritma Kriptografi Caesar Chiper dan Vigenere Chiper Untuk Keamanan Data,” vol. 2, no. 1, 2016.
[4] R. Febrianingsih and A. Hafiz, “Implementasi Kriptografi Berbasis Caesar Chiper Untuk Keamanan Data.”
[5] A. Anto Tri Susilo, “Penerapan Algoritma Asimetris RSA Untuk Keamanan Data Pada Aplikasi Penjualan CV. Sinergi Computer Lubuklinggau Berbasis Web,” Jurnal SIMETRIS, vol. 9, no. 2, 2018.
[6] B. Schneier, Applied cryptography : protocols, algorithms, and source code in C. Wiley, 1996.
[7] M. Algoritma Caesar Cipher, H. Dwi Purnomo, and I. Sembiring, “Modifikasi Algoritma Caesar Cipher pada Kode ASCII dalam Meningkatkan Keamanan Pesan Teks,” JIFOTECH (Journal of Information Technology), vol. 2, no. 1, 2022.
[8] F. Maqsood, M. Ahmed, M. Mumtaz Ali, and M. Ali Shah, “Cryptography: A Comparative Analysis for Modern Techniques,” 2017. [Online]. Available: www.ijacsa.thesai.org
[9] S. Godara, S. Kundu, and R. Kaler, “An Improved Algorithmic Implementation of Rail Fence Cipher,” International Journal of Future Generation Communication and Networking, vol. 11, no. 2, pp. 23–32, Mar. 2018, doi: 10.14257/ijfgcn.2018.11.2.03.
[10] T. Zebua and S. Limun, “Analisa Dan Implementasi Algoritma Triangle Chain Pada Penyandian Record Database,” Pelita Informatika Budi Darma, vol. III, no. 2 , 2013.
