Deteksi Pemalsuan Wajah Berbasis ResNet-50 dengan Fitur LBP-HOG-CTA CbCr: Kinerja Intra-Dataset dan Generalisasi Antar-Dataset
Keywords:
CTA CbCr, Deep Learning, Face Spoof Detection, HOG, LBP, Resnet-50,Abstract
Sistem pengenalan wajah masih menghadapi kerentanan terhadap serangan pemalsuan, khususnya dalam kondisi pencahayaan yang buruk, yang dapat mengancam keandalan proses verifikasi identitas. Tantangan utama terletak pada kesulitan sistem dalam membedakan wajah asli dan tiruan, yang semakin kompleks akibat mudahnya akses terhadap citra wajah melalui media sosial serta permasalahan domain shift pada data dunia nyata. Penelitian ini mengusulkan metode Face Spoofing Detection (FSD) berbasis deep learning dengan memanfaatkan arsitektur ResNet-50 yang dikombinasikan dengan teknik ekstraksi fitur Local Binary Pattern (LBP), Histogram of Oriented Gradients (HOG), dan Chromatic Textural Analysis (CTA CbCr) diharapkan dapat menangkap perbedaan halus antara ciri asli dan palsu. Evaluasi dilakukan secara menyeluruh terhadap beberapa dataset publik (CASIA-FASD, NUAA, OULU) serta dataset pribadi yang merepresentasikan kondisi aktual di lapangan. Hasil pengujian intra-dataset menunjukkan performa yang sangat baik pada akurasi dan F1-score, di mana CTA CbCr menghasilkan performa paling optimal dalam sebagian besar skenario (EER dan HTER mencapai 0,00 pada CASIA dan OULU). Namun, hasil pengujian lintas dataset dan pada data pribadi menunjukkan penurunan kinerja yang signifikan, menandakan adanya tantangan serius terkait domain shift. Meskipun pelatihan dengan multi-dataset memberikan peningkatan generalisasi pada beberapa dataset publik, performa pada data pribadi tetap terbatas. Penelitian ini menekankan potensi kuat dari kombinasi ResNet-50 dengan fitur tekstur dan warna dalam lingkungan yang terkendali, namun juga menunjukan kebutuhan pendekatan adaptasi domain yang lebih robust untuk penerapan FSD di dunia nyata.References
A. S. Budi, Suma’inna, and H. Maulana, “Pengenalan Citra Wajah Sebagai Identifier Menggunakan Metode Principal Component Analysis (PCA),” JURNAL TEKNIK INFORMATIKA, vol. 9, no. 2, 2016.
Sumijan, P. A. W. Purnama, and S. Arlis, Teknologi Biometrik: Impementasi pada Bidang Medis Menggunakan Matlabs. Solok: INSAN CENDEKIA MANDIRI , 2021.
A. Olayiwola et al., “Optimization Of An Identity Access control System Using Biometric Techniques,” International Journal of Progressive Sciences and Technologies (IJPSAT, vol. 27, no. 2, pp. 647–653, 2021, [Online]. Available: http://ijpsat.ijsht-journals.org
Y. Wang, F. Nian, T. Li, Z. Meng, and K. Wang, “Robust face anti-spoofing with depth information,” J Vis Commun Image Represent, vol. 49, pp. 332–337, Nov. 2017, doi: 10.1016/j.jvcir.2017.09.002.
N. Daniel and A. Anitha, “Texture and quality analysis for face spoofing detection,” Computers and Electrical Engineering, vol. 94, Sep. 2021, doi: 10.1016/j.compeleceng.2021.107293.
Z. Wang et al., “Domain Generalization via Shuffled Style Assembly for Face Anti-Spoofing,” IEEE , Mar. 2022, [Online]. Available: http://arxiv.org/abs/2203.05340
K. Balamurali, S. Chandru, M. S. Razvi, and V. Sathiesh Kumar, “Face Spoof Detection Using VGG-Face Architecture,” in Journal of Physics: Conference Series, IOP Publishing Ltd, Jun. 2021. doi: 10.1088/1742-6596/1917/1/012010.
P. Jaswanth, P. Y. chowdary, and M. V. S. Ramprasad, “Deep learning based intelligent system for robust face spoofing detection using texture feature measurement,” Measurement: Sensors, vol. 29, Oct. 2023, doi: 10.1016/j.measen.2023.100868.
Z. Zhang, J. Yan, S. Liu, Z. Lei, D. Yi, and S. Z. Li, “A Face Antispoofing Database with Diverse Attacks,” 2012.
L. Li, X. Feng, Z. Xia, X. Jiang, and A. Hadid, “Face spoofing detection with local binary pattern network,” J Vis Commun Image Represent, vol. 54, pp. 182–192, Jul. 2018, doi: 10.1016/j.jvcir.2018.05.009.
X. Shu, H. Tang, and S. Huang, “Face spoofing detection based on chromatic ED-LBP texture feature,” Multimed Syst, vol. 27, no. 2, pp. 161–176, Apr. 2020, doi: 10.1007/s00530-020-00719-9.
W. Zhang and S. Xiang, “Face anti-spoofing detection based on DWT-LBP-DCT features,” Signal Process Image Commun, vol. 89, Nov. 2020, doi: 10.1016/j.image.2020.115990.
Y. Arti and A. M. Arymurthy, “Face Spoofing Detection using Inception-v3 on RGB Modal and Depth Modal,” Jurnal Ilmu Komputer dan Informasi, vol. 16, no. 1, pp. 47–57, Mar. 2023, doi: 10.21609/jiki.v16i1.1100.
Y. Kong, X. Li, G. Hao, and C. Liu, “Face Anti-Spoofing Method Based on Residual Network with Channel Attention Mechanism,” Electronics (Switzerland), vol. 11, no. 19, Oct. 2022, doi: 10.3390/electronics11193056.
S. Megawan and W. S. Lestari, “Deteksi Spoofing Wajah Menggunakan Faster R-CNN dengan Arsitektur Resnet50 pada Video (Face Spoofing Detection Using Faster R-CNN with Resnet50 Architecture on Video),” 2020. [Online]. Available: https://www.idiap.ch/dataset/replayattack.
F. A. Pujol, M. J. Pujol, C. Rizo-Maestre, and M. Pujol, “Entropy-based face recognition and spoof detection for security applications,” Sustainability (Switzerland), vol. 12, no. 1, pp. 1–18, Jan. 2020, doi: 10.3390/SU12010085.
Published
How to Cite
Issue
Section
License
The copyright to this article is transferred to Universitas Mercu Buana (UMB) if and when the article is accepted for publication. The undersigned hereby transfers any and all rights in and to the paper including without limitation all copyrights to UMB. The undersigned hereby represents and warrants that the paper is original and that he/she is the author of the paper, except for material that is clearly identified as to its original source, with permission notices from the copyright owners where required. The undersigned represents that he/she has the power and authority to make and execute this assignment.
We declare that this paper has not been published in the same form elsewhere.
Furthermore, I/We hereby transfer the unlimited rights of publication of the above-mentioned paper as a whole to UMB. The copyright transfer covers the right to reproduce and distribute the article, including reprints, translations, photographic reproductions, microform, electronic form (offline, online) or any other reproductions of similar nature.
The corresponding author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. This agreement is to be signed by at least one of the authors who have obtained the assent of the co-author(s) where applicable. After submission of this agreement signed by the corresponding author, changes of authorship or in the order of the authors listed will not be accepted.
Retained Rights/Terms and Conditions
Although authors are permitted to re-use all or portions of the Work in other works, this does not include granting third-party requests for reprinting, republishing, or other types of re-use.
Our Articles are licensed under CC BY-NC
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
