YUZ, HARAKAT VA VAQTGA BOG‘LIQ MODELLAR ASOSIDA KO‘P MODALLI BIOMETRIK UZLUKSIZ AUTENTIFIKATSIYA TIZIMI: MUAMMOLAR VA YECHIMLAR
Alayev R. H. & Nurmatov A. S.
Annotatsiya
Ushbu tezisda web-kamera video oqimi asosida foydalanuvchini sessiya davomida uzluksiz autentifikatsiya qilish masalasi bo‘yicha olib borilgan so‘nggi ilmiy ishlar tahlil qilinadi. Uzluksiz autentifikatsiya odatiy bir martalik autentifikatsiyadan farqli ravishda, foydalanuvchini seans davomida haqiqiy yoki haqiqiy emasligini passiv va uzluksiz tekshiruvchi yondashuv bo‘lib, u biometrik, xulqiy va kontekstga oid sensor belgilariga tayangani sababli xavfsizlik bilan birga maxfiylik muammolarini ham keltirib chiqaradi [1]. Tahlil doirasida yuz biometrikasi (CNN/embedding), harakat signallari (bosh holati, kichik harakatlar kabi) hamda vaqtga bog‘liq modellar (LSTM/GRU va Transformer) asosida ko‘p modalli (multimodal) autentifikatsiya arxitekturasi qanday shakllanishi, qaysi bosqichlarda xatoliklar yuzaga kelishi va ularni kamaytirish yechimlari yoritiladi. Ko‘p modalli yondashuvning motivatsiyasi — yakka modal tizimlardagi samaradorlik pastligi va taqdimot hujumlari (presentation attacks) xavfi — ham nazariy, ham amaliy manbalar orqali asoslanadi [2]. Shuningdek, spoofing/replay va deepfake tahdidlariga qarshi PAD/tiriklikni aniqlash standart metrikalari, real vaqt sharoitida resurs cheklovlari [3] va tasvir sifati masalalari bo‘yicha xulosalar beriladi.
Adabiyotlar
- Baig, A.F., Eskeland, S. (2021). Security, Privacy, and Usability in Continuous Authentication: A Survey. Sensors, 21(17), 5967.
- Ammour, N., Bazi, Y., Alajlan, N. (2023). Multimodal Approach for Enhancing Biometric Authentication. Journal of Imaging, 9(9), 168.
- Zeeshan, N., Moradpoor, N., La Spada, L. (2025). Continuous Authentication in Resource-Constrained Devices via Biometric and Environmental Fusion. Sensors, 25(18), 5711.
- Qin, Z., Zhao, P., Zhuang, T., Deng, F., Ding, Y., Chen, D. (2023). A survey of identity recognition via data fusion and feature learning. Information Fusion, 91, 694–712.
- Schroff, F., Kalenichenko, D., Philbin, J. (2015). FaceNet: A Unified Embedding for Face Recognition and Clustering.
- Deng, J., Guo, J., Yang, J., et al. (2019). ArcFace: Additive Angular Margin Loss for Deep Face Recognition.
- Grother, P., Ngan, M., Hanaoka, K. (2019). FRVT Ongoing: Face Recognition Quality Assessment—Concept and Goals.
- NIST. Face Recognition Vendor Test (FRVT): Program Overview and Reports.
- Grother, P. (2022). Face Recognition Vendor Test (FRVT) Part 8: Summarizing Demographic Differentials (NIST IR 8429).
- ISO/IEC 30107-3:2017. Biometric presentation attack detection — Testing and reporting.
- Busch, C. The ISO/IEC 30107-3 standard for testing of Presentation Attack Detection (PAD).
- Amerini, I., et al. (2025). Deepfake Media Forensics: Status and Future Challenges.