EKSTREMAL YORITILGANLIK SHAROITIDA YUZ TASVIRI ASOSIDA SHAXSNI IDENTIFIKATSIYALASHDA CHUQUR O‘QITISH USULLARI
Abstract
Ushbu maqolada ekstremal yoritilganlik (o‘ta qorong‘i, haddan tashqari yorqin va teskari yorug‘lik) sharoitida yuz tasviri asosida shaxsni identifikatsiyalash muammolari va ularning sun’iy intellektga tayangan yechimlari tadqiq etilgan. An’anaviy kompyuterli ko‘rish algoritmlarining yorug‘lik o‘zgarishlariga nisbatan past samaradorligi tahlil qilinib, zamonaviy chuqur o‘qitish (Deep Learning) usullari taklif etilgan.
References
1. NIST Interagency Report 8331, Face Recognition Accuracy with Face Masks. // National Institute of Standards and Technology, 2021.
2. Wang M., Deng J. MaskedFace-Net: A Dataset for Masked Face Recognition. // IEEE International Conference on Computer Vision (ICCV), 2020.
3. Huang G., Li Z., Wang D. Adaptive Illumination Correction for Face Recognition in Unconstrained Environments. // IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 42, No. 3, 2020, pp. 567-582.
4. Kumar M.A., Bhattacharya S. Attention-based Convolutional Neural Network for Occluded Face Recognition. // IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 2345-2354.
5. Deng J., Guo J., Xue N., Zafeiriou S. ArcFace: Additive Angular Margin Loss for Deep Face Recognition. // IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 4685-4694.
6. Zhu X., Lei Z., Liu X., Shi H., Li S.Z. Face Recognition with Mask: Challenges and Solutions. // IEEE Transactions on Biometrics, Behavior, and Identity Science, Vol. 4, No. 2, 2022, pp. 152-165.
7. Kamilov M.M., Fozilov Sh.X. Tasvirlarga raqamli ishlov berish usullari va ularning amaliy qo'llanilishi. // Fan va texnologiya’ nashriyoti, Toshkent, 2019, 256 b.
8. Mamatov N.S., Niyozmatova N.A. Yuz tasviri asosida shaxsni identifikatsiyalash algoritmlarini ishlab chiqish. // Muhammad al-Xorazmiy nomidagi TATU axborotnomasi, №4, 2020, 78-85 b.
