O‘ZGARUVCHAN YORUG‘LIK VA FON SHAROITLARIDA YUZNI ANIQLASH VA TANIB OLISHNING GIBRID MODELI
Keywords:
yuzni aniqlash, yuzni tanib olish, gibrid model, chuqur o'rganish, CNN, HOG, LBP, yorug'lik normalizatsiyasi, real vaqt qayta ishlash.Abstract
Ushbu maqolada o‘zgaruvchan yorug‘lik va fon sharoitlarida yuzni aniqlash va tanib olish uchun mo'ljallangan gibrid model taqdim etiladi. Dolzarbligi shundaki, real dunyo ilovalarida mavjud tizimlar past yoritilganlik, murakkab fonlar va okkluziya sharoitlarida aniqligini yo‘qotadi. Muammoni hal qilish uchun Retinex asosidagi yorug'lik normalizatsiyasi, HOG, LBP va CNN xususiyatlarini birlashtiradigan ko'p kanalli yondashuv hamda og'irlikli ansambl klassifikatsiyasi taklif etiladi. Natijada LFW ma'lumotlar to'plamida 94.7% aniqlik va 31 FPS tezligiga erishildi. Tadqiqotning ilmiy yangiligi — yorug'lik normalizatsiyasi bilan ko'p kanalli gibrid arxitekturani birlashtirishdan iborat bo'lib, bu mavjud usullarga nisbatan 12-15% aniqlikni oshiradi. Amaliy ahamiyati — xavfsizlik tizimlari, bank autentifikatsiyasi va tibbiy diagnostika sohalarida qo'llanilishi mumkin.
References
[1] Zhao W., Chellappa R., Phillips P.J., Rosenfeld A. Face recognition: A literature survey // ACM Computing Surveys. — 2003. — Vol. 35, No. 4. — P. 399–458. DOI: 10.1145/954339.954342
[2] Kortli D., Tjahjadi T. A survey of face recognition applications // Pattern Recognition Letters. — 2020. — Vol. 131. — P. 232–240. DOI: 10.1016/j.patrec.2019.12.016
[3] Wang M., Deng W. Deep face recognition: A survey // Neurocomputing. — 2021. — Vol. 429. — P. 215–244. DOI: 10.1016/j.neucom.2020.10.081
[4] LeCun Y., Bengio Y., Hinton G. Deep learning // Nature. — 2015. — Vol. 521. — P. 436–444. DOI: 10.1038/nature14539
[5] Viola P., Jones M. Rapid object detection using a boosted cascade of simple features // Proc. IEEE CVPR. — 2001. — Vol. 1. — P. 511–518. DOI: 10.1109/CVPR.2001.990517
[6] Dalal N., Triggs B. Histograms of oriented gradients for human detection // Proc. IEEE CVPR. — 2005. — Vol. 1. — P. 886–893. DOI: 10.1109/CVPR.2005.177
[7] Ojala T., Pietikäinen M., Mäenpää T. Multiresolution gray-scale and rotation invariant texture classification with LBP // IEEE Trans. Pattern Anal. Mach. Intell. — 2002. — Vol. 24, No. 7. — P. 971–987. DOI: 10.1109/TPAMI.2002.1017623
[8] Taigman Y., Yang M., Ranzato M., Wolf L. DeepFace: Closing the gap to human-level performance in face verification // Proc. IEEE CVPR. — 2014. — P. 1701–1708. DOI: 10.1109/CVPR.2014.220
[9] Schroff F., Kalenichenko D., Philbin J. FaceNet: A unified embedding for face recognition and clustering // Proc. IEEE CVPR. — 2015. — P. 815–823. DOI: 10.1109/CVPR.2015.7298682
[10] Wang H. et al. Additive margin softmax for face verification // IEEE Signal Process. Lett. — 2018. — Vol. 25, No. 7. — P. 926–930. DOI: 10.1109/LSP.2018.2831071
[11] Zhang Z., Luo P., Loy C.C., Tang X. Facial landmark detection by deep multi-task learning // Proc. ECCV. — 2014. — P. 94–108. DOI: 10.1007/978-3-319-10599-4_7
[12] Ranjan R., Patel V.M., Chellappa R. Hyperface: A deep multi-task learning framework // IEEE Trans. Pattern Anal. Mach. Intell. — 2019. — Vol. 41, No. 1. — P. 121–135. DOI: 10.1109/TPAMI.2017.2781233
