RAQAMLI TASVIRLARNI INTERPOLATSION TIKLASHNING OCR TIZIMLARI ANIQLIGIGA TA’SIRINI KOMPLEKS TAHLIL QILISH
Keywords:
interpolatsiya, raqamli tasvir, OCR, PSNR, SSIM, MSE, EPI, FSIM, CER, WER, adaptiv algoritm, tasvirni tiklash.Abstract
Mazkur maqolada raqamli tasvirlarni interpolatsion tiklash algoritmlarining optik belgilarni tanib olish (OCR) tizimlari aniqligiga ta’siri kompleks eksperimental va analitik yondashuv asosida o‘rganiladi. Tadqiqot doirasida bilinear, bicubic, spline, Lanczos hamda taklif etilgan adaptiv interpolatsiya usullari qiyosiy tahlil qilinib, ularning samaradorligi tasvir sifati metrikalari (PSNR, SSIM, MSE, EPI, FSIM) hamda OCR ko‘rsatkichlari (Character Accuracy, Word Accuracy, CER, WER) orqali baholanadi. Olingan natijalar interpolatsiya jarayonida strukturaviy invariantlarni, ayniqsa kontur va chegaraviy elementlarni saqlash OCR tizimlari samaradorligini sezilarli darajada oshirishini ko‘rsatadi.
References
UNCTAD. Digital Economy Report 2024: Shaping an Environmentally Sustainable and Inclusive Digital Future. United Nations Conference on Trade and Development (UNCTAD), Geneva, 2024.
Smith R. An Overview of the Tesseract OCR Engine. Proceedings of the Ninth International Conference on Document Analysis and Recognition (ICDAR), IEEE Computer Society, Curitiba, Brazil, 2007, pp. 629–633.
IDC. The Digitization of the World: From Edge to Core (Data Age 2025). International Data Corporation (IDC), Framingham, MA, USA, 2018.
NVIDIA. AI and Data Processing Trends: Accelerating Data-Centric Computing in the Era of Artificial Intelligence. NVIDIA Technical Report, Santa Clara, CA, USA, 2023.
DataReportal. Digital 2025: Uzbekistan – Insights into Internet Usage, Mobile Connectivity, and Digital Adoption. Kepios & DataReportal, 2025.
Downloads
Published
Issue
Section
How to Cite
