VIDEO TASVIRLARDA HARAKATDAGI OBYEKTLARNI ANIQLASHNING INTELLEKTUAL TIZIMINI ISHLAB CHIQISH

20260 citationsJournal Articlegreen Open Access

Authors

Bahodir Obilov · Kimyo International University in Tashkent

Abstract

Ushbu maqolada video oqimlarda harakatdagi obyektlarni aniqlash va kuzatish uchun zamonaviy intellektual tizim arxitekturasi taklif etiladi. Maqola YOLO, Faster/Mask R‑CNN, SSD, DETR kabi detektorlar hamda SORT/DeepSORT, FairMOT, ByteTrack singari ko‘p obyektli kuzatuv (MOT) usullari tahliliga tayangan holda, real vaqt talablari (FPS), aniqlik ko‘rsatkichlari (mAP, IDF1) va resurs cheklovlari o‘rtasidagi muvozanatni yoritadi. Shuningdek, O‘zbekistonda yo‘l harakatini monitoring qilish bo‘yicha nashrlardan amaliy misollar keltiriladi.

Topics & Keywords

Engineering and Agricultural Innovations Advanced Computational Techniques in Science and Engineering Education, Innovation and Language Studies

UN Sustainable Development Goals

Sustainable cities and communities

Publication Details

Published in: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.19376141

Field-Weighted Citation Impact: 0.00