SUN’IY INTELLEKT YORDAMIDA MOBIL KAMERA TASVIRLARIDA OBYEKTLARNI ANIQLASH VA ULARNING INGLIZCHA NOMLARINI OVOZLI IFODALASH TIZIMINI ISHLAB CHIQISH
Keywords:
mobile camera, object detection, open-vocabulary detection, OWLv2, Android, CameraX, Text-to-Speech, computer vision, school supplies, English vocabulary, multimodal learning, temporal filtering.Abstract
This article presents the conceptual and technical architecture of a
system designed to recognize school supplies in a real-world environment using an
Android mobile device camera and to provide voice output of the English names of
the recognized objects; it also discusses the results of pilot testing the artificial
intelligence component on static images. The proposed architecture integrates an
image stream generated via CameraX, a server-side open-vocabulary object detection
module, a result filtering module, and Android Text-to-Speech technology
