Asosiy kontentga oʻtish
Boshqa

Integration of Computer Vision and Machine Learning Technologies for Accurate Classification of Objects in Images

Karimov Sardor Ilkhom UgliFerghana State UniversityKholiknazarov Rashidjon KhomidjonovichFerghana Polytechnical InstituteErgashev Otabek Mirzapo’LatovichFergana State Technical University,Department of Computer Engineering and Artificial Intelligence,Fergana,Uzbekistan
2025

Ochiq kirish: yoʻq

ABI

Annotatsiya

This paper presents an in-depth analysis of modern object-based image classification methods based on the integration of VGG19, U-Net, and MobileNet deep learning architectures. Particular attention is paid to a comparative study of their structural and functional characteristics: the high ability to extract informative features in VGG19, the effectiveness of segmentation analysis in U-Net, and the computational compactness and energy efficiency of MobileNet. A comprehensive experiment was conducted on open datasets using key quality assessment metrics—Accuracy, Precision, Recall, and F1-score. The results of the analysis confirmed that the combined use of these architectures provides an optimal balance between classification accuracy, processing speed, and computational performance.

Hali tarjima qilinmagan

Identifikatorlar

Iqtiboslar va manbalar