Application of Deep Neural Networks in Recognition of Selected Types of Objects in Digital Images
Year: 2025
Type: article
Abstract: The presented article focuses on the application of deep neural networks for recognizing selected types of objects in digital images. Various learning techniques, network architectures, and hyperparameters were analyzed to optimize the detection quality. This study compared supervised learning, self-supervised learning, and transfer learning methods, with YOLOv8 showing the highest effectiveness in the transfer learning approach. This article is of a typical application nature. The results included in it can be used by a potential reader for other applications related to object recognition in digital images.
Source: Applied Sciences
DOI: 10.3390/app15147931
Authors: Artur Babiarz, Michał Bugaj
Institution: Silesian University of Technology
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W4412480740