Work

Effect of Frame Processing Frequency on Object Identification Using MobileNetV2 Neural Network for a Mobile Robot

Year: 2024

Type: article

Abstract: This article is devoted to the analysis of the relationship between frame processing frequency, identification speed, recognition accuracy and system resources. The article presents a mathematical description of the dependencies, which allows for a quantitative assessment of the effect of changing the processing frequency on these parameters. A program code based on the Python language was developed, which integrates the MobileNetV2 model for practical testing and optimization. The conducted experiments allow us to identify optimal settings that provide a balance between recognition accuracy and processing speed, taking into account system resources. The results of the study provide useful recommendations for improving computer vision systems in mobile robots

Source: Multidisciplinary Journal of Science and Technology

DOI: 10.5281/zenodo.13270677

Authors: Dmytro Gurin, Vladyslav Yevsieiev, Svitlana Maksymova, Amer Abu-Jassar

Institutions: Kharkiv National University of Radio Electronics, V. N. Karazin Kharkiv National University, Ajloun National University

Language: English

Topic: Advanced Neural Network Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: diamond

OpenAlex ID: W6949194437

FWCI:

Cites:

Cited by:

Related to: