INFORMATION SCIENCE

Comparative analysis of filtration algorithms for images of architectural plans

N. A. Aksionova

Francisk Skorina Gomel State University

Abstract: The use of smoothing filters for pre-processing images of architectural plans is considered. A comparative analysis of Gaussian and Roudin – Osher – Fatemi (ROF) filters based on the Chambolle model is carried out. The software modules are implemented in the Python programming language using OpenCV. The results showed that for pre-processing not very noisy images it is better to use a Gaussian filter, and for images with a high noise level it is better to use the ROF filter, which prevents the loss of special corner points.

Keywords: preprocessing, filtering, noise, Gaussian blur, Gaussian filter, Rudin – Osher – Fatemi filter, Chambolle model, Python, OpenCV.

UDC: 004.932:004.94

Received: 17.05.2024

DOI: 10.54341/20778708_2024_3_60_86