Open AccessApplication of spectral and spatial processing methods to sonar images
Author(s) -
А. В. Кокошкин,
Е. П. Новичихин,
Ilia V. Smolyaninov
Publication year - 2021
Publication title -
radioèlektronika, nanosistemy, informacionnye tehnologii
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.204
H-Index - 5
eISSN - 2414-1267
pISSN - 2218-3000
DOI - 10.17725/rensit.2021.13.377
Subject(s) - sonar , speckle noise , speckle pattern , synthetic aperture sonar , computer vision , artificial intelligence , noise (video) , grayscale , computer science , brightness , image quality , remote sensing , image (mathematics) , optics , physics , geography
The paper proposes the use of the method of renormalization with limitation (MRL) for suppressing the speckle noise of images obtained using sonar. The method is tested on real images obtained by the interferometric side-view sonar. The principal possibility of a significant reduction in the speckle noise level is found due to the fact that the MRL renormalizes the spectrum of the sonar image to the universal reference spectrum (URS) model, which is a model of the spectrum of a "good" quality grayscale image. To increase the overall sharpness of the image, after applying the MRL, it is proposed to use spatial brightness transformations. The study allows us to conclude that the application of MRL to sonar images can significantly reduce speckle noise.