Technique for remove harmonic noise on digital images

Аuthors

Gusev V. Y.^*, Krapivenko A. V.

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

*e-mail: gusev3@mail.ru

Abstract

In this paper have considered the problem to remove harmonic noise from image. Such noise can appear on the satellite photos owing to the nature of the equipment (the so-called dark current drift). For remove this noise, we use filtering in frequency domain after application Fourier transform. Using this approach offers significant advantages over the methods of spatial filtering when using the luminance values of pixels and can be a noticeable loss of useful information in the form of reduced sharpness and blur boundaries. When using frequency filtration to remove the periodic noise are removed mostly just noise, as cut is only a small portion of the spectrum of whole image. And this method is useful enough, because exist algorithms for fast Fourier transform (FFT).

Keywords:

frequency filtering of images, fast Fourier transform, periodic noise in the images.

Download