A METHOD FOR FILTERING IMAGE NOISE IN A VIDEO SURVEILLANCE SYSTEM

Abstract

: The article discusses the features of video information processing, using the two-way filtering method to attenuate the visibility of noise in images in a video surveillance system. The two-way filtering method allows you to eliminate low-contrast details of small sizes from the image, while at the same time keeping the light boundaries created by large brightness differences clear. This approach made it possible to reduce the size of the processed area without loss of processing quality for low-intensity noise. In the experimental part of the article, a comparative analysis of the two-way filtering method for suppressing various types of noise on images in a video surveillance system is carried out. Images distorted by noise with a probability of 20% were used for modeling. Numerical evaluation of the quality of image noise purification based on the peak signal-to-noise ratio (PSNR) and standard deviation (SD) showed that the proposed methods show the best processing result in all considered cases compared to known approaches. The results obtained in the article can find wide practical application in the processing of satellite and medical images, geophysical data and other applications of digital image processing.

Keywords:

:video surveillance system, image processing, impulse noise, filtering, peak signal-to-noise ratio, standard deviation.