Abstract:
The article deals with the problem of image processing using pulse noise. These interferences occur in an unstable communication channel between an aircraft or a spacecraft where sensors and primary signal processing devices are located and a ground processing station where final signal processing is done. The solution of this problem is based on the theory of signals at finite time intervals. This article deals with the methods of frequency filtering based on Walsh transform. The method of quasi two-dimensional filtering is based on application of Walsh transform to image rows or columns. The article provides a comparative analysis of this problem solution by spatial and frequency methods of image processing. To assess the quality of image recovered, the mean square error is used, being the most popular and common metrics in routine practice.
