Digital Images Preprocessing for Optical Character Recognition in Video Frames Reconstructed from Compromising Electromagnetic Emanations from Video Cables

Abstract

This work presents an analysis of the performance of two different preprocessing techniques for improving the quality of video frames previously reconstructed from compromising electromagnetic emanations produced by video interfaces connecting external monitors. Each technique depends on parameters such as white and black thresholds, kernel size of blur filter, contrast gain, brightness level, among others, and for this reason, an optimization process is applied by implementing an evolution $strategy$.