Method for Assessing the Quality of Images Generated by Remote Monitoring Devices for Rolling Stock and Railway Tracks

The application of the two–dimensional Walsh-Hadamard transform is considered to evaluate the quality of images generated by remote monitoring devices for rolling stock and railway tracks. Purpose: to analyze the uneven distribution of the pixel spectrum intensity along the diagonal. This will facilitate a quantitative assessment of the degree of linear distortion in images of facilities belonging to “Russian Railways” JSC. Methods: image spectrum analysis by Walsh-Hadamard transformation. Results: the applicability of the proposed method for assessing the severity of linear distortions has been demonstrated. Practical significance: the proposed method can be applied for detecting various types of linear distortions, with the possibility of its subsequent implementation into automatic detection algorithms.