Method of image enhancement by splitting-signals

The method of tensor representation of an image with respect to the Fourier transform and its application for image enhancement is described. The method is based on the fact that a two-dimensional (2D) image can be represented by a set of 1D signals that split the 2D Fourier transform of the image into different groups of frequencies. Each splitting-signal carries information of the spectrum in a specific group. The processing of the image is reduced to processing splitting-signals. The effectiveness of such an approach is illustrated through processing the image by the α-rooting method of enhancement. We propose to enhance the image by processing only one or a few splitting-signals, to achieve image enhancement which, in many cases, can exceed enhancement by the α-rooting method. The selection of such splitting-signals is described.