Comparison of denoising methods in fault diagnosis with Gabor order tracking
B. L. Su, Xialan He · 2017
Based on simulations of denoising of Gabor coefficient spectrum with Gaussian dual windows, threshold filter denoising and morphological Filter denoising methods have been implemented. With conventional Gabor order tracking algorithm, Gaussian window applied as analysis window and synthesis window, a trail-and-error threshold filter denoising method reconstructed original signal without noise, the choose of value of threshold filter is a waste of time. However, morphological Filter denoising is put forward as a alternative vision solutions, flood fill algorithm looks for all nodes in the array that are connected to the start node by a path of the target color and keep this area unchanged. Mask operation sets unselected region to zero, simulations indicate that should be rapid and effective.