Application of blind signal processing for noise suppression in a laser speckle triangulation

In engineering and manufacturing often raises the issue of evaluation of the geometric parameters for solving problems of technical diagnostics and nondestructive testing. One of the most recent applicable methods for noncontact measurement of the geometric profile of the object is a method of laser triangulation. The main problem with this method is the presence of noise speckles due to which there is a deterioration of accuracy. Complexity to deal with such noise is that they are random and depend on the surface being measured. The use of classical methods of filtration used in image processing tasks, in this case is impossible, since it leads to increased uncertainty. The way out of this situation is the application of blind treatment, in particular the methods of blind signal separation, which do not distort the measurements.