Search

It is proposed to use weight coefficients of the bilateral filter to calculate the measure of similarity of image regions in optical flow calculation algorithms. The efficiency of using this measure is studied by an example of a three-dimensional recursive search algorithm. Weight coefficients are used for calculating the optical flow with subpixel accuracy by the Lucas–Kanade algorithm. A possibility of reducing the optical flow calculation error by means of using the proposed approaches is demonstrated by an example of processing of various types of images. A method of choosing the parameters of the weight functions of the bilateral filter is described and analyzed.

An algorithm of fatigue crack detection in optical images taken in fatigue tests of materials is proposed and tested. The algorithm is designed for automation of measurements of the crack propagation parameter and tracing the crack tip position in the course of cyclic loading for the purpose of shifting the optical system with respect to the examined sample surface to the “region of interest.” It is found that the coordinates of the image fragment containing the crack can be determined with a mean error of 1.93% of the total size of the raster. Testing of the algorithm on model images shows that the mean error of determining the crack tip position is smaller than 56 pixels.

An approach to automatic determination of operation parameters of a hierarchical 3D recursive search (3DRS) algorithm is proposed and tested. A comparative study of the computational speed and noise immunity of the 3DRS algorithm used for constructing displacement vector fields, including the use of the Gaussian pyramids (hierarchical search), is performed. It is shown that application of the hierarchical 3DRS algorithm with operation parameters determined in this work can substantially increase its noise immunity and reduce the computational cost.

An algorithm of constructing displacement vector fields, based on the incremental approach to estimating displacements of image fragments, is proposed. Application of this algorithm for processing model and experimental optical images is analyzed. Noise immunity of the incremental algorithm for determining displacements is compared with that of the combined algorithm developed by the authors. It is shown that the proposed incremental algorithm ensures a more noise-immune estimate of displacements in processing images of a deformation relief essentially varying in time.

The use of the Bezier surface as a smoothing filter for vector fields is studied. Advantages of the Bezier surface over the moving average filter are determined. The algorithm of adaptation of the filter aperture size is modified. The efficiency of adaptation of the aperture size in vector field processing in the problem of estimating the material strain by the method of digital image correlation is considered.

The influence of adaptive bilateral filtration on the calculation of the fractal dimension of optical images is estimated. Filtration parameters are chosen on the basis of various criteria. The use of the maximum dispersion of the probability as the most effective parameter of estimating the image filtration quality is considered.

Various filtration methods are compared by an example of processing of model and experimental vector fields. A criterion for estimating the filtration efficiency is proposed. It is demonstrated that combined vector filters ensure more efficient filtration than other popular methods.

The influence of various methods of obtaining surface images on the calculated value of their fractal dimension as a quantitative characteristic of the surface state is studied. It is demonstrated that images obtained both by a scanning electron microscope and by a photocamera are characterized by a noticeable noise level, which alters the behavior of the fractal dimension. Various methods of correction of the observed effect are discussed.