A procedure for determination of the microstructural probabilistic characteristics from micrographs and their impact on subsequent results

In this study a methodology for the experimental determination of probabilistic characteristics induced by injection molding short-fibre reinforced composites is presented. This approach based on pattern recognition algorithm fibre detection two-dimensional micrograph. Since fibres are elliptic shape when dealing with arbitrary oriented cross-section, method ellipses used. The stochastic microstructure described probability density functions length, diameter, and orientation, which extracted from For accurate results, sample preparation major significance, otherwise microstructural not captured correctly. result orientation over cross-section shows typical layered structure microstructure. Furthermore, function diameter present as expected normal distribution. general, results in good agreement literature other scientific works.