The physical and mechanical properties of a polycrystalline material depend on its microstructure characteristics such as the size and morphology of grains. In practice, different imaging methods are used to visualize the grain structure of such materials. To analyze microstructural changes in case of applied stress and to predict its performance in a given application, the quantitative information about the grain structure must be taken into account. In this work, an effcient and reproducible algorithm, which automatically detects grains in different types of microstructure images, is proposed. Due to the diversity between the analyzed images and a limited number of labeled data, a clustering patch-based approach is followed. The algorithm aims to distinguish between patches in homogeneous grain areas and those lying on grain boundaries through Gaussian Mixture Modeling. The identified groups of grain patches are used to create the seed image for a Seeded Region Growing algorithm, enabling nally a pixelwise image segmentation.

Electrical measurement of degradation in metal films induced by high thermo-mechanical stress is not possible. Therefore, different imaging methods are used in practice to visualize the changes in material microstructure. In this work, SEM (Scanning Electron Microscopy) cross section images of the metal layer of interest that illustrate the fatigue induced degradation and material microstructure are analyzed. We propose an unsupervised algorithm for detection and quantitative assessment of the damage in mentioned images. In the first stage of the algorithm, the metal layer of interest is extracted from the background using k-Means method. In the second stage, the non-local means (NL-means) denoising method with automatically computed standard noise deviation followed by post-processing and k-Means is used to detect the damage patterns. Visual and quantitative evaluation of results reveals that the algorithm provides robust and plausible results.