Automatic characterization of titanium dioxide nanotubes by image processing of scanning electron microscopic images

The presented article deals with the assessment of titanium dioxide nanotube microscopic images by means of image-processing methods. Inner diameter, wall thickness, and fraction of intertubular space are among the basic parameters characterizing the quality of nanostructured material. These parameters are especially important during the process of nanomaterial development. Nanostructures were prepared by electrochemical oxidation on the surface of Ti-6Al-4V alloy. Results of this process are greatly influenced by the chosen experimental conditions, so objective evaluation of experimental results is very important for the optimal setting of experimental parameters. Image-processing methods could be successfully applied to microscopic images to obtain the aforementioned objective characterization criteria. Various methods, such as object classification, image filtering, and mathematical morphology, could be taken into consideration with respect to a specific type of image data. In addition, the algorithm proposed uses the advantages of adaptive thresholding, watershed transform, low-pass filtering, mathematical morphology and cluster analysis. The nanotube wall thickness evaluation is suggested by two different approaches for the characterization of variable quality of the images processed.