Hybrid edge detection method on TEM images of nanoparticles

Abstract

General application of the fundamental part of digital image processing is widely used in various fields of science. With the rapid development of microscopic imaging technology, the need for the robust segmentation of nanoparticles increased significantly. Due to large differences in the number and form of nanoparticles in TEM images, most of the existing methods are only capable of segmenting certain types of nanoparticles. Taking into account the characteristics of nanoparticle materials, this paper gives some suggestions for a new edge detection algorithm. As an assessment of similarity of the proposed algorithm in the paper, the analysis was carried out both through the spatial and through the amplitude resolution of the digital image. With the aim of obtaining additional accurate results, the analysis was conducted through images of nanoparticles of different degrees of magnification. Images of this type provide different levels of noise in digital image, therefore the final analysis gave an assessment of the degree of resistance to the proposed edge detection algorithm for noise occurring in nanoparticle images. All the obtained results indicate a high degree of utility of the processed TEM images, both in spatial and amplitude domain.