| Automated Particle sizing of TEM Images with Particle-Based Labeling Algorithm |
| Paper ID : 1338-UFGNSM-FULL |
| Authors: |
|
elham tahmasebi * 1st floor, No-2, 9th Golbarg- north Golsar, Golriz st, Kharazmi st, Molasadra st, Tehran,Iran |
| Abstract: |
| Transmission electron microscopy (TEM) or X-ray diffraction is commonly used for the characterization and measuring particle size distributions. But manual analysis of the TEM image is extremely labor-intensive, while this method is limited by human inspections. Here, a new method, based on the image processing, is developed for automatic determination of the distribution of nanoparticles in TEM images. In the present method, the pre-processing procedures, including median filter, Wiener filter and histogram equalization, are used to improve the quality of the image, then, according to the auto-defined threshold, the image is binarized. In the consequent procedures, the particle-based labeling technique, which works according to the labeled connected component, has been applied to the binary image, to distinguish, separate and count the nanoparticles. The results were compared with the manual measurement, while showing a good agreement with the manual measurements. The merits of the method are: the ability to analysis accurately the TEM images with a very high number of particles, a low computational cost and fast processing, which makes it suitable to process large stacks of images, such as those generated by TEM, in a relatively short time. |
| Keywords: |
| Nano particles, TEM image, Image processing, Noise reduction, Particle labeling |
| Status : Paper Accepted (Oral Presentation) |
