Files
Abstract
One among the non-exhaustive list of applications where image processing is used is Nanotechnology. To visualize and characterize the nanostructures, high resolution Scanning Electron Microscope (SEM) and Atomic Force Microscopy (AFM) are ideally used. One of the crucial steps in image processing the nanostructures is image segmentation. There are various ways to attain segmentation. One way is the statistical approach. In this approach, we fit the foreground and background histogram data to the Gaussian curves and the parameters are found. The optimal threshold is found by solving a set of equations using the six parameters of the mixture of two Gaussian distribution. We have used Expectation-Maximization (EM), kmeans, combination of EM and kmeans, minimum distance method to determine the optimal threshold using the statistical approach. The comparisons of all these techniques are thoroughly studied for the simulated data as well as the real nanostructure images. We have also proposed a faster, preparation insensitive objective method for measuring the length, orientation and density of the single walled carbon nanotubes and the diameter of the nanoparticles using image processing techniques.