In this activity, students will learn about the use of image processing methods to analyze Scanning Electron Microscopy images using a technique known as Image Segmentation and the SEM Image Processing Tool. The purpose of this tutorial is demonstrate several methods for image masking: identifying regions of the image that are similar to each other and grouping them accordingly. The examples provided here show microscopy images of graphene grown on metal subtrates using chemical vapor deposition. Students will explore simple color masking, template matching, and k-means as a way to identify the regions of the images that correspond to graphene and those that correspond to the substrate.

Users of this teaching resource should view the companion lecture Image Segmentation for Graphene Images which reviews various clustering algorithms anworks for classifying pixels. Throughout the presentation, the strengths and weaknesses of different techniques is reviewed as is their relation to the structure of the data.

This resource contains multiple documents for download. The Download link above will download a bundle of all documents as a zip archive file. To view a list of all documents click on the Supporting Docs tab. This resource consists of the following items:

1. Procedure for using SEM Image Processing Tool
2. Example SEM images
3. Discussion questions for the tutorial

Researchers should cite this work as follows:

• Joshua A Schiller (2020), "SEM Image Segmentation Tutorial using SEM Image Processing Tool," https://nanohub.org/resources/33729.

  BibTex | EndNote

1. convolutional neural networks (CNN)
2. Gr-ResQ
3. machine learning
4. materials science
5. nanomanufacturing
6. nanoMFG Node
7. neural networks
8. scanning electron microscopy (SEM)