A team of researchers has applied deep learning to scanning electron microscopy to develop a super-resolution imaging technique. It can convert a low-resolution microstructure image into a super-resolution image.
In modern-day materials research, scanning electron microscopy images play a crucial role in developing new materials, from microstructure visualization and characterization to numerical material behavior analysis. However, due to hardware limitations, acquiring high-quality microstructure image data may be exhaustive or highly time-consuming. It may affect the accuracy of subsequent material analysis; therefore, it is paramount to overcome the technical limitations of the equipment.
The team developed a faster and more accurate microstructure imaging technique using deep learning. Using a convolutional neural network, the researchers enhanced the resolution of the existing microstructure image. The new approach reduces the imaging time significantly (256x) compared to the conventional system. Accurate microstructure characterization and finite element analysis can restore the morphological details of the microstructure.
In addition, super-resolution imaging verified that the morphological details of the microstructure could be restored with high accuracy through microstructure characterization and finite element analysis. Through the EBSD technique developed in this study, researchers anticipate the time it takes to develop new materials will be drastically reduced.
Related Content: AI Tool Analyzes Microscopy Images