SRFS:Super Resolution in Fast-Slow Scanning Mode of SEM Based on Denoising Diffusion Probability Model

Zelin Wang,Rui Feng,Jikai Mao,Xingyu Tao,Deao Zhang,Xuliang Shan,Jianguang Zhou
DOI: https://doi.org/10.1109/tim.2024.3351247
IF: 5.6
2024-01-01
IEEE Transactions on Instrumentation and Measurement
Abstract:Scanning electron microscopy (SEM) is the most commonly used and important tool for material characterization, it provides two scanning mode, fast scanning mode obtain images rapidly, but these images are low resolution and lack of details, slow scanning mode obtain images with high resolution and rich details, but it is time-consuming. This limits the use of SEM, in order to solve the above trade-off problem, we propose a diffusion-based super resolution (SR) network SRFS. To the best of our knowledge, this is the first time of image super resolution in fast-slow scanning mode of SEM. Comparing to previous SR methods, SRFS can generate SR images with LPIPS of 0.1414 and FID of 0.59 which provide the best subjective vision quality and the most accurate image details. SRFS allow us to obtain high resolution SEM images with rich and accurate details fast and continuously. By using SRFS, we can achieve accurate process monitoring under SEM.
What problem does this paper attempt to address?