Machine Vision Applied to Backscattered Electron Imaging

Electron beam processing usually needs to be carried out in a vacuum environment. The influence of high temperature and high concentration of metal vapor makes it difficult for traditional machine vision observation systems to perform long-term observation. In this paper, a machine vision algorithm was used, a quadrupole plate sensor and image fusion algorithm were adopted, and the combination of software and hardware was realized. Finally, we solved the problems of scale imbalance and slice model scanning order in the imaging process and successfully realized clear observation of the surface in all directions. The backscattered electron signal imaging technology has been successfully applied to the electron beam processing process, which meets the online observation requirements of the electron beam processing process.