Three-dimensional reconstruction method for image of scanning electron microscope

Abstract

The invention provides a three-dimensional reconstruction method for the image of a scanning electron microscope. The method comprises the steps of providing a micro-nano structure; obtaining the image of a scanning electron microscope of the micro-nano structure; according to the image of the scanning electron microscope, obtaining a rough graph of a three-dimensional structure; according to the image of the scanning electron microscope, constructing a reflection electron intensity database; according to the reflection electron intensity database, correcting the rough graph of the three-dimensional structure to obtain a corrected graph of the three-dimensional structure; obtaining the boundary information of the micro-nano structure and correcting the corrected graph of the three-dimensional structure based on the boundary information so as to obtain the fine graph of the three-dimensional structure.

Description

technical field [0001] The invention relates to the field of semiconductors and scanning electronic imaging, in particular to the three-dimensional reconstruction technology of microscopic images viewed from above by micro-nano devices. Background technique [0002] Scanning Electronic Microscopy (SEM) technology plays a very important role in microscopic imaging of micro-nano devices. It is currently the most widely used micro-nano structure observation and measurement technology in the field of integrated circuits. It has high resolution and imaging speed Fast, less damage to the device structure and other advantages. The imaging of surface topography by SEM is mainly based on the secondary electrons generated by the inelastic scattering of electrons and atoms, which escape from the surface and are then collected by the detector. Since the secondary electrons can only escape from the range of 10nm below the surface, SEM can better reflect the surface topography and is cur...

AI Technical Summary

Problems solved by technology

However, this method has many shortcomings, such as the SFS algorithm is accompanied by a huge amount of calculation and calculation time; the SFS algorithm is particularly sensitive to SEM images with noise; the deviation between various assumptions and reality in the process of solving ill-conditioned equations; The surface topography relationship expression has a large error for micro-nano structures with a size as low as tens of nanometers, etc.

[0006] Based on the above problems, it is urgent to provide a more accurate and efficient SEM-based three-dimensional reconstruction method, which can solve the SEM-based three-dimensional reconstruction problem when the size of the micro-nano pattern structure is reduced to tens of nanometers.

Images