Synthetic High-Resolution Image Generation via Sparse Representation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scanning electron microscope (SEM) technology faces limitations in achieving high resolution while maintaining fast scanning speeds, particularly in the chip inspection industry, where the inherent physical scanning resolution is linked to the output image resolution, hindering the ability to quickly scan large areas with high resolution.
Innovation Solution
The method involves generating a synthetic high-resolution image using a charged particle microscope by performing a dictionary learning process to create a mapping between low-resolution and high-resolution atoms, allowing for the enhancement of low-resolution images to achieve high-resolution quality without the need for high-resolution mode scanning, and detecting defects based on sparse representations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution mode scanning is used, then image resolution is improved, but acquisition time increases and field of view decreases
Solution Approach 1:
The patent creates a synthetic high-resolution image by copying and transforming low-resolution image data through dictionary learning and sparse representation. Instead of directly acquiring high-resolution data through time-consuming scanning, the system synthesizes high-resolution images from low-resolution inputs, effectively copying the desired high-resolution output without the time cost of direct high-resolution acquisition
Solution Approach 2:
The patent transforms the resolution parameter relationship by using dictionary learning to establish a mapping between low-resolution and high-resolution atomic representations. This parameter transformation allows the system to generate high-resolution images (finer second resolution) from low-resolution inputs (coarser first resolution) by changing the resolution parameter through computational processing rather than physical scanning
2Measurement precision
If high-resolution mode scanning is used, then image resolution is improved, but the field of view decreases
Solution Approach 1:
The patent adds a computational dimension to the imaging process by introducing dictionary learning and sparse representation transformations. This dimensional transformation allows the system to overcome the physical trade-off between resolution and field of view by operating in a transformed feature space where high-resolution details can be synthesized across the entire low-resolution field of view without the constraints of physical high-resolution scanning geometry
3Productivity
If low-resolution mode scanning is used, then acquisition time is reduced, but image resolution deteriorates
Solution Approach 1:
The patent introduces sparse representation and dictionary atoms as intermediary elements between low-resolution input and high-resolution output. These intermediaries act as a bridge, allowing information from low-resolution scans to be transformed and reconstructed into high-resolution images through the mediating process of sparse coding and dictionary-based synthesis
Data Source
AI summary
A method for generating a synthetic image of a region of an object, includes: generating, by a charged particle microscope, a charged particle microscope image of the region of the object; calculating a sparse representation of the charged particle microscope image; wherein the sparse representation of the charged particle microscope image comprises multiple first atoms; generating the synthetic image of the region, wherein the synthetic image of the region is formed from multiple second atoms; wherein the generating of the synthetic image of the region is based on a mapping between the multiple first atoms and the multiple second atoms; wherein the charged particle microscope image and the multiple first atoms are of a first resolution; and wherein the synthetic image of the region and the multiple second atoms are of a second resolution that is finer than the first resolution.