Sparse Particle-Beam Image Segmentation for Faster Substrate Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image segmentation techniques for raster-based imaging applications, such as electron microscopy, face challenges in achieving high throughput due to long data acquisition times and require computationally intensive methods that may lose valuable image information.
Innovation Solution
A method and system for segmenting features from sparse imaging data using a particle beam to impinge a substrate surface at sensing locations, measuring intensity values, calculating estimated intensities based on proximal locations, and solving optimization problems to segment images, incorporating penalty functions and noise characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sparsely sampled data is used to increase image acquisition speed, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent introduces an intermediary computational model that maps sparsely sampled sensor data to a complete image representation. This model acts as a mediator between the limited measurements and the desired full-resolution image, using physical constraints and optimization algorithms to reconstruct missing information without requiring dense sampling.
Solution Approach 2:
The patent transforms the problem from direct image reconstruction to solving for underlying physical parameters (such as electron interaction cross-sections and material properties) that generate the observed signals. By changing parameters from pixel intensities to physical quantities, the system can interpolate missing data more accurately.
2Productivity
If computational algorithms are used to construct complete images from partial data, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent segments the image reconstruction problem into distinct computational stages: forward modeling of electron interactions, optimization-based parameter estimation, and final image synthesis. This segmentation allows each stage to be optimized independently and enables parallel computation where applicable.
Solution Approach 2:
The patent performs preliminary computational actions by pre-calculating electron interaction cross-sections and sensitivity matrices before actual imaging. These pre-computed components are stored and reused during reconstruction, significantly reducing the computational burden during data acquisition and processing.
3Measurement precision
If noise reduction is applied to improve image quality, then measurement precision is improved, but loss of information increases
Solution Approach 1:
The patent applies different processing strategies to different regions of the image based on local characteristics. High-confidence regions receive minimal processing, while low-confidence regions undergo more aggressive reconstruction. This local differentiation preserves genuine image features while reducing noise in problematic areas.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for rapid image acquisition and accurate segmentation of substrate features, maintaining image quality and connectivity, reducing noise, and enhancing the efficiency of image reconstruction from partial data.
Implementation Method 1
impinging the substrate surface with a particle beam at each of a plurality of sensing locations
Implementation Method 2
measuring at each of the plurality of sensing locations, by each of at least two particle sensors, an intensity value associated with post-impingement particles resulting from the impinging
Data Source
AI summary
Described are systems and methods for segmenting images which comprise impinging a substrate surface with a particle beam at each of a plurality of sensing locations which define a subset of locations within an area of interest of the substrate surface. An intensity value associated with post-impingement particles resulting from the impinging is measured and the measured intensity based on the intensity value of the sensing location is calculated. For each of a plurality of estimated locations which define a further subset of said area of interest and a corresponding estimated intensity based on at least one of the following corresponding to one or more locations proximal to the estimated location is calculated. The plurality of estimated locations is each segmented based on the corresponding estimated intensity, each of the sensing locations, and based on the corresponding measured intensity, to correspond to one of the plurality of features.