Spectrum Processing Device for High S/N Phase Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing spectrum processing methods for scanning electron microscopes face challenges in accurately identifying the composition of phases due to low signal-to-noise (S/N) ratios in spectrum imaging data, particularly when measurement time is short.
Innovation Solution
A spectrum processing device and method that acquire spectrum imaging data, compare pixel spectra with a representative spectrum, and extract multiple pixel spectra based on the comparison results to generate a phase spectrum with a high S/N ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If measurement time is extended to improve S/N ratio, then composition identification accuracy is improved, but productivity decreases
Solution Approach 1:
The patent performs preliminary selection of representative spectra and extraction of candidate pixel spectra before final composition identification. By pre-processing the spectrum imaging data to identify and extract high-quality spectra that represent each phase, the system prepares optimized input data for composition analysis, enabling accurate identification without requiring extended measurement time.
Solution Approach 2:
The patent creates a phase spectrum by combining multiple extracted pixel spectra that represent the same phase. Instead of relying on a single representative spectrum, the system generates a composite phase spectrum that aggregates signal information from multiple pixels, effectively increasing the S/N ratio through signal accumulation without requiring longer measurement time at each pixel.
2Adaptability or versatility
If representative spectrum is selected from spectrum imaging data, then phase distribution mapping is enabled, but S/N ratio of the representative spectrum remains low
Solution Approach 1:
The patent merges multiple pixel spectra that correspond to the same phase by combining their signal information to generate a phase spectrum. This consolidation of spectral data from multiple pixels increases the S/N ratio of the resulting phase spectrum, enabling more accurate composition identification while maintaining the phase distribution mapping capability provided by the representative spectrum approach.
3Measurement precision
If multiple pixel spectra are extracted and combined, then S/N ratio of phase spectrum is improved, but device complexity increases
Solution Approach 1:
The patent segments the spectrum imaging data by identifying and extracting pixel spectra that belong to different phases through comparison with representative spectra. This segmentation approach organizes the complex data into phase-specific groups, making the subsequent combination process more systematic and manageable, thereby reducing the perceived complexity of the overall processing workflow.
Solution Approach 2:
The patent employs a feedback mechanism where extracted pixel spectra are compared against representative spectra to verify phase membership before inclusion in the phase spectrum generation. This feedback loop ensures that only spectra representing the target phase are combined, improving the accuracy of the phase spectrum while providing a structured approach that simplifies the complex task of spectral data integration.
Data Source
AI summary
A spectrum processing device includes a data acquiring unit configured to acquire, for each of pixels expressing positions on a specimen, spectrum imaging data in which a pixel spectrum based on a signal from the specimen is stored; an extraction unit configured to compare, for each of the pixels, the pixel spectrum and a representative spectrum selected from the spectrum imaging data, and extract a plurality of the pixel spectra from the spectrum imaging data, based on a comparison result; and a spectrum generating unit configured to generate a phase spectrum, based on the plurality of the extracted pixel spectra.