FTIR Spectral Matching Using Normalized Local Change
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating spectral similarity in Fourier-transform infrared (FTIR) spectroscopy are susceptible to offsets, baseline sloping, and variations in absorbance, leading to inaccurate comparisons of FTIR spectra.
Innovation Solution
A novel method using normalized local change (NLC) values to calculate spectral similarity, focusing on local characteristics of spectra to reduce bias from large peaks and tolerate variations in absorbance, employing a range and floor value to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Euclidean distance is used to calculate spectral similarity, then the calculation is simple and direct, but the method is susceptible to offset, baseline sloping, and variations in absolute absorbance values
Solution Approach 1:
The patent transforms the spectral comparison from using absolute absorbance values to using normalized local change (NLC) values. This parameter transformation makes the comparison invariant to offset, baseline sloping, and scaling variations while maintaining sensitivity to spectral shape differences. The NLC is calculated as the normalized difference between adjacent absorbance values, effectively removing the influence of absolute value variations.
Solution Approach 2:
The patent introduces an intermediary transformation step that converts raw absorbance spectra into NLC spectra before comparison. This intermediary representation (NLC values) serves as a mediator that eliminates the harmful effects of offset and baseline variations while preserving the essential spectral features for identification.
2Reliability
If correlation coefficient or dot product methods are used to tolerate offset and baseline variations, then robustness against artifacts improves, but these methods give excessive weight to larger peaks creating bias
Solution Approach 1:
The patent segments the spectral information by focusing on local changes (differences between adjacent points) rather than treating the entire spectrum as a single vector. This segmentation into local derivative-like features allows each region to contribute equally to the similarity calculation, preventing large peaks from dominating the overall comparison.
Solution Approach 2:
The patent changes the parameter from absolute absorbance values to normalized local change values. This transformation equalizes the contribution of different spectral regions by emphasizing shape information while suppressing the influence of peak magnitude, thereby eliminating the bias toward large peaks that plagues correlation-based methods.
3Reliability
If spectral comparison methods are made more tolerant to artifacts through preprocessing and scaling, then reliability improves, but the complexity of the method increases
Solution Approach 1:
The patent achieves robustness through a single parameter transformation (calculating normalized local changes) rather than requiring multiple preprocessing steps like baseline correction, normalization, and scaling. This unified approach simplifies the overall method while maintaining tolerance to offset, baseline sloping, and absorbance variations.
Solution Approach 2:
The patent extracts only the essential information needed for spectral identification by focusing on local changes rather than processing the entire spectrum with multiple transformations. This extraction of key features (NLC values) eliminates the need for complex preprocessing while preserving the essential spectral characteristics.
Data Source
AI summary
Spectra matching is widely used in various applications including the search for a spectrum of an unknown or subject material, chemical, or compound in an existing spectral database and quality control by means of comparing the spectra of products with standards. New systems and methods are described for identifying an unknown compound by calculating the similarities of Fourier-transform infrared (FTIR) spectra of organic compounds. The systems and methods incrementally calculate the spectral similarity based on the local spectral shapes. This reduces the bias caused by uneven weighing of large or broader peaks. In addition, the new systems and methods tolerant to the common issues in spectra matching including baseline offset, baseline sloping, and deviations in wavenumber axis alignment, suggesting its robustness and practical applicability.