# Raman Signal Extraction from CARS Spectra Using a Learned-Matrix Representation of the Discrete Hilbert Transform

Source: [Raman Signal Extraction from CARS Spectra Using a Learned-Matrix Representation of the Discrete Hilbert Transform](<https://www.sota2.com/research/paper/68fa421a1e209131ce8ffb22>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T03:37:08.074Z

Authors: Charles H. Camp Jr

Published: 2022-04-01T19:05:33.000Z

## Abstract

Removing distortions in coherent anti-Stokes Raman scattering (CARS) spectra due to interference with the nonresonant background (NRB) is vital for quantitative analysis. Popular computational approaches, the Kramers-Kronig relation and the maximum entropy method, have demonstrated success but may generate significant errors due to peaks that extend in any part beyond the recording window. In this work, we present a learned matrix approach to the discrete Hilbert transform that is easy to implement, fast, and dramatically improves accuracy of Raman retrieval using the Kramers-Kronig approach.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Raman Reconstruction on Methanol Real-world CARS sample](<https://www.sota2.com/research/sota/raman-reconstruction-on-methanol-real-world-cars-sample.md>): Raman Reconstruction; Methanol Real-world CARS sample; MSE: 0.0315.

- [Raman Reconstruction on Synthetic Raman Spectra (test)](<https://www.sota2.com/research/sota/raman-reconstruction-on-synthetic-raman-spectra-test.md>): Raman Reconstruction; Synthetic Raman Spectra (test); MSE: 0.0814.

- [Raman Reconstruction on Acetone Real-world CARS sample](<https://www.sota2.com/research/sota/raman-reconstruction-on-acetone-real-world-cars-sample.md>): Raman Reconstruction; Acetone Real-world CARS sample; MSE: 0.2663.

- [Raman Reconstruction on DMSO Real-world CARS sample](<https://www.sota2.com/research/sota/raman-reconstruction-on-dmso-real-world-cars-sample.md>): Raman Reconstruction; DMSO Real-world CARS sample; MSE: 0.0157.

- [Raman Reconstruction on Ethanol Real-world CARS sample](<https://www.sota2.com/research/sota/raman-reconstruction-on-ethanol-real-world-cars-sample.md>): Raman Reconstruction; Ethanol Real-world CARS sample; MSE: 0.0351.

- [Raman Reconstruction on Isopropanol Real-world CARS sample](<https://www.sota2.com/research/sota/raman-reconstruction-on-isopropanol-real-world-cars-sample.md>): Raman Reconstruction; Isopropanol Real-world CARS sample; MSE: 0.0607.

- [Raman Reconstruction on Toluene Real-world CARS sample](<https://www.sota2.com/research/sota/raman-reconstruction-on-toluene-real-world-cars-sample.md>): Raman Reconstruction; Toluene Real-world CARS sample; MSE: 0.1129.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2204.00666>)

[PDF](<https://arxiv.org/pdf/2204.00666.pdf>)
