“A review on spectral data preprocessing techniques for machine learning and quantitative analysis” (2025), by Chunsheng Yan. Chemistry research paper. Original abstract below, unchanged. No free PDF was found.

A review on spectral data preprocessing techniques for machine learning and quantitative analysis

Chunsheng Yan · May 29, 2025 · Chemistry

Abstract

Spectroscopic techniques are indispensable for material characterization, yet their weak signals remain highly prone to interference from environmental noise, instrumental artifacts, sample impurities, scattering effects, and radiation-based distortions (e.g., fluorescence and cosmic rays). These perturbations not only significantly degrade measurement accuracy but also impair machine learning-based spectral analysis by introducing artifacts and biasing feature extraction. This review provides a systematic evaluation of critical spectral preprocessing methods-encompassing cosmic ray removal, baseline correction, scattering correction, normalization, filtering and smoothing, spectral derivatives, and advanced techniques like 3D correlation analysis-highlighting their theoretical underpinnings, performance trade-offs, and optimal application scenarios. The field is undergoing a transformative shift driven by three key innovations: context-aware adaptive processing, physics-constrained data fusion, and intelligent spectral enhancement. These cutting-edge approaches enable unprecedented detection sensitivity achieving sub-ppm levels while maintaining >99% classification accuracy, with transformative applications spanning pharmaceutical quality control, environmental monitoring, and remote sensing diagnostics.

Source page DOI: 10.1016/j.isci.2025.112759