Visible Diffuse Reflectance Spectroscopy Coupled with PCA–LDA for Non-Destructive Paper Age Prediction

Darvishi Ghazanchi, Reza; Tayyeb, Zohair; Tavassoli, Seyed Hassan

doi:10.48308/piadm.2026.244285.1021

Visible Diffuse Reflectance Spectroscopy Coupled with PCA–LDA for Non-Destructive Paper Age Prediction

Document Type : Original Article

Authors

Reza Darvishi Ghazanchi ¹

Zohair Tayyeb ²

Seyed Hassan Tavassoli ¹

¹ Laser and Plasma Research Institute, Shahid Beheshti University, Tehran, Iran

² Department of Persian Language and Culture, University of Tehran

10.48308/piadm.2026.244285.1021

Abstract

Age prediction of paper samples belonging to the same commercial brand but manufactured in different years was carried out using non-destructive Visible Diffuse Reflectance Spectroscopy (DRS) combined with PCA–LDA-based chemometric analysis. Initially, the classification model was trained using the reflectance spectra of paper samples collected from different production years. The performance of the developed model was subsequently evaluated using unknown samples that were not included in the training stage. Blind test results demonstrated that the model was capable of correctly predicting the age of 50% of the unknown samples. Considering that the temporal difference between some samples was as small as one year, this level of prediction accuracy can be regarded as significant. The findings indicate that visible diffuse reflectance spectroscopy, coupled with chemometric techniques, has considerable potential as an accessible, cost-effective, and non-destructive approach for paper age prediction. Unlike previous studies that commonly employed temporal intervals of four to five years, the present work successfully investigated naturally aged paper samples with intervals as small as one year.

Keywords

Paper Age Prediction

Visible Diffuse Reflectance Spectroscopy

PCA–LDA

Chemometric Analysis

Non-Destructive Analysis

Paper Classification