Development of a Computationally Enhanced Micro-Spectrometer Using Disordered Frosted Cellophane: Applications in Biotechnological Analysis

Abstract

This study introduces an innovative low-cost micro-spectrometer designed using disordered frosted cellophane, enhanced by advanced computational reconstruction techniques. The primary challenge in the development of cost-effective spectrometric devices lies in maintaining high accuracy and reliability amidst external and internal noise influences during spectral data reconstruction. To address this, we have integrated Local Weighted Scatterplot Smoothing (LOWESS) with the Singular Value Decomposition (SVD) algorithm to refine the stability and accuracy of the spectral curve solutions. Our comparative analysis reveals that spectra reconstructed using the LOWESS-optimized SVD algorithm demonstrate minimal peak shifts of 5.58 nm and Full Width at Half Maximum (FWHM) deviations of 2.51 nm when compared to those obtained from conventional commercial spectrometers. These findings indicate that our computationally enhanced micro-spectrometer achieves spectral resolutions on par with traditional, more expensive commercial devices. This approach not only reduces the cost of spectrometric analysis but also opens new avenues for deploying spectrometric solutions in resource-constrained settings, thereby expanding the potential applications of biotechnology in environmental monitoring, medical diagnostics, and quality control processes. This study exemplifies how computational innovation can bridge the gap between cost-efficiency and high-performance in the development of biotechnological tools.