Low-Cost Optoelectronic System for IR Spectral Acquisition Based on Band Selection

Infrared (IR) spectroscopy provides insights beyond the visible spectrum. Current efforts to reduce spectrometry costs primarily target the visible range. Deep learning (DL) algorithms could optimize optoelectronic designs and significantly reduce costs for specific IR applications. The study proposes a low-cost optoelectronic system for IR spectral acquisition by optimizing IR sensors through band selection, aiming for a 98.07% cost reduction compared with commercial IR sensors. Promoting the decentralization of IR spectroscopy is particularly beneficial for resource-constrained settings lacking general-purpose IR spectrometers.