Monolithic RGB-NIR Perovskite Photodetector for Fused Multispectral Sensing and Imaging

The multispectral fusion of near-infrared (NIR) and visible red-green-blue (RGB) photons can enhance target identification under weak light conditions. Nevertheless, the crosstalk between NIR and RGB photons in a traditional pixelated sensor impedes their practical application, while using complex algorithms and optical filters would significantly increase the cost, form factor, and frame latency. In this work, a delicate monolithic RGBN (RGB-NIR) multispectral photodetector (PD) is proposed on the basis of perovskite materials without complicated algorithms or optical filters. The multispectral response toward selective RGBN signals in this monolithic PD pixels can be achieved by switching the polarity of the applied bias, affording the following benefits: I_on/I_off ratio of >10⁴, detectivity of >10¹⁰ Jones, crosstalk of -74 dB, and fast response with -3 dB > 10³ Hz. Moreover, proof-of-concept imaging of the iris and periocular with successful recognition in multispectral fusion further confirms its potential for identity authentication.