Bio-inspired spectral adaptive visual devices: A new paradigm for structure-defined functionality
Authors: BAO Youyou, ZHAO Yuhan, WU Daixuan, TIAN He
In recent years, the rapid development of artificial intelligence has driven the widespread deployment of visual systems in complex environments such as autonomous driving, security surveillance, and medical diagnosis. However, existing image sensors—such as CMOS and CCD devices—intrinsically suffer from the limitation of fixed spectral response. Especially in environments with strong glare, haze, or dust, external spectral conditions often severely mismatch the device's design range, leading to significant degradation in image quality and a sharp drop in target recognition accuracy. While algorithmic post-processing (such as color bias correction or background suppression) can mitigate these issues, algorithm approaches typically introduce computational latency and increased energy consumption, making them unsuitable for edge computing or high-speed scenarios. Achieving real-time adaptation to environmental spectral changes at the hardware level remains a major bottleneck in the intelligentization of visual systems. Zhao et al.[1] noted in their review that biomimetic sensing technologies are gradually breaking through the limitations of traditional sensors in complex environments, particularly in underwater visual systems, where structure-inspired approaches are increasingly important in defining device functionality. In 2024, Ouyang et al.[2] published a study in Nature Electronics proposing a biomimetic spectral adaptive visual device inspired by the spectral regulation mechanism of Pacific salmon. This design innovatively adopts a filterless, single-structure stacking approach, enabling the switching of the primary response spectral band within the device by adjusting the bias voltage, thereby defining spectral sensing functionality at the structural layer. The filterless stacking approach achieves 'depth-tunable' response through material heterostructures, not only avoids the volume and complexity issues of traditional multi-channel schemes but also constructs 'hardware-adaptive' sensing capabilities at the device level, opening up new avenues for the development of next-generation visual systems. The design of this spectra-adapted vision sensor draws inspiration from the efficient spectral adaptation strategies found in the biological world. Migratory salmon must transition from turbid inland freshwater to clear marine environments during their lifecycle, where the spectral compositions of visible and infrared light bands differ significantly as shown in Fig. 1(a). Salmon adjust the ratio of photoreceptor proteins with vitamin A1 and A2 structures to flexibly switch sensitivity between short-wavelength and long-wavelength light. The spectral sensitivity switching process is catalyzed by specific enzymes (such as Cyp27c1) without altering retinal structure, representing a typical 'intrinsic invariance with functional tunability'.
