A cyber-physical system using thin film filters and Bayesian inference to accurately estimate light wavelengths with 99% accuracy, suitable for vision systems
https://nu.technologypublisher.com/files/sites/mark-saulich-1280x1600.jpg
Cone cells are natural photoreceptor cells in the retinas of vertebrate eyes. They respond differently to light of different wavelengths and thus responsible for color vision, a sensation in our brain in response to the incidence of mixed wavelength values (i.e. non-monochromatic light) incident on the cone cells. Similarly, this nanotechnology/AI-enabled system will be able to accurately identify colors from different sources.
In this invention, Northeastern University researchers presented a cyber-physical system for accurately estimating the wavelength of any monochromatic light in the range 325−1100 nm by applying Bayesian inference on the optical transmittance data from a few low-cost, easy-to-fabricate thin film “filters” of layered transition metal dichalcogenides (TMDs) such as MoS2 and WS2. Wavelengths of tested monochromatic light could be estimated with only 1% estimation error over 99% of the stated spectral range, with lowest error values reaching as low as a few ten parts per million (ppm) in a system with only 11 filters. By stepwise elimination of filters with the least contribution toward accuracy, mean estimation accuracy of ∼99% could be obtained even in a two-filter system. Furthermore, this invention provides a statistical approach for selecting the best “filter” material for any intended spectral range based on the spectral variation of transmittance within the desired range of wavelengths. Researchers demonstrate that calibrating the data-driven models for the filters from time to time overcomes the minor drifts in their transmittance values, which allows for use of the same filters indefinitely. This work not only enables the development of simple cyber-physical photodetectors with high accuracy color estimation, but also provides a framework for developing similar cyber-physical systems with drastically reduced complexity.