Deep Learning-Based Miniaturized All-Dielectric Ultracompact Film Spectrometer

Abstract

Conventional benchtop spectrometers with bulky dispersive optics and long optical path lengths display limitations where the significance of miniaturization, real-time detection, and low cost transcend the ultrafine resolution and wide spectral range. Here, we demonstrate a miniaturized all-dielectric ultracompact film spectrometer based on deep learning working in the single-shot mode. The scheme employs 16 spectral encoders with simple five-layer film stacks where merely the thickness of the intermediate high-index modulation layer is varied to realize unique encoded transmission spectra. Structural parameters as well as transmission spectra of the filters are predesigned to guarantee weak correlation and highly efficient encoding. Leveraging a trained reconstruction network, the absolute spectra of various nonluminous samples are successfully reconstructed excluding the emitting spectrum of the light source and the spectral response of the detector. The remarkable reconstructed spectral imaging result for the color board is presented and the reconstructed spectra match well with the measured ones for different patches using the identical network. We utilized the least number of spectral encoders ever since to guarantee efficient encoding, along with the single thickness-variant modulation layer, which shows potential for mass, rapid, large-area production by combining deposition with nanoimprint. Instead of the synthetic Gaussian line shape spectra, a training dataset composed of diverse spectrum types is adopted to achieve fine generalization of the trained reconstruction network. In addition, by retraining the neural network, the reconstruction network is modified to fit for the actual filter functions of the spectral encoders, thus better reconstruction performance. The proposed miniaturized spectrometer has great prospects in the fields of consumer electronics, environmental monitoring, and disaster prevention.

Miniature Spectrometer Based on Gold Nanorod-Polyvinylpyrrolidone Film

“…Moreover, the noise that exists in each step of the spectrometer operations may have great effects on the reconstruction results, including the noises introduced by the detector, stray light, and measurement error during the characterization stage. , Therefore, it is necessary to employ a reliable reconstruction algorithm that can solve underdetermined problems with high noise tolerance. Several reconstruction algorithms have been developed in prior studies, such as nonnegative least-squares, simulated annealing, sparse optimization, and machine learning. − In this report, the Tikhonov regularization algorithm with non-negative constraints was applied to obtain a good reconstruction performance. The reconstructed spectra were compared with the reference measured by a commercial spectrometer (Ocean Optics USB2000+UV–vis), where the root-mean-square error (RMSE) was calculated to evaluate the quality of the reconstruction.…”

Random broadband filters based on combination of metasurface and multilayer thin films for hyperspectral imaging

“…For the resonant narrowband filters, it is challenging to achieve high spectral resolution due to the limited number of spectral channels corresponding to the number of filters [13][14][15][16][17]. More efficient approaches utilize random broadband micro/nano filters, which could be achieved via quantum dot arrays [18,19], photonic crystal plate arrays [20,21], multilayer thin films filters [22][23][24], bandgap tunable nanowires [25,26] and metasurface arrays [27][28][29]. These alternatives encode the spectral information of incident light as a series of responses at different locations on detector.…”

Miniature computational spectrometer with a plasmonic nanoparticles-in-cavity microfilter array

“…Many types of microfilters, such as quantum dots 21 , 22 , photonic crystals 23 – 26 , plasmonic encoders 27 , plasmonic rainbow chips 28 , metamaterials 6 , 11 , 15 , 29 , 30 , liquid crystals 31 , random structures 32 , 33 , multilayer film filters and interference filters 12 , 34 – 38 , and 3D-printed microoptics 39 , have been demonstrated to be useful for computational spectrometry. Multilayer film filters have good flexibility in tailoring optical spectral responses 12 , 36 , 37 . However, their fabrication processes using many steps of deposition/patterning are less scalable in the fabrication of large-scale distinct filter arrays.…”