Abstract:Emerging computational spectrometers eliminate the need for optical and mechanical components, and achieve portable, real‐time spectral measurements. This new design enables the miniaturization and cost reduction of spectrometers, making the popularization of hyperspectral camera possible. This review summarizes the principles and recent developments of computational spectrometers, and discusses the future opportunities and challenges. The extensive applications of spectrum analysis across various fields have rendered the traditional desktop spectrometers unable to meet the market demand for portability and instantaneity. Reducing the size of spectrometers has become a topic of interest. Based on this trend, a novel type of computational spectrometer is developed and has been widely studied owing to its unique features. Such spectrometers do not need to integrate complex mechanical or optical structures, and most of them can achieve spectrum analysis by the properties of the material itself combines with the reconstruction algorithm. Impressively, a single‐detector computational spectrometer has recently been successfully realized based on in situ modulation of material properties. This not only enables the further miniaturization of the device, but also means that the footprint‐resolution limitation which has always existed in the field of hyperspectral imaging has been broken, opening a new era of image analysis. This review summarizes the classifications and principles of various spectrometers, compares the spectrum resolution performances of different types of spectrometers, and highlights the progress of computational spectrometers, especially the revolutionary single‐detector spectrometer. It is expected that this review will provide a positive impact on expanding the boundary of spectrum analysis and move hyperspectral imaging forward.

Undersampling Raster Scans in Spectromicroscopy for reduced dose and faster measurements

High-content, high-throughput imaging of single-cell metabolism by sub-sampled hyperspectral SRS microscopy

Advances in sparse dynamic scanning in spectromicroscopy through compressive sensing

Optimization of Image Quality and Acquisition Time for Lab-Based X-ray Microtomography Using an Iterative Reconstruction Algorithm

Compressive Sensing for Dynamic XRF Scanning

Compressed sensing for rapid tabletop X-ray absorption spectroscopy

Hyperspectral X-Ray Imaging For Nanometrology

Assessment of Subsampling Strategies in Microspectroscopy of Environmental Microplastic Samples

Towards scanning nanostructure X‐ray microscopy

Inpainting vs denoising for dose reduction in scanning-beam microscopies

Optical synthetic sampling imaging: Concept and an example of microscopy

The advantages of sub-sampling and Inpainting for scanning transmission electron microscopy

Accelerated Nano-Optical Imaging through Sparse Sampling

Low-latency Visual Previews of Large Synchrotron Micro-CT Datasets

Hyperspectral acquisition with ScanImage at the single pixel level: Application to time domain coherent Raman imaging

On-the-fly Data Assessment for High Throughput X-ray Diffraction Measurement

Emerging Computational Micro‐Spectrometers — From Complex System Integration to Simple In Situ Modulation

Robust retrieval of material chemical states in X-ray microspectroscopy

A High-Throughput Imaging Spectrometer Based on Over-Scanning

Fast X-ray micro-tomography for low-Z materials

Methods for energy dispersive x-ray spectroscopy with photon-counting and deconvolution techniques