Limited by the current technological development level of infrared detectors, it is difficult for the infrared multispectral imaging systems to meet the requirements of practical engineering application indicators such as low structural complexity and high transmittance at the same time. In order to promote the engineering application level of computational imaging technology in the infrared spectrum, this paper proposes a snapshot computational spectral imaging technology scheme based on reflective coding mask; this scheme utilizes the built-in reflective coding mask in traditional imaging system: Firstly, complete the design and molding of the coding mask according to the index requirements, complete the information calibration of the coding mask in the spatial dimension and the spectral dimension by building a high-precision spectral calibration module, and then combine the compressive sensing image restoration algorithm according to the calibration results, finally realize the Imaging and spectral data extraction of targets. Laboratory indoor imaging shows that this scheme can realize broad-spectrum imaging of indoor targets and extraction of multi-spectral information of targets. Compared with the technical scheme of traditional optical filter and grating infrared spectrometer, it has the technical advantages of simple structure and high transmittance.
With the development of computational imaging technology, computational spectral imaging technology based on coding masks has attracted the attention of researchers due to its advantages in obtaining multi-dimensional information. However, the current research on computational spectral imaging technology in the infrared band is relatively scarce. In the research of this paper, in order to solve the problem that the signal-to-noise ratio of the spectral calibration in the infrared spectral calibration of the coding mask is too low, an innovative spectral calibration scheme based on the reflective coding mase is proposed. Firstly, according to the infrared band and spectral resolution index, the mask surface model coding technology is used to design the surface of the mask, according to the above scheme, the reflective mask can be processed, then the spectral modulation of the specific speckle field distribution can be finished base on the corresponding infrared point light source generation module; secondly after completing the spectral calibration of the key points on the encoding mask surface ,the spectral calibration signal reconstruction of the remaining pixels of the encoding mask can be finished base on the improved bilinear interpolation algorithm, this work can greatly save the spectrum calibration time; finally, the effect of the speckle field algorithm base on compressed sensing is used to verify the effect of the spectrum calibration structure. The results show that, compared with the traditional spectrum calibration technology, the high-precision infrared spectrum modulation and calibration technology based on the reflective coding mask used in this paper can greatly improve the signal-to-noise ratio of the spectrum calibration signal and save the time required for the spectrum calibration.