Terahertz single-pixel imaging method and terahertz single-pixel imaging system

Abstract

The invention provides a terahertz single-pixel imaging method and a terahertz single-pixel imaging system. The method comprises the steps of modulating terahertz waves through a mask, and finally obtaining a reconstructed image. The mask is obtained through the following steps: obtaining an original image, and performing image coding on the original image through a preset mask to obtain a presetcoding result; inputting the preset coding result into a self-coding network to obtain a preset decoded image; constructing a minimum difference network according to the preset decoded image and the original image, optimizing the minimum difference network through iterative learning, and judging whether the minimum difference network is converged or not: outputting a self-encoding network parameter if the minimum difference network is converged, and learning again if the minimum difference network is not converged; and obtaining the mask according to the self-encoding network parameter. Through a deep learning network, the problem of contradiction between imaging time and imaging quality of terahertz single-pixel imaging is solved, and high-quality reconstruction of the image under a low sampling ratio is realized while the imaging time is shortened.

Description

technical field [0001] The present invention relates to the field of terahertz single-pixel imaging, in particular to a terahertz single-pixel imaging method and system thereof. Background technique [0002] Terahertz waves (frequency 1-10THz or wavelength 30μm-3000μm) refer to coherent electromagnetic radiation between the microwave band and the optical band. Terahertz waves can penetrate a variety of optically opaque materials, such as semiconductors, plastics, and paper, and can identify hidden objects in them with high contrast, such as metals, internal structures of objects covered by external materials, etc. Furthermore, terahertz photon energies are very low, making this radiation safe for use on both the human body and sensitive samples. Despite showing great application potential, terahertz imaging technology has not been practically applied, the main reason being the high equipment cost. Unlike the field of visible light, the development of terahertz pixelated de...

AI Technical Summary

Problems solved by technology

[0004] Some scholars proposed to combine 180nm VO2 film and Hadamard mask coding to realize terahertz super-resolution image reconstruction, and use compressed sensing algorithm to realize image reconstruction under 20% modulation mask. longer time

There are also people who use 220um silicon-based graphene modulators and Fourier fringes to realize sub-wavelength terahertz image reconstruction. Taking advantage of image sparseness, image reconstruction under 10% modulation mask is realized by collecting low-frequency coefficients and inverse Fourier transform, but due to Missing high-frequency information, the picture is blurry

In the prior art, there is also the use of silicon total internal reflection prisms and Hadamard masks to achieve near-real-time terahertz single-pixel video, and the use of under-sampling technology to reduce sampling time by taking advantage of the sparse characteristics of the Hadamard field, but there is also the problem of deterioration of reconstructed images

[0005] To sum up, in the prior art, there is a common problem of contradictory imaging speed and imaging quality for terahertz single-pixel imaging, and the relationship between imaging speed and imaging quality cannot be balanced.

Images