| Sheared-beam imaging(Sheared-beam imaging,SBI)technology breaks through the traditional optical imaging concept,illuminating the target by three coherent laser beams which are laterally displaced at the transmit plane,and reconstructing the target image by returning laser speckle signals.The effect of atmospheric turbulence can be overcome to the maximum extent in terms of technical mechanism,and images of remote targets approaching near diffraction limit can be obtained without adaptive optics and imaging lenses.It has the advantages of high imaging resolution and high real time.However,in order to reconstruct high-resolution images,the detector array must have a large number of sensor elements,which are very difficult to implement.In order to reconstruct a clear image,the number of acquisitions required for data acquisition is high,which affects the imaging efficiency,and these key problems seriously restrict the practical application of SBI technology.Based on the coding and decoding principle of SBI system,this dissertation proposes a new system of SBI technique based on space-time sparse reconstruction by optimizing the data sampling and image reconstruction methods.The space-time sparse reconstruction algorithm is studied,which breaks through the constraint that emission shear length and the array spacing of detector elements are equal,reduces the requirement of imaging system,and improves data acquisition efficiency,speed and imaging quality.The main research contents of this dissertation include:1.The effect of low order phase distortion introduced by the initial phase of laser and atmospheric disturbance on image quality is analyzed through the study of SBI principle,which provides a theoretical basis for the suppression of target aliasing in the space-domain sparse construction method.The constraint conditions of the traditional three-beam imaging system are expounded,the parameters of the imaging system in various typical scenarios are calculated,the necessity of spatial sparse reconstruction algorithm in space-time domain is demonstrated,and a non-reference image quality evaluation method suitable for evaluating SBI speckle images is presented.2.Aiming at the problem of large amount of data acquisition of echo signal in SBI system,an image reconstruction method based on "L" type three-beam timedomain sparse sampling is proposed,and the sparse sampling of echo signals in the time-domain is realized by compressed sensing.The spectrum leakage of the reconstructed signal is suppressed by all-phase FFT spectrum analysis,and the amplitude demodulation method based on Fourier transform improves the amplitude demodulation accuracy.Then,an image reconstruction algorithm based on cross type five-beam time-domain sparse and rapid sampling is proposed,which uses the centrally symmetrical beam array to reconstruct 8 target images in a single exposure,and the speckle effect of the reconstructed image is quickly suppressed.Compared with the image reconstruction method based on "L" type three-beam sparse sampling in the time-domain,the number of samples taken in proposed method is reduced from 19 to 3 times,and the number of samples required for the echo signal of SBI system is greatly reduced.3.Aiming at the problem that the number of detector elements in SBI system is too large to be developed,four image reconstruction methods based on space-domain sparse sampling are proposed.By optimizing the arrangement of coherent lasers,the space-domain sparse sampling of echo signal is realized according to the space distribution of echo signal.Firstly,an image reconstruction method based on "P" type five-beam space-domain sparse sampling is proposed,and the number of detector elements is reduced by half under the premise of ensuring that the image resolution and image quality remain unchanged.Secondly,an image reconstruction method based on“T" type four-beam space-domain sparse sampling is proposed,the phase difference of other beat frequency components is used to replace the phase difference that is missing due to the reduction of the beam,so that a new phase correlation is established between the adjacent points of the target spectrum,and the quality of the reconstructed image is improved on the basis of the traditional "T" type four-beam image reconstruction method,and the Strehl ratios of the reconstructed image is increased by more than 20%.Then,a image reconstruction method based on sparse sampling in two-dimensional space-domain is proposed,which further reduces the number of detector elements.When the number of lasers emitted from the detector array is Mmultiplied by N(M>2,N>2),the number of detector elements required for the proposed method is only 1/[(M-1)(N-1)]as large as that of the traditional three-beam image reconstruction method,and the target image can be reconstructed,moreover,the image quality is almost the same.4.In order to further improve the sparse reconstruction capability of the imaging system,an image reconstruction method based on time-space joint sparse sampling is proposed.Experimental results show that the image quality of "T" type four-beam space-time sparse reconstruction algorithm is better than the traditional "T" type fourbeam space-time sparse reconstruction algorithm.In addition,the factors affecting the quality of space-time joint sparse imaging are analyzed,and the application conditions of the three types of sparse imaging techniques studied in this dissertation,that is,timedomain sparse imaging,space-domain sparse imaging and space-time joint sparse imaging.The SBI space-time joint sparse reconstruction technology proposed in this dissertation solves some key problems encountered by traditional SBI technique in the engineering application of remote moving target imaging,and provides theoretical and technical support for the development of SBI sparse reconstruction system.At the same time,it lays a certain technical foundation for the promotion and application of this technology in marine exploration,astronomical observation,medical imaging and other fields. |
PDF Full Text Request