Color Fourier Single-pixel Imaging

Single-pixel imaging is a new type of imaging technology,which obtains the information of the object through single-pixel detector and reconstructs the object image by calculating the imaging method.In the past decade,single-pixel imaging has received more and more attention,because single-pixel detectors are easier to manufacture than two-dimensional array sensors,and it is expected to solve the difficulty of two-dimensional array sensors in non-visible wavelengths imaging.Single-pixel imaging technology has important scientific significance and application value in the field of optical imaging.Currently,there are few single-pixel imaging technologies that can obtain the color information of objects,and the existing color single-pixel imaging requires a lot of measurement times,leading to a long time in the experimental process,which greatly reduces the imaging efficiency and hinders the application of related research work.In order to solve the problem that the current two-dimensional color single-pixel imaging scheme has many measurement times and long acquisition time,a two-dimensional color Fourier single-pixel compression imaging scheme is proposed in this paper.Based on the sparsity of natural objects in the Fourier transform domain,color Fourier basis patterns based on Bayer filter are designed.The spatial information and color information of objects can be compressed and encoded in the low frequency and high frequency parts of the Fourier transform domain by using this pattern for lighting.The effectiveness of the proposed method is prove d by simulation and experiment.When the number of measurements is far less than the total number of image pixels,the proposed method can reconstruct a clear,well-colored two-dimensional image.Based on two-dimensional color Fourier single-pixel imaging,the research of three-dimensional color Fourier single-pixel imaging is further carried out.By modulating the detection optical path of single-pixel imaging,the depth information of the target object is further coded,and the color,depth and spatial information of the object are compressed and coded in the high,medium and low frequency parts of Fourier transform domain respectively.The effectiveness of the proposed method is verified by experiments.When the number of measurements is far less than t he total number of image pixels,the proposed method can reconstruct a three-dimensional image with clear details,high precision and perfect color.Based on the study of color Fourier single-pixel imaging,this paper further develops the study of multi-spectral Fourier single-pixel imaging,designs the spatial light modulation filters which can be used to obtain four-wavelength and nine-wavelength multi-spectral images,and proposes the correspondin g image reconstruction algorithm,and finally the effectiveness of the proposed method is verified by numerical simulation.