Underwater Fourier single pixel imaging based on water degradation function compensation method

The detection environment of underwater optical imaging is relatively complex. The forward scattering, back scattering and absorption greatly reduce the imaging quality of underwater optical imaging. Single pixel imaging is considered to be a suitable technique for underwater optical imaging because of its high noise resistance. The serious problem of underwater single pixel imaging system is that the structural light is required for illumination, and the forward scattering distorts the pre-generated speckle, when the speckle travels underwater. Therefore, the resolution of the reconstruction results decreases, making the reconstruction results blurry. In order to reduce the influence of the forward scattering on single pixel system, the reconstruction process of Fourier single pixel imaging should be improved. Underwater degradation function of Fourier single pixel imaging system was estimated in the spectrum domain, and then target spatial spectrum inversion was implemented based on the estimated degradation function. The image of the target can be obtained by transforming the target spatial spectrum with Fourier transform. The validity of the proposed method was proved by theoretical analysis and experimental results. Utilizing the proposed method, the influence of the forward scattering was reduced and the quality of reconstruction results of underwater Fourier single pixel imaging was improved.