OAM-Basis underwater single-pixel imaging based on improved GAN at a low sampling rate

Our study introduces a pioneering underwater single-pixel imaging (SPI) approach that employs Orbital Angular Momentum (OAM) basis as sampling scheme and a dual-attention residual U-Net Generative Adversarial Network (DARU-GAN) as reconstruction algorithm. This method is designed to addresses the challenges of low sampling rates and high turbidity typically encountered in underwater environments. The integration of OAM basis sampling scheme and the improved reconstruction network not only enhances reconstruction quality but also ensures robust generalization capabilities, effectively restoring underwater target images even under the stringent conditions of a 3.125% sampling rate and 128 NTU turbidity. The integration of OAM beams' inherent turbulence resistance with DARU-GAN's advanced image reconstruction capabilities make it as an ideal solution for high-turbid underwater imaging applications.