Inaccuracy Amplified: Compressed Sensing Under Experimental Misalignment

Compressed sensing (CS) with single-pixel detectors and optical phased arrays (OPAs) allows efficient image reconstruction by exploiting structured speckle patterns. This study investigates how spatial misalignments affect reconstruction fidelity. We numerically simulate the alignment errors of different magnitude and evaluate the reconstructed image using both the structural similarity index (SSIM) and the peak signal-tonoise ratio (PSNR). These findings highlight the sensitivity of compressed sensing to spatial misalignments, emphasizing the need for precise control in experimental setups to ensure reliable image reconstruction.