Efficient single-pixel imaging of axisymmetric targets withLaguerre–Gaussian speckles

This paper presents an efficient single-pixel imaging method for symmetric targets utilizing Laguerre–Gaussian (LG) speckle patterns. The proposed approach eliminates the requirement for imaginary-component measurements, which are essential in conventional complex speckle imaging techniques. By incorporating an angular degree of freedom into the LG speckle modes, the symmetry axis of the speckle can be precisely aligned with the target’s axisymmetric orientation. This alignment results in the elimination of the imaginary component in the bucket signal, thereby enabling complete image reconstruction using solely real-part measurements. Both numerical simulations and experimental validation demonstrate that for axisymmetric targets, high-quality images can be successfully recovered using only real-component data, achieving peak signal-to-noise ratio and structural similarity index measure comparable to or even slightly surpassing those obtained from full complex reconstruction. Furthermore, we develop a rotation angle estimation method that determines object orientation using only four measurements, validated with both binary and grayscale targets. Collectively, this work establishes an efficient imaging framework for symmetric objects and provides what we believe to be new perspectives for rotational alignment sensing applications.