Adaptive aperture pupil-inspired foveated single-pixel imaging for multi-posture moving targets

In nature, animals have evolved various shapes of pupils to detect moving targets with diverse postures effectively. Motivated by the concept of the adaptable aperture pupil, we present a pupil-inspired foveated single-pixel imaging approach featuring a variable foveal, enhancing the correlation between the fovea regions and targets in the scene. Furthermore, this pupil-inspired foveated strategy not only regulates the field of view by mimicking the expansion and contraction of the iris but also aids in target tracking by foveal saccade mimicry. In this way, our method improves the maximum utilization of high-resolution areas by 63%. The methodologies outlined in this study optimize the arrangement of pixel cells during the sampling process and promise to be a practical method enabling single-pixel imaging in dynamic scenes.