Image-free single-pixel fast localization of multi-target using small window projection

Multi-target localization using projections in multiple directions is an effective strategy for distinguishing the positions of multiple targets, which will lead to time consumption in single-pixel localization. In order to achieve fast tracking of multiple targets, we propose a Fourier single-pixel detection method based on small-window projections. Our method avoids projecting in multiple directions globally, and in most cases, tracking multiple targets can be accomplished by projecting in only two directions within a small window corresponding to a single target. So, due to the small projection window, only a few projections are required to complete the localization. Simulation results show that this method can accomplish the multi-target tracking task. At a spatial resolution of $256 \times 256$ pixels, four targets are tracked simultaneously, and a minimum of only 32 projections are required to complete one tracking frame with a tracking error of no more than one pixel. This technique provides an effective scheme for real-time fast-tracking of multiple targets.