Long-Range Fast Single-Pixel Localization of Multiple Moving Targets

Single-pixel imaging is an innovative technique for object imaging that employs non-scanning single-pixel detectors. Most methods for multi-target localization by single-pixel imaging involve extensive light field modulation, leading to long sampling times, and they lack long-distance experimental verification. This paper proposes a multi-target localization method with high pixel resolution and ultra-low sampling rates, wherein the selection of positioning speckle patterns is optimized using Fourier spectrum characteristics to reduce the sampling rate. Long-range field experiments validated our scheme which, in contrast to conventional methods, significantly reduced the number of required samples by up to tenfold and slashed the sampling rate by as much as a hundredfold. In field experiments at a distance of 258.5 meters, we accurately located multiple moving targets with a localization rate of 133 frames per second at a pixel resolution of 768 × 1024 pixels and modulation frequency of 8kHz, while maintaining an extremely low sampling rate of only 0.00763%.