Detector-multiplexed infrared sparse point-target imaging via field-dependent PSF encoding

Wide-field infrared sparse point-target surveillance is constrained by the size, cost, and readout complexity of infrared focal-plane arrays. This paper proposes a detector-multiplexed infrared imaging architecture for sparse point-like targets, in which multiple sub-fields are folded onto a common small-format detector and distinguished by field-dependent point spread function (PSF) encoding. A cylindrical modulation element is introduced at the intermediate image plane to generate controllable off-axis aberrations, producing spatially variant PSFs whose orientation and aspect ratio vary deterministically with field angle. The ZEMAX optical model combines a secondary imaging system, a non-sequential folding module, and a cylindrical modulation stage to establish the sub-field-to-detector mapping. A PSF-codebook decoding framework is further introduced to quantify PSF distinguishability and recover the originating sub-field of sparse point targets from centroid and morphology features. Simulation and experimental results show controllable PSF aspect-ratio variation from approximately 1.9 to 3.4 while preserving sufficient energy concentration for small-target detection. The application-oriented decoding evaluation, including classification accuracy, localization error, confusion matrix, and signal-to-noise ratio sensitivity, is reported in the added validation section. The proposed architecture therefore provides a practical optical front end for detector-multiplexed infrared sparse point-target imaging, while its use for dense extended scenes remains outside the present scope.