Search

High-flux measurement characteristics of compressed sensing (CS) imaging causes the imaging system prone to be disturbed by quantization. To realize high-quality CS imaging with limited detector bits, an improved imaging method combining sparse measurements and multiple dithers is proposed to reduce the dynamic range of the measured signals and increase that of effective detection. Simulations and experiments show that compared with traditional CS imaging, the proposed system decreases reconstruction errors caused by quantization distortions and may reduce the required number of detector bits to 1. The effects of detector noise and system parameters are discussed to validate the feasibility and performance of this method.

Compressive sensing (CS) has been used in LiDAR systems utilizing one single-photon-counting avalanche diode. We demonstrate an unexpected grayscale inversed image of an object at an unchosen depth, which appears in the reconstruction of the infrared single-pixel LiDAR system due to the pile-up effect. A correction algorithm and the sparse measurement are proposed and experimentally verified to effectively reduce the photon pile-up influence, so that the negative images can be completely removed. The correction methods in this research can improve the accuracy and the flexibility of the single-pixel LiDAR systems employing detectors with a low maximum light count.