Incoherent Active Convolved Illumination Enhances the Signal-to-Noise Ratio for Shot Noise: Experimental Evidence

Abstract

Imaging is indispensable for nearly every field of science, engineering, technology, and medicine. However, measurement noise and stochastic distortions pose fundamental limits to accessible spatiotemporal information despite impressive tools such as structured-illumination microscopy (SIM), stochastic optical reconstruction microscopy or photoactivated localization microscopy (STORM or PALM), and stimulated-emission-depletion (STED) microscopy. How to combat this challenge ideally has been an open question for decades. Inspired by a virtual-gain technique to compensate losses in metamaterials, active convolved illumination (ACI) has recently been proposed to significantly improve the signal-to-noise ratio and hence the data acquisition. In this technique, the light pattern of the object is superimposed with a correlated auxiliary pattern, the function of which is to reverse the adverse effects of losses, noise, and random distortion based on their spectral characteristics. Despite enormous implications in statistics, any experimental evidence verifying the theory of this novel technique has been lacking to date. We find experimentally that ACI boosts not just the resolution limit and image contrast but also the resistance to pixel saturation. The results confirm the previous theories and may open up horizons in a wide range of disciplines from atmospheric sciences, seismology, biology, statistical learning, finance, and information processing to quantum noise beyond the fundamental boundaries.