Demonstration of computational ghost imaging through fog

Abstract

Imaging through fog is an active and challenging topic in the fields of optical imaging and image processing. Ghost imaging (GI), as an indirect optical imaging technique, can acquire clear images of unknown objects hidden in some hostile environments, such as underwater, turbulence, and scattering media. Here the computational ghost imaging (CGI) technique is employed to retrieve images of objects hidden in a foggy environment. Based on the extended Huygens-Fresnel principle, the imaging formula for the CGI system through fog is developed with a Lorentz shaped incoherent source. The fog model is established by combining atmospheric turbulence and atmospheric scattering. Two objects with different slit widths are applied in the numerical examples and the effects of the propagation distance, turbulence strength, and scattering particle concentration are analyzed in detail. The results show that the quality of retrieved images is hardly degraded in the CGI system through fog under short-distance conditions. While the effect of fog on imaging quality becomes more pronounced as the distance increases. To improve the quality of the CGI system through fog, the Lorentz shaped incoherent source is effectively employed to obtain better ghost images than the widely used Gaussian source, especially in long-distance conditions. Thus the work is very useful for promoting CGI real applications, such as marine navigation, road transportation safety, and remote sensing imaging.