Computational ghost imaging through a dynamic scattering medium based on a convolutional neural network from simulation

Article Optics

Color computational ghost imaging by deep learning based on simulation data training

Zhan Yu et al.

Summary: This paper presents a new color computational ghost imaging strategy using a single-pixel detector and training by simulated dataset. The experimental results show that this method can accurately recover image information and correct color distortion with good generalization.

APPLIED OPTICS (2022)

添加到收藏夹

Article Optics

Transition from the ballistic to the snake regime of a femtosecond laser through a turbid medium via Monte Carlo simulation

Yuhu Ren et al.

Summary: This study investigates the transition from ballistic to snake diffusion of a femtosecond laser through a turbid medium using Monte Carlo simulation. The results show that the transition depth is close to the surface of the scattering volume and is dependent on the scattering coefficient of the medium. Unlike the intensity decay governed by Beer-Lambert law for ballistic photons, the intensity of snake photons initially increases with increasing thickness of the turbid medium and then decreases for all scattering coefficients ranging from 10 to 20 cm(-1). The balance point between the intensity of ballistic and snake photons depends on the scattering coefficient and the half-acceptance angle, and the product of the transmission depth and scattering coefficient is approximately a constant (around 2) for a specific simulation configuration.

LASER PHYSICS LETTERS (2022)

添加到收藏夹

Article Optics

Expanding the memory effect in scattering imaging by manipulating photon distributions

Zhengjia Zhang et al.

Summary: This paper proposes a speckle correlation technique based on the memory effect to recover objects from multiple scattered light without the information of the scattering media. The autoshrinking algorithm is used to expand the field of view in scattering imaging, and a pretreatment iteration scheme is developed to achieve high accuracy reconstruction. Experiments and simulations successfully confirm the effectiveness of the method.

PHYSICAL REVIEW A (2022)

添加到收藏夹

Article Optics

Color computational ghost imaging based on a generative adversarial network

Yang Ni et al.

Summary: The novel color computational ghost imaging scheme simplifies experimental setup and shortens acquisition time, effectively removing noise and color distortion with the use of a generative adversarial network, resulting in better visual quality and realistic textures in the final images.

OPTICS LETTERS (2021)

添加到收藏夹

Article Optics