Resolution enhancement of ghost imaging in atom vapor

Ghost translation: an end-to-end ghost imaging approach based on the transformer network

Artificial intelligence has recently been widely used in computational imaging. The deep neural network (DNN) improves the signal-to-noise ratio of the retrieved images, whose quality is otherwise corrupted due to the low sampling ratio or noisy environments. This work proposes a new computational imaging scheme based on the sequence transduction mechanism with the transformer network. The simulation database assists the network in achieving signal translation ability. The experimental single-pixel detector's signal will be 'translated' into a 2D image in an end-to-end manner. High-quality images with no background noise can be retrieved at a sampling ratio as low as 2%. The illumination patterns can be either well-designed speckle patterns for sub-Nyquist imaging or random speckle patterns. Moreover, our method is robust to noise interference. This translation mechanism opens a new direction for DNN-assisted ghost imaging and can be used in various computational imaging scenarios.

Controllable superbunching effect from four-wave mixing process in atomic vapor

Correlation property of light limits the performance in related applications such as the visibility of ghost imaging or intensity interferometry. To exceed these performance limits, we here manipulate the degree of second- and higher-order coherence of light by changing controllable variables in four-wave mixing (FWM) process. The measured degree of second- and third-order coherence of the output light beams considerably exceed those of the incident pseudothermal light. Namely superbunching effects, g⁽²⁾(0) value up to 7.47 and g⁽³⁾(0) value up to 58.34, are observed experimentally. In addition, strong second- and third-order cross-correlation exist between the output light beams. Further insights into the dependence of superbunching light on the temperature of Rb vapor, the laser detuning and the optical power of all the incident light beams show that it can serve as a light source with a tunable superbunching degree.

Spatially modulated thermal light in atomic medium for enhanced ghost imaging

Recent years have seen vast progress in image modulation based on atomic media, with potential applications in both classical optical imaging and quantum imaging regions. However, there have been few investigations of how thermal light images interact with an electromagnetically induced transparent medium. In this letter, we experimentally demonstrate pseudo-thermal light modulation on coherent population trapping conditions in ⁸⁷Rb vapor. By introducing the Laguerre-Gaussian beam as the control beam and the encoded speckle as the probe beam, we obtained sharper speckle patterns after the atom cell compared with that in free space. The spatially modulated thermal light was then used to enhance the image resolution in ghost imaging of which the resolution was enhanced by factor 3, since the ghost image resolution is heavily reliant on the speckle’s transverse coherent length. Our results are promising for potential applications in high resolution ghost imaging and image metrology, image processing and biomedical imaging.