Group frame neural network of moving object ghost imaging combined with frame merging algorithm
The nature of multiple samples to extract correlation information limits the applications of ghost imaging of moving objects. A novel multi-to-one neural network is proposed and the concept of "batch frame" is introduced to improve the serial imaging method. The neural network extracts more correlation information from a small number of samples, thus reducing the sampling ratio of the ghost imaging technique. We combine the correlation characteristics between images to propose a frame merging algorithm, which eliminates the dynamic blur of high-speed moving objects and further improves the reconstruction quality of moving object images at a low sampling ratio. The experimental results are consistent with the simulation results.