Research On Computational Ghost Imaging Reconstruction Model For Large Field Of View Detection

Large field of view detection is widely used in meteorological monitoring,aerospace,and military fields because of its wide coverage and abundant information.However,the large field of view detection under "point-to-point" traditional optical imaging faces many problems such as complex systems,atmospheric molecular scattering,and diffraction limit limitations.The requirements for imaging conditions are more stringent.The ghost imaging technology that uses second-order or even higher-order correlation of light fields to detect object information has great potential for applications in remote sensing imaging and radar detection due to its advantages of non-local imaging and anti-turbulence interference.However,ghost imaging in a large field of view has the characteristics of complex information and huge amount of data,which has a great impact on the imaging quality of ghost imaging.At the same time,its huge sampling data volume makes data transmission and storage difficult during imaging.It is not conducive to the practical application of ghost imaging in large field of view detection.In order to solve the above problems,this paper takes speckle patterns as the research object,and studies the ghost imaging reconstruction model that can reduce the amount of sampled data under the framework of computational ghost imaging,so that the ghost imaging technology can meet the requirements of large field of view detection.The research contents of this article are as follows:(1)Exploration on related theories of ghost imaging: Based on the basic concepts and related principles of ghost imaging,the working mechanism of computational ghost imaging framework is introduced,the main problems of ghost imaging for large field of view are analyzed.The characteristics of some ghost imaging methods suitable for large field of view are analyzed,and the main strategies for reducing the amount of sampled data of ghost imaging are given.It lays a theoretical foundation for the establishment of ghost imaging models suitable for large-field scenes and reduces the blindness of model establishment.(2)Research on the relationship between speckle pattern resolution and ghost imaging: By analyzing the characteristics of speckle patterns with different resolutions,the relationship between speckle pattern resolution and the amount of sampled data and imaging quality is analyzed;A ghost imaging reconstruction model based on local low resolution speckle pattern and a ghost imaging model based on sub-pixel shift of low resolution speckle pattern are proposed.Experiments show that the model has a significant effect on reducing the amount of ghost imaging sampling data and improving imaging quality.(3)Research on the relationship between filtering operation and ghost imaging: The relationship between the transmittance of the object to be measured and the imaging quality is explored,and the significance of edge reconstruction in ghost imaging for reducing the number of samples and effectively extracting scene information is discussed;The effects of different filtering methods and filtering at different stages on ghost imaging results are analyzed.A edge reconstruction model of ghost imaging based on speckle pattern high-pass filtering is proposed.Experiments verify that the model can reduce the amount of data by reducing sampling.