Lensless imaging via LED array based computational ghost imaging

Lensless imaging uses a mask plate instead of a conventional lens and reconstructs an image based on the point spread function (PSF), which allows for a compact optical path. This method relies heavily on accurate estimation of the PSF of the mask plate, which entails meticulous assembly of the mask plate and calibration of the PSF. Computational ghost imaging (CGI) reduces the need for spatial resolution of the detector by modulating the object multiple times. Combining the characteristics of the two methods, we demonstrate a lensless imaging model based on an LED array, named Quasi-Uniform PSF Model. In this model, the PSF is quasi-uniform in the field of view and can be accurately calculated. Inspired by traditional CGI algorithms, we proposed a deconvolution image reconstruction algorithm suitable for our lensless imaging model. The exact PSF of this model is discussed in numerical simulations. We built a prototype, and reconstructed recognizable images with a resolution of 32 x 32 pixels. The results of the experiment were in agreement with the simulation results. Our lensless imaging model eschews lenses, masks, or array detectors, and can evaluate the accurate PSF.