Improved resolution integral imaging using random aperture coding based on compressive sensing

The low-resolution of the three-dimensional (3D) image is the main limitation for the development of integral imaging. We proposed a resolution enhanced integral imaging method by using a randomly coded aperture based on compressive sensing (CS). The compressive integral imaging model is derived based on wave optics theory. Instead of sensing raw pixel data, the recording device measures the compressive samples of the observed 3D scene through a random phase mask attached to each lenslet. The coded elemental images are captured with a single exposure and the super-resolution elemental images can be reconstructed by an optimization algorithm. In this paper, we use CS in the acquisition of the elemental images, and ultimately on improving the image quality of reproduced 3D scene either by optical experiment or computational method. Both the theoretical model and experimental results show the feasibility of the proposed method.