Defocus and geometric distortion correction for projected images on a curved surface

In this paper, a correction method of defocus and geometric distortion for projected images on nonideal projection surfaces is proposed. To offer distortion-free images on curved projection surfaces, a camera captures the projected image to estimate the distortion. To compensate for defocus, which is due to the change in distance from the projector to the projection surface, the space varying point-spread function of the projector is modeled using a thin lens model. The performance of the proposed algorithm is verified from experiments when the projection surface is curved; the defocus is compensated for using the space varying point-spread function while correcting for the geometric distortion. Employing a point-spread function model, which depends on the distance from the projector to the projection surface, is quite simple and accurate; therefore, a focused image could be obtained using space varying deconvolution. The aim of our proposed method is for application to mobile projectors.