Effective reconstruction of a partially occluded 3-D target by using a pixel restoration scheme in computational integral-imaging

In this paper, we propose an effective approach for visibility-enhanced reconstruction of a partially occluded 3-D target by using the pixel-restoration method in the computational integral-imaging system. In the proposed method, elemental images (EIs) of the 3-D target that are partially occluded by a foreground object are picked up and transformed into sub-images (SIs). By using the block-matching algorithm then, the occluded target regions of each sub-image are estimated and removed. After that, the missing pixels in each sub-image are re-established by using the proposed pixel-restoration method. Finally, these restored SIs are converted back into the newly modified EIs, from which the resolution-enhanced target can be reconstructed with the computational integral-imaging reconstruction (CIIR) method. Successful experiments with the test objects confirm the feasibility of the proposed method.

► The method is proposed for visibility-enhanced reconstruction in the integral-imaging system.
► A block-matching algorithm is used to estimate the occluded regions of each sub-image.
► The occluded regions can be removed based on the estimated depth maps.
► The proposed method is used to minimize the effect of the occluded region on the 3-D object.