Hardware structure of 2D to 3D image conversion system for digital archives

Abstract

In this paper, we proposed a system that can automatically generate the depth map of the 2D image. First we segment the object by Sobel edge detection using different thresholds. Then thin down the image with Z.S thinning method. The image transformed into a grid image that edge of the object becomes a pixel. By using the gradient depth map, we fill the grid with its depth value. After automatically generated the depth map of the original 2D image, we combined depth maps with different thresholds to get the final depth map and perform the 2D to 3D conversion hardware structure and system design of digital for classical antiques and document.