A fast image reconstruction algorithm using adaptive R-tree segmentation and B-splines

Abstract

The image reconstruction using adaptive R tree based segmentation and linear B- splines is addressed in this paper. We used our own significant pixel selection method to use a combination of canny and sobel edge detection techniques and then store the edges in an adaptive R tree to enhance and improve image reconstruction. The image set can be encapsulated in a bounding box which contains the connected parts of the edges found using edge-detection techniques. Image reconstruction is done based on the approximation of image regarded as a function, by B-spline over adapted Delaunay triangulation. The proposed method is compared with some of the existing image reconstruction spline models. © 2012 Springer-Verlag.