Superresolution Testing and Results 1.2 2006/12/20 15:12:26 US/Central 2006/12/22 01:04:30.616 US/Central Jennifer Ann Gillenwater jgillenw@rice.edu J. Ryan Stinnett jryans@rice.edu Elica Skorcheva elicas@rice.edu Jennifer Ann Gillenwater jgillenw@rice.edu J. Ryan Stinnett jryans@rice.edu Elica Skorcheva elicas@rice.edu Richard G. Baraniuk richb@rice.edu Mark A. Davenport md@rice.edu results superresolution Using projective flow and Delauany triangulation to accompish image registration and interpolation, we achieved the following superresolution results. The following images illustrate the effects of the projective flow and Delaunay triangulation algorithms we used to implement image superresolution. Our testing process was as follows:Take several pictures of the same object with a digital camera (no tripod -> images will need registration). Downsample the images to create low-resolution (LR) copies. Register all LR images with respect to the first LR image. Interpolate to obtain the HR image.We also tested using LR images from external sources. An example is given below.Original images:

LR image 1 (Source: 7)

LR image 2 (Source: 7)

LR image 3 (Source: 7)

LR image 4 (Source: 7)

HR image result:

HR image

The MATLAB code used to generate the above HR image is available here.