220 - Super-Resolution Imaging : Use of Zoom as a Cue M.V....

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Super-Resolution Imaging : Use of Zoom as a Cue M.V. Joshi and Subhasis Chaudhuri Department of Electrical Engineering Indian Institute of Technology - Bombay Powai, Mumbai-400076. India mvjoshi, sc @ee.iitb.ac.in Abstract In this paper we propose a novel technique for super- resolution imaging of a scene from observations at different zooms. Given a sequence of images with different zoom fac- tors of a static scene, the problem is to obtain a picture of the entire scene at a resolution corresponding to the most zoomed image in the scene. We model the super-resolution image as a Markov random field (MRF) and a maximum a posteriori estimation method is used to derive a cost func- tion which is then optimized to recover the high resolution field. Since there is no relative motion between the scene and the camera, as is the case with most of the super- resolution techniques, we do away with the correspondence problem. 1. Introduction Availability of high spatial resolution images is often de- sirable in most computer vision applications. Be it remote sensing, medical imaging, robot vision, industrial inspec- tion or video enhancement (to name a few), operating on high resolution images leads to a better analysis in the form of lesser misclassification, better fault detection, more true- positives, etc. However, acquisition of high resolution im- ages is severely constrained by the drawbacks of sensors that are commercially readily available. Thus, images ac- quired through such sensors suffer from aliasing and blur- ring. Aliasing occurs as a consequence of insufficient den- sity of the detector array which causes sampling of the scene at less than Nyquist rate, while blurring occurs due to in- tegration of the sensor point spread function (PSF) at the sensor surface. Hence, one must resort to image process- ing methods to construct a high resolution image from one or more available low resolution images. Super-resolution refers to the process of producing a high spatial resolution image from several low resolution images. It includes up- sampling the image thereby increasing the maximum spa- tial frequency that can be represented and removing degra- dations that arise during image capture, viz., aliasing and blurring. The effect of aliasing differs with zooming. Thus one can use zoom as cue for generating high resolution im- ages at the lesser zoomed area of a scene. Now we review some of the prior works on super- resolution imaging. Many researchers have tackled the super-resolution problem for both still and video images, e.g., [8, 9, 14, 18] (see [16] for details). Tsai and Huang [18] were the first to propose a frequency domain approach to re- construction of a high resolution image from a sequence of undersampled low resolution, noise-free images. Kim et al. discuss a recursive algorithm, also in the frequency domain,
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/22/2010 for the course MI IP taught by Professor Vladbalan during the Spring '10 term at Universidad del Rosario.

Page1 / 6

220 - Super-Resolution Imaging : Use of Zoom as a Cue M.V....

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online