The following information was submitted:
Transactions: INTERNATIONAL JOURNAL of CIRCUITS, SYSTEMS and SIGNAL PROCESSING
Transactions ID Number: 19-895
Full Name: Paula Mariela Tristan
Position: Ph.D. Candidate
Age: ON
Sex: Female
Address: Pinto 399
Country: ARGENTINA
Tel: +54 2293 439682
Tel prefix:
Fax:
E-mail address: ptristan@exa.unicen.edu.ar
Other E-mails: ptristan3@gmail.com
Title of the Paper: Automatic Improvement of Image Registration Using High Information Content Pixels
Authors as they appear in the Paper: Paula Tristan, Ruben Wainschenker and Jorge H. Doorn
Email addresses of all the authors: ptristan@exa.unicen.edu.ar,rfw@exa.unicen.edu.ar,jdoorn@exa.unicen.edu.ar
Number of paper pages: 8
Abstract: Image registration is currently used to refer the activity of make to coincide two or more observation of the same scene. This task is fundamental for several areas of image processing, such as: remote sensing, processing of medical images and video processing, among others. There are numerous publications addressing the image registration problem, from several points of view. It involves choosing one image as reference. The remaining images must be shift to match the one chosen as reference. Two images match between them whenever every region of the original scene is mapped on the same set of pixels in both images. The registration problem consists in finding the geometric transformation to accomplish the required modification of the image. Existing methods may be classified as: based on characteristics and based on areas. The first group of methods requires the intervention of an expert who chooses a set of distinctive points of the image, while the second group !
of methods automatically uses all pixels of the image. This does not take into account the fact that the information content of different pixels may be rather different. Pixels with low content of information degrade the contribution of the pixels with high information content. Information content is measured in relation with its contribution to define the geometric transformation. This paper details a method that automatically discards pixels with low content of information focusing the calculation on those with high information content.
Keywords: Image Registration, Image Matching, Super-Resolution.
EXTENSION of the file: .pdf
Special (Invited) Session: Automatic Pixel Selection Criteria for Image Registration
Organizer of the Session: 104-161
How Did you learn about congress:
IP ADDRESS: 190.178.226.47
