Please use this identifier to cite or link to this item:
|Title:||The Little Algorithm that Grew: Scaling the Morphological Image Compositing Algorithm to Meet the Chanllenges of Processing Large Image Data Sets|
|Authors:||BIELSKI CONRAD; GRAZZINI JACOPO; SOILLE PIERRE|
|Citation:||Proceedings of the 9th International Conference on Geocomputation - GeoComputation 2007: The Art and Science of Solving Complex Spatial Problems with Computers p. Paper 1A5|
|URI:||http://ncg.nuim.ie/geocomputation/papers.html; hhttp://ncg.nuim.ie/geocomputation/sessions/1A/1A5.pdf; ttp://ncg.nuim.ie/geocomputation/materials/Geocomputation2007-Proceedings.pdf|
|Type:||Articles in periodicals and books|
|Abstract:||Image mosaicing (also called image compositing) is a technique used to join two or more images together. Recently, Soille (2006) developed a morphological image compositing algorithm which is being used to automatically generate a panEuropean image mosaic based on fine spatial resolution satellite imagery. While this is not the only image mosaicing algorithm available today (Szeliski 2004, Soille 2006, Price 2006) it does generate a mosaic with the following required characteristics: visually pleasing and research grade. Visually pleasing in this case means that the seam lines are difficult to detect visually and research grade means that the radiometry of the original image data is not changed in the resulting mosaic. This talk will describe the evolution of the morphological image compositing algorithm in order to deal with computer memory limitations to make it work on large image data sets and the adaptations to make it available within a grid computing environment. The challenges and solutions of this approach will be described as well as the benefits of this algorithm for use in processing large image data sets.|
|JRC Institute:||Sustainable Resources|
Files in This Item:
There are no files associated with this item.
Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.