Title :
Classification of segmentable images using the PDD function
Author :
Uchill, Joseph ; Assadi, Amir
Author_Institution :
Dept. of Math., Wisconsin Univ., Madison, WI, USA
Abstract :
Due to the need to search large databases of images, image classifiers have been developed to sort images not only by subject matter, but also into several more general classes such as photographs, text, and images. These sorting methods suggest that there are inherent and identifiable characteristics unique to photographs or text. A more fundamental problem to solve is whether there are features which combine these separate classes into the greater class of segmentable images:images which have been generated as a means of communicating visual data. A new type of classifier is proposed; one which separates segmentable images from the greater set of possible pixel combinations.
Keywords :
image classification; image segmentation; visual communication; visual databases; image database; pixel density decomposition; segmentable images classification; sorting methods; visual data communication; Cameras; Computer graphics; Computer vision; Cryptography; Dictionaries; Humans; Image databases; Image segmentation; Pixel; Sorting;
Conference_Titel :
Image Processing, 2003. ICIP 2003. Proceedings. 2003 International Conference on
Print_ISBN :
0-7803-7750-8
DOI :
10.1109/ICIP.2003.1247297