Title :
Unsupervised image segmentation using the minimum description length principle
Author :
Sheinvald, Jacob ; Dom, Byron ; Niblack, Wayne ; Steele, David
Author_Institution :
IBM Res. Div., Almaden Res. Center, San Jose, CA, USA
fDate :
30 Aug-3 Sep 1992
Abstract :
Considers the problem of unsupervised multiband image segmentation specifically, using the MDL criterion, an associated complexity measure. According to the MDL principle, the best estimates are those that result in the most compact encoding of the image. An important advantage of such an approach is that it is virtually free of the need to choose arbitrary thresholds, which are typical of many segmentation techniques, and which, in many cases, need to be interactively adjusted in order to get satisfactory results
Keywords :
image coding; image segmentation; MDL criterion; encoding; minimum description length principle; segmentation techniques; unsupervised multiband image segmentation; Gaussian distribution; Image coding; Image segmentation; Integrated circuit modeling; Jacobian matrices; Labeling; Layout; Pixel;
Conference_Titel :
Pattern Recognition, 1992. Vol.II. Conference B: Pattern Recognition Methodology and Systems, Proceedings., 11th IAPR International Conference on
Conference_Location :
The Hague
Print_ISBN :
0-8186-2915-0
DOI :
10.1109/ICPR.1992.201875