Title :
Quick construction of efficient morphological operators by computational learning
Author_Institution :
Inst. de Matematica e Estatistica, Universidade de Sao Paulo, Brazil
fDate :
2/13/1997 12:00:00 AM
Abstract :
A new technique is introduced to construct generic morphological operators quickly and automatically using input-output pairs of images as training samples. This process can be modelled as generalised PAC (probably approximately correct) learning. A new representation for morphological operators, much more efficient than the traditional one, is also proposed
Keywords :
image processing; learning systems; mathematical morphology; computational learning; generalised PAC learning; generic morphological operators; input-output pairs of images; probably approximately correct learning; training samples;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19970199
