Open AccessCo-occurrence Matrix and fractal dimension for image segmentation
Author(s) -
Beatriz S. Marón
Publication year - 2012
Publication title -
revista de matemáticas
Language(s) - English
Resource type - Journals
eISSN - 2215-3373
pISSN - 1409-2433
DOI - 10.15517/rmta.v19i1.2104
Subject(s) - artificial intelligence , fractal dimension , dimension (graph theory) , computer vision , fractal , segmentation , pattern recognition (psychology) , object (grammar) , computer science , image texture , matrix (chemical analysis) , image segmentation , fractal analysis , scale space segmentation , operator (biology) , image (mathematics) , segmentation based object categorization , texture (cosmology) , image processing , mathematics , pure mathematics , mathematical analysis , biochemistry , chemistry , materials science , composite material , repressor , transcription factor , gene
One of the most important tasks in image processing problem and machine vision is object recognition, and the success of many proposed methods relies on a suitable choice of algorithm for the segmentation of an image. This paper focuses on how to apply texture operators based on the concept of fractal dimension and cooccurence matrix, to the problem of object recognition and a new method based on fractal dimension is introduced. Several images, in which the result of the segmentation can be shown, are used to illustrate the use of each method and a comparative study of each operator is made.