New Windows based Color Morphological Operators for Biomedical Image Processing

20th Argentinean Bioengineering Society Congress

物理学;生物科学

Pastore, Juan^1,2,4 ; Bouchet, Agustina^1,2,4 ; Brun, Marcel^1,2 ; Ballarin, Virginia^1,3

Digital Images Processing Lab. ICyTE, UNMdP, Argentina^1

Department of Mathematics, School of Engineering, UNMdP, Argentina^2

Department of Electronics, School of Engineering, UNMdP, Argentina^3

CONICET, Argentina^4

关键词: Color morphological operators;    Color morphology;    Color perception;    Image processing and computer vision;    Lattice structures;    Morphological image processing;    Morphological operator;    Window-based;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/705/1/012023/pdf
DOI  :  10.1088/1742-6596/705/1/012023

学科分类：生物科学（综合）

来源: IOP

【 摘 要 】

Morphological image processing is well known as an efficient methodology for image processing and computer vision. With the wide use of color in many areas, the interest on the color perception and processing has been growing rapidly. Many models have been proposed to extend morphological operators to the field of color images, dealing with some new problems not present previously in the binary and gray level contexts. These solutions usually deal with the lattice structure of the color space, or provide it with total orders, to be able to define basic operators with required properties. In this work we propose a new locally defined ordering, in the context of window based morphological operators, for the definition of erosions-like and dilation-like operators, which provides the same desired properties expected from color morphology, avoiding some of the drawbacks of the prior approaches. Experimental results show that the proposed color operators can be efficiently used for color image processing.