Title :
Image segmentation based on evaluation of candidate regions
Author :
Tracz, Pawel ; Szczepaniak, Piotr S.
Author_Institution :
Inst. of Inf. Technol., Lodz Univ. of Technol., Lodz, Poland
Abstract :
In this paper, a method of image segmentation is proposed. It is a region growing segmentation based on the assessment of candidate regions. It uses the concept of co-occurrence matrix. The method is illustrated with the example of liver locally affected by cancer cells manifested on the computer tomography image.
Keywords :
cancer; computerised tomography; image segmentation; liver; matrix algebra; medical image processing; cancer cells; candidate region evaluation; computer tomography image; cooccurrence matrix; image segmentation; liver; region growing segmentation; Biomedical imaging; Brightness; Cancer; Image segmentation; Lattices; Liver; Object segmentation; co-occurrence matrix; medical image segmentation; region growing;
Conference_Titel :
Cybernetics (CYBCONF), 2015 IEEE 2nd International Conference on
Conference_Location :
Gdynia
Print_ISBN :
978-1-4799-8320-9
DOI :
10.1109/CYBConf.2015.7175936
