Title :
Cell Image Processing Based on Distance Transform and Regional Growth
Author :
Duan Jun ; Bai, Qinglong
Author_Institution :
Inf. Inst., Inner Mongolia Univ. of Sci. & Technol., Baotou, China
Abstract :
When picking up the characteristics of cancer cells from cell image, there are the phenomenon of cell adhesion and the phenomenon of similar pixel depth between nucleus and cytoplasm, which will greatly influence the accuracy of recognition. This article discusses feasibility of splitting adhesion among cells of fluid cancer cells with the method of distance transform, and distinguishing the nucleus and cytoplasm with regional growth method. Experimental results show that these methods mentioned in this article are effective for the division of adhesion cells, and they are suitable for the division between the cytoplasm and nuclei.
Keywords :
cancer; cellular biophysics; image recognition; medical image processing; transforms; cell adhesion; cell image processing; cell nucleus; cytoplasm; distance transform; fluid cancer cell; image recognition; nucleus; regional growth method; Adhesives; Cancer; Image segmentation; Shape; Transforms; Cell image processing; Image segmentation; distance transform; regional growth;
Conference_Titel :
Internet Computing for Science and Engineering (ICICSE), 2010 Fifth International Conference on
Conference_Location :
Heilongjiang
Print_ISBN :
978-1-4244-9954-0
DOI :
10.1109/ICICSE.2010.20
