Title :
Studies on discrimination and segmentation method of touching milk somatic cell
Author :
Na, Su ; Heru, Xue
Author_Institution :
Coll. of Comput. & Inf. Eng., Inner Mongolia Agric. Univ., Hohhot, China
Abstract :
The stack phenomenon existed in milk somatic cell pictures and the shape irregularity. In this paper, firstly a suitable shape factor selected judges whether it occurred in touching. This is in order to reduce false segmentation. An improvement watershed method is proposed which is mainly used in resolving the over-segmentation problem. The idea is obtaining the central points in the local maximum regions (seed points), and merging the redundancy seed points. Then the small areas are deleted after watershed segmentation. Finally segmentation points and segmentation lines are selected appropriately, too. Experiment shows the method can accurately and quickly separate touching milk somatic cell. The effect is obvious.
Keywords :
cellular biophysics; dairy products; feature extraction; image segmentation; medical image processing; image processing; local maximum regions; redundancy seed points; shape factor; shape irregularity; stack phenomenon; touching milk somatic cell; watershed segmentation; Agricultural engineering; Dairy products; Educational institutions; Euclidean distance; Human factors; Image processing; Image segmentation; Merging; Pixel; Shape; milk somatic cell; touching cell discrimination; touching cell segmentation;
Conference_Titel :
Bioinformatics and Biomedical Technology (ICBBT), 2010 International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6775-4
DOI :
10.1109/ICBBT.2010.5478932