Count And Adhesion Cell Image Segmentation Study

Along with the development of the computer technology, computer image processing isplaying a more and more important role in clinical diagnosis and treatment. Digital imageprocessing and pattern recognition technology are now widely used in microorganismfields.Computer-aided medical cell image processing is more accurate than visualinspection, and reducing subjective interference.Computer-aided medical cell image processing includes image collection, imagepretreatment, image segmentation, feature extraction , feature analysis and output, etc. It isessential to segment the images right for accurate image data results. And the quality of thesegment will influence the image analysis, recognition and explanation directly, and it isthe difficult point and key step in automatic deciphering and analysis.However, although a lot of scientists who have researched image segmentation profoundlyadvance a number of valid segment algorithms, none of them are best suitable to all thetested images. In this paper a segmentation based on the range convert is adopted whichhelps to count and segment numerous conglutination cells. This method has been proved tohave a better segment result by avoiding the traditional Watershed Algorithm which maysegment excessively.