Object-oriented segmentation for microscopy images of lung tissue

Abstract

Image segmentation is a very important process in image processing; and it plays an extremely important role in the field of image recognition, image analysis and image understanding. The quality of segmentation will directly affect the efficiency and results on identification, analysis and understanding. According to the information entropy (IE) in information theory and the theory of mathematical morphology (MM), we analyze and process the microscopy image of lung tissue. Also with the idea of object-oriented, we give a reasonable definition for the target area of image, and provide a better solution for separating the target and background in an effective way.