基于数学形态学的细胞图像分割方法研究及实现  

Research and Realization of Cell Image Segmentation Method Based on Mathematical Morphology

作　　者：张燕红[1] 李瑛[1] 张燕宁 王凤芹[1] 

机构地区：[1]海军航空工程学院基础部计算机教研室,山东烟台264001 [2]北京天圆泰工程造价咨询有限公司烟台分公司,山东烟台264001

出　　处：《计算机与现代化》2013年第7期135-137,163,共4页Computer and Modernization

摘　　要：图像分割是数字图像处理的一项重要技术。本文结合边缘检测和数学形态学提出一种新的图像分割方法,并应用到细胞图像中,借助Matlab平台,验证该方法的有效性。在检测细胞边缘的过程中运用数学形态学的相关知识,对图像进行膨胀、填充等处理,有效地弥补了单一分割算法存在的分割边缘不够精细、细胞内部孔洞较多等缺点。在显示阶段对图像进行腐蚀操作,使得细胞边缘更加光滑,分割结果更加自然,大大提高了分割图像的可读性。Image segmentation is an important technology for digital image processing.This paper presents a new image segmentation method based on edge detection and mathematical morphology,and verifies its effectiveness in cell image with Matlab platform.Using the knowledge of mathematical morphology during the detection of cell edge,image processing of expansion and filling effectively makes up defects of single segmentation algorithm such as segmentation edge insufficiently sophisticated and cell internal more holes.The image corrosion in the display stage makes the cells more smooth edges and the segmentation results look more natural.It greatly improves the readability of the segmented image.

关 键 词：图像分割 边缘检测 数学形态学 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]