基于形态学重构的多结构元细胞图像边缘检测  

A Cell Edge Detection Algorithm Based on Morphological Reconstruction and Multi-structure

作　　者：张鑫[1] 陈伟斌[2] 

机构地区：[1]温州医学院信息与工程学院,浙江温州325000 [2]温州大学计算机学院,浙江温州325000

出　　处：《计算机仿真》2009年第8期216-219,294,共5页Computer Simulation

摘　　要：细胞图像边缘检测结果为细胞形态学分析提供依据。针对传统边缘检测算法在细胞图像边缘检测中存在的问题。为了改善图像细节丢失的缺点,提出一种基于形态学重构的边缘检测算法。利用形态学重构运算保持边缘的良好特性,采用多结构元方案,设计形态学重构滤波器对细胞图像进行去噪处理,利用形态学梯度检测算子获取重构后的细胞图像边缘,对获得的多路细胞图像边缘进行加权处理,最终检测出细胞图像边缘。仿真结果表明算法检测效果优于传统边缘检测算子检测效果,检测出的细胞图像边缘连续且一致。The result of cell edge detection is an objective evidence for cell morphology analysis. To resolve the problem of traditional cell edge detection, a cell edge detection algorithm based on morphological reconstruction and Multi -structure is proposed. Making use of the super performance of the morphological reconstruction and multi - structure, a morphological reconstruction filter is designed to eliminate the noise. And the image of cell edges is obtained with the morphological gradient operators. Further more, the final cell edge image is obtained with weighted disposal. The experimental results indicate that the new method is better than the traditional edge detection method, and the cell image edges are consecutive and identical.

关 键 词：数学形态学 形态学重构 细胞图像 边缘检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]