基于粗糙集与水平集的细胞智能识别应用研究  

Research of Intelligent Identification for Cell Based on Level Set and Rough Set

作　　者：张二元[1] 孙艳红[2] 

机构地区：[1]重庆电子工程职业学院物联网学院,重庆401331 [2]重庆交通大学信息科学与工程学院,重庆400074

出　　处：《计算机测量与控制》2014年第12期4124-4126,共3页Computer Measurement &Control

基　　金：重庆市教委科学技术研究项目(KJ110401)

摘　　要：针对细胞图像特性微弱,提出改进的Chan-Vese水平集分割算法,利用水平集演化曲线实现对微生物感染的细胞核边缘的准确分割;针对宫颈细胞图像特征,首先提取细胞图像样本的五维特征向量(纹理、几何形状特征以及灰度图像等),然后根据粗糙集理论建立决策表,用于细胞图像的微生物识别,并直接从训练样本图像中导出决策规则;应用所获取的规则对微生物测试样本图像进行分类;实验结果表明该方案应用在宫颈细胞智能识别上可以达到较好的效果,在工程上应用是可行的。In the case of nucleus image with weak speciality,introducing improved Chan-Vese level set segmentation algorithm,using the level set evolution curve to the edge of the microbial nuclei realize accurate segmentation.The 5 dimensional feature vectors of intensity,texture and geometry characteristics for four kinds of cervical cells image microorganisms are extracted.The key technology of Rough Set theory is described.The decision table of cervical cells image microorganisms recognition is established,the decision rules are obtained from the training sample images directly.The test samples of cervical cells image microorganisms have been classified with application of decisions rules.The experimental results can be seen that the solution achieve good effect in automatic intelligent identification of Liquid-Based Cervical Cell Images,and is feasible in engineering.

关 键 词：水平集 粗糙集 智能识别 细胞 图像处理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]