摘要
本文给出了宫颈癌细胞图像的处理、分析、特征提取及癌细胞的识别算法。从近50张涂片中输入共513幅各类细胞来验证该算法,识别率达到真阴性97.5％和真阳性90.9％,结果表明该算法完全可用于临床识别系统。
To be used with a cell image analysis system, the algorithms for image preprocessing, feature extraction and cell recognition are given here. The classification results show that the recognition rate is better than 97.5% (real negative) and 90.9% (real positive).
出处
《计算机工程》
CAS
CSCD
北大核心
1991年第2期29-33,72,共6页
Computer Engineering
