宫颈上皮细胞图象的计算机分类

通过对约300个宫颈上皮细胞图象的分析研究，结果表明:(1)当将对象分为三类(正常、核异质与癌)时,三级线性判决的效果明显地优于一级判决;(2)从纯数学角度选取的非视觉特征对分类是重要的.在适当地选择训练集且限定每次判决所用特征数为五时,正确识别率可达75%.

The Computerized Sorting for Cervical Epitheliums

About 300 cervical epithelium images were analysed ,and studied as a whole. Theresults have shown that 1) when these images are devided into three classes (normal,dysplastic, cancer), tertiary linear decision effect is much better than the primary leveldecision and 2) non-visual features that is acquired from pure-mathematical methodare important for sorting.When numbers of feature to each decision is restricted to five and training setsare preperly selected, the right recognition ratio achieved is 75%.