Abstract
The recognition of pre-cancer abnormalities in Pap smear images by computer is discussed. Classification systems discussed treat these images as the source of feature vectors that are multi-modal and can be classified using the Bayes decision model with knowledge of the various subclasses that comprise the sample. A cell classification system must be capable of using information about the various subclasses of normal and abnormal cells. Two parametric decision rules are shown which are applicable to this multi-modal problem, and are suitable for the classification required for automated detection of atypical cells in a cervical smear. Test results from a series of experiments indicate that correct recognition rates of about 85% can be achieved on the atypical cells, while maintaining an error rate of about 1% on the normal cells.
Original language | English (US) |
---|---|
Pages | 476-448 |
Number of pages | 29 |
State | Published - 1978 |
Event | Proc IEEE Comput Soc Conf Pattern Recognition Image Process - Chicago, IL, USA Duration: May 31 1978 → Jun 2 1978 |
Other
Other | Proc IEEE Comput Soc Conf Pattern Recognition Image Process |
---|---|
City | Chicago, IL, USA |
Period | 5/31/78 → 6/2/78 |
ASJC Scopus subject areas
- General Engineering