期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2014
卷号：7
期号：3
页码：181-192
DOI：10.14257/ijsip.2014.7.3.15
出版社：SERSC
摘要：When the cytological smear method is used to check cervical lesions, the irregularity of the cells contour is essential for smear image interpretation and has the vital significance for research on computer aided diagnosis of cervical lesions. Aimed at measuring the regularity of cells outlines, based on the characteristic that the normal cell smears is generally circular, oval, convex polygon, we put forward a cell contour irregularity feature extraction and measurement method based on linear geometric heat flow curve evolution, in which the curve of cells outline is dealt with linear geometric heat flow curve evolution to totally convex (i.e. no curvature zero crossing) and compared with the original contour to extract indexes which can describe irregularity features of cell contour. Experimental results show that the method proposed in this paper can better describe the irregularity of cell contour and distinguish cells of different lesions.
