Please use this identifier to cite or link to this item:
|標題:||Edge enhancement nucleus and cytoplast contour detector of cervical smear images||作者:||Yang-Mao, S.F.
|關鍵字:||cervical cancer;cervical smear screening;Gaussian noise;image;segmentation;impulse noise;scanning laser microscopy;human skin;confocal microscopy;segmentation||Project:||Ieee Transactions on Systems Man and Cybernetics Part B-Cybernetics||期刊/報告no：:||Ieee Transactions on Systems Man and Cybernetics Part B-Cybernetics, Volume 38, Issue 2, Page(s) 353-366.||摘要:||
This paper presents an edge enhancement nucleus and cytoplast contour (EENCC) detector to enable cutting the nucleus and cytoplast from a cervical smear cell image. To clean up noises from an image, this paper proposes a trim-meaning filter that can effectively remove impulse and Gaussian noises but still preserves the sharpness of object boundaries. In addition, a bigroup enhancer is proposed to make a clear-cut separation of the pixels lying in-between two objects. A mean vector difference enhancer is presented to suppress the gradients of noises and also to brighten the gradients of object contours. What is more, a relative-distance-error measure is put forward to evaluate the segmentation error between the extracted and target object contours. The experimental results show that all the aforementioned techniques proposed have performed impressively. Other than for cervical smear images, these proposed techniques can also be utilized in object segmentation of other images.
|Appears in Collections:||資訊管理學系|
Show full item record
TAIR Related Article
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.