Iterative thresholding for segmentation of cells from noisy images

WU, H.-S.
BARBA, J.
GIL, J.

*Department of Pathology, Box 1194, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029, U.S.A.
†Department of Electrical Engineering, City College of New York, 140th Street & Convent Avenue, New York, NY 10031, U.S.A.

Received 19 February 1997; accepted 25 August 1999

Correspondence to: Professor Hai-Shan Wu, Department of Pathology, Box 1194, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029, U.S.A. Tel: +1 (212) 241 0468; fax: +1 (212) 650 5342; e-mail: [email protected]

Journal of Microscopy 197(3):p 296-304, March 2000.

Summary

We introduce an iterative thresholding algorithm for the segmentation of cells from noisy cell images. The thresholding image, which is initially a constant, changes iteratively with both the previous segmentation and image local activity. Experimental results for both synthesized and real cell images are provided to demonstrate the performance of the algorithm.