Two intelligent algorithms applied to automatic chromosome incision

The state-of the-art in commercial chromosome image acquisition was mostly grayscale. Automated Giemsa-banded chromosome image analysis has been widely used in clinical and laboratory examination. The decomposition of touching and overlapping chromosomes is a rather difficult task during the whole analysis procedure. Two intelligent chromosome incision methods are presented in this paper: the contour characteristic method and incision based on Fourier transform. The former method makes full use of the characteristic of the object contours to be decomposed, while the latter method takes advantage of the Fourier transform. By the combination of these two methods, most cases of conglutinated chromosomes have been solved, so they can play an important and excellent role in the automatic analysis of chromosome images, and greatly alleviate the manual operation of the doctor in chromosome incision.