To the best of our knowledge, there is no previous work
on the segmentation of fluorescent microscopy images
depicting the yolk of transgenic zebrafish. Image
segmentation techniques based either on thresholding or on
determining the regional edges of images alone are not
sufficient to analyze these images. Methods of deformable
models [6][7][8] – defined as curves or surfaces that deform
under the influence of internal smoothness or external image
forces – may not be appropriate as they can be trapped to a
local minimum by spurious edges or noise when they are
initialized far from the object boundary. Finally, statistical
shape and appearance models [9][10][11] provide inaccurate