Abstract:
An automated method has been developed to identify the dark bands and light-inter bands in
the Polytene chromosomes microscopic images using the digital image processing
algorithms. Polytene chromosomes are specific interphase large chromosomes which have
alternation of dark bands and lighter inter bands morphology. Analysis of these banding
patterns is very important in many structural and functional biological researches such as
chromosomes mapping, identify small chromosome mutations, identify taxonomic species,
and classify the sibling species.
The algorithm of banding pattern identification and analysis follows a step-by-step strategy
and the overall system is modularized into three stages: preprocessing, segmentation and
feature extraction. The preprocessing stage contains the techniques to prepare the image for
segmentation; the techniques deal directly with the raw, possibly noisy, pixel values with de noising and contrast adjustment. The segmentation stage finds and establishes outlines of
specific dark/lighter-inter bands using rolling ball algorithm and automatic thresholding
techniques. The feature extraction stage obtains semantics of the banding pattern and useful
quantitative parameters from the segmented bands. In this work, some samples of
straightened Polytene chromosome microscopic images were analyzed with satisfactory
results.
