A region based decorrelation stretching method: Application to multispectral chromosome image classification

M-FISH (multiplex fluorescent in situ hybridization) is a multichannel channel chromosome imaging technique that allows the color discrimination of human chromosomes. Although M-FISH facilitates the visual detection of chromosome rearrangements, the success of this technique largely depends on the accuracy of the pixel-by-pixel classification. In this paper, we present a new method that enhances the classification ratio based on a region based decorrelation stretch transform. Utilizing the decomposition of the multichannel image into homogenous regions, the classification results improved after the application of the decorrelation stretch transform. Fifteen M-FISH images were used to test our method. To quantitatively estimate the enhancement we have used a Bayes pixel-by-pixel classifier and an improvement of 12.28% is achieved.