Chromosome classification for karyotype composing applying shape representation on wavelet packet transform

Karyotype is an image containing seven groups of chromosomes disposed in pairs. Since a chromosome photo presents the chromosome in random disposition the human interaction become necessary. As the digital image processing techniques have been improving, many techniques have been proposed to make the karyotyping an automatic process. The adopted method depending on the characteristics of the original chromosome image as color or distribution of the chromosome bands. This paper proposes a method for chromosome classification based on the chromosome shape. The chromosome shape is converted in a signature that is decomposed by the wavelet packet transform and its best basis. Then, the wavelet packet coefficients correspondent to the signature of chromosomes are compared in order to classify the chromosome in one of the six groups, A, B, C, D, E, F and G, of the karyotype. The results obtained show that the proposed method provides the classification of the chromosomes.