Iterative histogram matching algorithm for chromosome image enhancement based on statistical moments

Seyed Pooya Ehsani, Hojjat Seyed Mousavi, Babak Hossein Khalaj

Abstract: Vivid banding pattern of the chromosome image is a crucial part for diagnosis in karyotype medical test. Furthermore, thriving computer aided segmentation and classification depend on the initial image quality. In this paper, we propose an adaptive and iterative histogram matching algorithm for chromosome contrast enhancement especially in banding patterns which is one of the most important information laid in chromosome image. Objective histogram, with which the initial image needs to be matched, is created based on processes on the initial image histogram. Calculation of statistical moments of image histogram and determination of parameters in each step of iteration based on these moments form the algorithm structure. In order to assess the performance of the proposed algorithm, Density Profile (DP) and the contrast improvement ratio (CIR), are utilized. Experimental results indicate that the proposed method shows better results in both quantitative gauges of CIR and DP, and visual perception.