Abstract
Karyotyping is a common method in cytogenetic. Automatic classification of the chromosomes within the microscopic images is the first step in designing an automatic karyotyping system. This is a difficult task especially if the chromosome is highly curved within the image. The main aim of this paper was to define a new group of features for better representation and classification of chromosomes. this paper proposes classification & analysis of human chromosomes which includes the following steps i)we use image processing utilities and filter to remove noise .ii)the filtered image is then entered into segmentation algorithm to segment the image .iii)then the segments enter into two tracks for classifying chromosomes. the first one depends on image processing for measuring the length of chromosomes where the second one deals with initiating the feed forward neural network which is trained by means of back propagation algorithm. By using feed forward neural network and back propagation algorithm, width, position and the average intensity of chromosome was determined. back propagation algorithm achieves high accuracy with minimum training time, which makes it suitable for real-time chromosome classification in the laboratory.in our paper ,segmentation is done by using image processing and classification is done by using feed forward neural network and back propagation algorithm.
