Study On Segmentation Methods For Blood Cell Image Based On Genetic Algorithm

Many diseases clinical symptom is leukocytosis, intumescence of white blood cell or karyon's change of figure and quantity. So it is important to study the white blood cell figure and quantity. Correct segmentation on white blood cell is the precondition to recognize all kinds of the feature of white blood cell.Segmentation is one of the key procedures in digital image processing field and the basis of correct classification and recognition.Genetic algorithm (GA) is a sort of efficient, Parallel, full search method with its inherent virtues of robustness, parallel and self adaptive characters.Now it has been applied widely and perfectly in many study fields and engineering areas,in the image segmentation field, GA is usually used to get the threshold of image segmentation.The method based image segmentation,genetic algorithm's principle and application in the image division domain. In view of particularity about image data, take sharpens the overall situation search ability and the convergence rate as the goal, an improved genetic algorithm in image segmentation is proposed. In the new algorithms:coding with 2-dimention chromosome is adopted; initialization of population with stochastic and symmetrical methods is produced to keep the variety of the population; Otsu is used as fitness functions; in updated population aspect, a new individual is introduced.The Theory analyses quality evaluation of image segmentation.The improved genetic algorithm is compared with the traditional methods by region consistency, region contrast and running time.The experiment result shows that the method of the image segmentation based on genetic algorithm is much better. The method which makes the image goal clear, the division speed quick, and the division effect good, has certain application prospect.