Research On Tracking For Cell Based On Video Technology

Research of digital image processing and computer vision is developing very rapidly in recent years. It is now widely used in military and civil applications, and is becoming the most important way for intelligent machines to access external information and understand the world. Movement detection and object tracking,topics of this thesis, are the two most important applications of computer vision.In the pharmaceutical process ,the traditional cell movement research method was unsuitable in the technology fast developing period, it not only need some boring human manual operations but also be the use of dyeing chemicals,influence of cells movement, bring certain effect to pharmaceutical,so,it should need an objective and convenient method. really need an objective and convenient method. It was no doubt that using computer vision and image technology to analysis cell moving was a good idea, objective and precise, can be systematic and integrated with the equipment easily, so more and more scientists who studied cell moving paid much attention to it. scientists who studied cell moving paid much attention to it. This was the background of the subject.In this paper, based on theory of computer vision, motion detection and tracking of cells to make a certain amount of exploration, to provide new technical support for Cellular and biological study.In this paper, the author analyzes the characteristics of image,then study of classical motion detection and tracking algorithm, and application in cell detection and tracking. First, for motion detection According to the characteristics of cell image and using grayscale morphology to establish the background, and then through the background subtraction method to detect moving target. For adhesion of cells, put forward cell separation method based on context and effectively resolved Second,for inert cell tracking, the author studies of a related target tracking algorithm, and according to cell motion characteristics, using the classical MeanShift algorithm (Meanshift) track on the inert cells.Again,for the dividing cells of the track,according to the characteristics of cell division, with Meanshift tracking algorithm to one of the daughter cells, and then according to the characteristics of area distance for daughter cells after cell division to track.For some specific phenomenon of cells in the movement,such as such as how to track the transition of the cell and so on,the author studies of a related prediction model and puts forward the grey prediction model to predict the cells in a frame, while gives the corresponding theoretical analysis process, and then cell tracking.In order to verify the validity of the model, the paper also proposed using the centroid method to predict the cell location, and then with the Grey forecasting model predicted results were compared, The results indicate that grey forecasting model prediction cells than the qualitative method of forecasting position to accurately position.