The Research On The Macrophages Image Segmentation Algorithm And System Implementation

Macrophages, as a kind of immune cells in the body, plays a very important role in the body's resistance to infection, anti-tumor immune regulation, etc. Because the shape of the macrophages change is big. and number is large, if relys on the traditional cell analysis and detection methods, namely the doctor through the microscope to observe relevant information, not only to the doctor that demand is higher, but also very time consuming, and the obtained result also has some subjectivity. So far, domestic and foreign have developed medical cell image analysis systems are mostly for blood, urine and cancer cells. In this paper, we research macrophages image analysis system, using image processing technology and medical research, and the combination of the corresponding activation thing activated macrophages, the fluorescence microscope after taking pictures of the cell image, qualitative and quantitative image analysis, and we establish a set of macrophage image analysis system in the VC++environment.The main contents of this dissertation are as follows. The research situation of the image analysis system and the significance in medical applications are reviewed. The image preprocessing technologies are researched. The image is preprocessed with morphological processing algorithm to make the cell image bright, to extract the edge of macrophages, the image segmentation, the target area recognition, regional shape extraction, etc. Because the segmentation results directly influence the following parameters extraction, the watershed algorithm is studied. The color image watershed segmentation method based on the marking is studied, and at the same time, the force minimum technology is utilized to restrain the over-segmentation phenomenon. The watershed segmentation algorithm based on the distance transformation is studied, found that it is easy to cause the over-segmentation phenomenon because of the fluctuation of extreme value point. The method of the merger of the seeds point before the watershed transformation based on distance transformation is put forward, it can well restrain the over-segmentation phenomenon. Compared with the original method, the method can get a more accurate segmentation result, can meet the need of the actual analysis. The image analysis system structure and its processing module are introduced in detail. The analysis system is realized by using VC++language. The system software consists of image pretreatment, image segmentation and macrophages feature extraction function module, the performance of the analysis system meets the user's technical requirements. Finally, the analysis system can also realize the macrophages parameters extraction and can save and print information and so on. to provide medical personnel accurate judgment basis.