The Research Of Medical Image Processing Method Based On SV Mathematical Morphology

Mathematical morphology is a very important image processing method, but the traditional mathematical morphology uses the structural elements with fixed shape and size, which easily leads to "over-processing" or "inferior treatment" of the image. Whether we choose structural elements with right shape, size and direction directly affects the morphological processing results of image. In order to obtain the better effect, it is needed to design structural elements adapted to each local characteristics of the image and the corresponding processing algorithms according to different applications. The formal establishment of the ideal of Spatially Variant Mathematical Morphology (SVMM) can be traced to the concept and application of structure function proposed by Serra in1988, which has laid a solid foundation for the extension of application of mathematical morphology. In SVMM, the structural elements can automatically adjust its size and shape according to the local neighborhood feature and structure of detected points of the image.Medical image is complex, which contains a large amount of noise and information with large difficult in treatment. It has very high accurate in acquisition of information. Modern medical images, acquired by various medical imaging techniques, provide a wealth of information on patient symptoms. And quick achievement of information of medical image has an important significance for helping doctors make the right diagnosis and provide the prescription. Research on medical image processing theory and technology is of great practical value, not only for clinical improvement and development, but also for the development of image processing theory and technology. More and more countries of the world are paying more attention to this field.Research on medical image processing theory and techniques include medical image filtering, edge extraction, segmentation and3D medical image reconstruction, and so on. Firstly, this paper summarizes the theory of mathematical morphology and various methods in image processing. The second, we analyze the mathematical properties of SVMM in details, and demonstrates that it has good adaptability for medical image processing. The paper has also studied medical image edge detection based on the SVMM, and discussed its relevant theories and methods. The experimental results showed that the SVMM has the greater advantage in maintaining image detail, greater robustness for the salt and pepper noise, and better connective in extraction of the edge.