Medical Image Segmentation With ITK

As a new rising technology, image processing obtains rapid development in the past several years. Its applications can be found in various fields such as medical, military affairs, industry and so on. Segmentation is the most important aspect in image processing and computer vision field, which is always viewed as the basis of image analysis and image comprehension. The segmented result has a great affect on the capability of following recognition and analysis algorithms. However, since images' variety and complexity, until now there isn't a segmentation algorithm that fits all kind of images. For this reason, the segmentation technology has been a hotspot in image processing research fields for a long time.Segmentation of medical images is one of the important applications of image segmentation which obtains extensive concerns of many computer image processing researchers. The segmentation methods of medical images become the one-up research task in present image domain. We firstly retrospect the definition and classification of segmentation, then introduce some traditional and new segmentation methods in detail.The level set method based on geometric deformable model, which translates the problem of evolution of 2-D(3-D) close curve(surface) into the evolution of level set function in the space with higher dimension. However, as the traditional level set method just using the local marginal informations of the image, it is difficult to obtain a perfect result when the region has a fuzzy or discrete boundary, and the leaking problem was inescapable appeared. In this paper, we introduce Canny-Edge Level Set Segmentation method. The experimental results show that this improved algorithm can balance calculated efficiency and segmentation accuracy. Furthermore it can enhance the robustness of Level Set method.The hybrid segmentation approach integrates boundary-based and region-based segmentation methods that amplify the strength but reduce the weakness of both techniques. These algorithms achieve satisfactory segmentation results and play an important role in the automaticity of medical image segmentation. Image registration is the key technology in the imaging operation navigation software. We solve the automatic allocation sequence of the two sets of landmarks and give the ICP surface fusion algorithm of laser registration. The strongpoint of laser registration is retrospective, non-contact and automatic. The work of image registration is an important supplement to the work of image segmentation.