The Research And Applycation On 3-D Reconstruction Algorithm Based On CT Image

Medical CT image 3-d reconstruction is an important research branch in scientific visualization, is a very useful technique. In CT images for the original data, the target organ or tissue within the 2-d image information of refactorings, the use of computer graphics algorithm, reconstruction in three dimensional space, and then presented to the user. Doctor can make more intuitive, multi-angle and comprehensive analysis of the target organ or tissue transformation, to make a more accurate assessment of the patient condition and diagnose the illness. In addition to diagnosis, as well as simulation multi-angle scanning, radiation therapy, surgical navigation, the foundation of the new emerging medical technologies such as remote surgery.At the same time, based on 3-d reconstruction of point cloud data research has made great progress. Point cloud data is the most intuitive to the real world entities, the description of the natural way. At present, the use of a laser scanning device to obtain the machine parts of 3-d information, and then based on these information for the construction of a 3-d model has been very mature. In recent years, especially, the further development of 3-d measurement technology, location information acquisition device of civilian and the emergence of 3-d printing technology, more make point clouds reconstruction has become a hot topic. Further promote the research and development of the human imagination, it is allowed as long as the input of any building model, this model can be the corresponding three-dimensional entity "print", put on your desk.Based on the CT scan data of human body big legs in raw data, CT data is converted into BMP format image first, then the image preprocessing, threshold processing, morphology processing, thinning and other operation), then the canny operator for image edge detection, search by using the method of neighborhood search the outermost contour in the image. The outermost contour as the basis of the data.In this paper, using the extracted contour, based on the cubic B spline curve for 3-d reconstruction of CT images, generate a layer and interlayer continuous second derivative of the human body curve of the thigh bone; This way the generated surface fairing sexual level is higher, pane uniform, can block noise, etc.This article will be treated as emerging the thinking of "three dimensional point cloud" is applied to the contour points are extracted from CT, put forward a feasible new direction of 3-d reconstruction of CT images; The outline on the sample point as the point cloud processing. This article USES the Crust algorithm for contour sampling point surface reconstruction, the end result is a triangular mesh, and the present main support 3-d printers STL format of triangular mesh model is appropriate, can be combined with 3-d printer to print the reconstruction of the organ. The method with the traditional CT image 3-d reconstruction based on curve interpolation, compared with fast speed (no need to solve the equations), topological structure adaptability is strong, reduce the human intervention. But this method is sensitive to noise. The paper finally, discussed the further work in the future.