Search

Surface reconstruction based on point cloud data has been extensively studied. In this paper, the original data has a lot of noise and without normal vector information, so we use Poisson surface reconstruction algorithm to create 3D heart model, which has good robustness for noisy and irregular point cloud data. Then correct the shape by removing or adding points from the model surface to perfect the heart model, and reconstruct the new model by Power Crust algorithm based on the correctional surface point cloud data with normal vector, which is quick, accurate and efficient and very suitable for fast model correction. The results not only accurately display the spatial relationship of the heart, but also can be discretionary scaling and rotation in 3D space. The visual 3D graphical structures of the reconstructed heart model can display diseases (cardiopathy & arrhythmia) and enable catheter navigation in real time, which can help doctors to detect and diagnose diseases effectively, and improve the accuracy and security of medical diagnosis.

In this paper, it has researched on the heart model, which reconstructed by passion algorithm from unorganized points, and it mainly research the distance between heart and catheter. It has some improvement in distance function. These point data obtained through the catheter moving on the heart, and the doctor treat the heart by the catheter too. The paper has provided help to the process of treatment, because the job measuring distance between catheter and heart is very important. When the distance is zero, it has got the optimal effect, at the same time optimal treatment can been obtained as the catheter touch the heart. It is three-dimension heart model in this paper, and the distance is also three-dimension straight distance. The catheter head is simulated by sphere with a certain radius, which must be considered when calculate the distance. It is an important technology in medical system, which has been achieved some achievement. The main development platform is VTK combined C++ language.

Unlike other 3D image visualization system based on Windows, this paper takes full use of the image processing advantages on MAC OS, with the library of development of the platform VTK, to achieve the medical reconstruction system. In this paper, we will not only discuss the environment building of VTK on MAC OS, but also achieve some functions of medical image system. The stability and feasibility of the system are being verified by the simulation results. (4 pages)

In order to meet the demand of real-time transmission of the terrain data and efficient display, based on the different view field of high altitude and surface, firstly, a general(efficient to both surface and high altitude) view field algorithm is proposed. This algorithm promotes the traditional vision clipping algorithm, efficiently determining the visibility of the view field, the LODs of the key nodes and previsible region when position of the vision moves. Pre-scheduling method is used to keep the calculation in order. Then progressive transmission strategy is adopted to transmit the processed data. Finally, the validity of the algorithm is approved by experiment.

This paper implements a novel surface rendering algorithm for three-dimensional reconstruction of medical images. Traditional Marching Cubes (MC) algorithm shows limits for isosurface extraction by using surface configurations of a cube, because it costs much time to detect null cube when abstracting an isosurface. This has an important effect on the reconstruction speed. To overcome the problem, an improved MC algorithm based on image segmentation is proposed. It uses segmented images as input, aiming at reducing the time of detecting null cube and finding out those cubes crossed by the isosurface as soon as possible. Extensive experiments show that the efficiency of three-dimensional reconstruction can be improved a lot by using the novel algorithm.