The Regularized Marching Cubes Algorithm for 3D Reconstruction from Medical Image

The standard marching cubes(MC) algorithm is not suitable for real-time rendering of amounts of triangles.An improvement of MC named regularized marching cubes(RMC),which can significantly reduce the number of triangles,is presented in this paper.Though constructing a look-up table to find isosurface,this algorithm needn't judge every cube of the volume data so as to play a judging role in reducing the number of cube,whereby saving the computing time.Meanwhile,in the process of merging the isosurfaces,by determining the isosurface seed and setting thresholds,the RMC algorithm adaptively merges surfaces contained in the judged cells and can reach expected results of different precision.The new method can speed up the three-dimensional reconstruction of medical images as long as the error is within user-specified error range.