Abstract
Most of the known algorithms can render images very quickly, however, only few of them are suitable for real-time volume rendering, and among them the shear-warp factorization is the most promising method. In this paper the shearing unit has been built and implemented in VHDL. The entire unit has a module structure, which can be easily modified and adjusted to any size of data set and its representation. The design contains two derived parts: standard image module and intermediate image module. Each part, in turn, consist of the controller (or control path) and the data path. Both controllers are composed concurrently and synchronized by two in/out signals. The volume slice is performed as the bitstream of voxels' stack. Each voxel is multiplied by scaling coefficients of the transformation matrix, which is chosen respectively to the viewing axis, then added to the intermediate image and packed into registers.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have