The Improved Parallel Ray Casting Algorithm in Embedded Multi-core DSP System
An improved parallel ray casting algorithm in embedded multi-core DSP system is proposed in this paper. In order to speed up the process of intersection, the algorithm takes advantage of the improved bounding volume hierarchy (BVH) technology. By parallelizing and optimizing ray casting algorithm on TMS320C6678 of TI, the computing efficiency greatly increases. Master-slave mode is adopted to schedule the tasks and balance the load between 8 cores in parallel processing. Moreover, the code and Cache is optimized on the hardware platform. The experimental result shows that our method has good performance and exceeds the result of OpenMP (Open Multiprocessing). This paper proposes a new method to run rendering algorithm on embedded platform efficiently. The rendering algorithm usually runs on GPU, due to GPU’s powerful computing capability.
KeywordRay casting BVH DSP Parallel computing Task scheduling Load balancing
- 1.Bruckner, S., Groller, M.E.: VolumeShop: an interactive system for direct volume illustration. In: Proceedings of the IEEE Visualization, Minneapolis, pp. 67l–678 (2005) Google Scholar
- 2.Goswami, P., Zhang, Y., Pajarola, R., Gobbetti, E.: High quality interactive rendering of massive point models using multi-way kd-trees. In: 18th Pacific Conference on Computer Graphics and Applications (PG), pp. 93–100. IEEE (2010)Google Scholar
- 3.Kraus, M., Strengert, M., Klein, T., Strengert, M.: Adaptive sampling in three dimensions for volume rendering on GPUs. In: 6th International Asia-Pacific Symposium on Visualization 2007, pp. 113–120. IEEE (2007)Google Scholar
- 4.Ling, T., Zhi, Y.Q.: An improved fast ray casting volume rendering algorithm of medical image. In: 4th International Conference on Biomedical Engineering and Informatics, pp. 109–112. IEEE (2011)Google Scholar