Abstract :
Accuracy is one of the most important requirements of a reconstructed surface. However, this criterion needs high computation too since large number of data points is involved. Therefore, distribution of data points is the best way to solve this problem. In this paper, dyadic rational technique is employed in segmenting the data points into several processors. Then, the surface reconstruction process is done independently in each processor. Finally, all the surfaces are combined in the master processor. This project used low-cost self-developed parallel laboratory named Ars Cluster in UiTM Shah Alam. Therefore, the parallel process in this project is time-saving and cost-saving.
