![](bg5.png)
Figure 8: reconstructed surface with composite contours
Figure 9: The reconstructed surface with composite
contours
Figure 9(a) shows the generated bunny using the
method developed in this paper, and Figure 9(b) is the
generated bunny using CPU-alone by (Abdul Hadi,
Ibrahim, Yahya, & Md Ali, 2013). In Figure 9(a), the
texture on the bunny body and paws are clearer. This
is because Figure 9(a) consists of 500 surface points
per slice, while Figure 9(b) has only 75 surface points
per slice due to the memory limitation of the CPU.
The CPU limitation also has disadvantage in
generating composite contour. Thus, some branching
part for example ear in Figure 9(b) is not smoothly
joined and can be obviously seen.
4 CONCLUSION
In this paper, an algorithm for generating composite
contour on the CPU-GPU platform has been
developed. The composite contour occurs when
the adjacent image slices have different number of
contours. The decrement of the processing time and
the improvement of the speedup of the developed
algorithm suggest that the CPU-GPU platform is
suitable to be employed in the composite contour
generation since the process involves huge number of
data points.
The capability of GPU also allows the number
of surface points to be high enough to produce the
smoothed and accurate surface. Furthermore, the
fitted cubic beta-spline surface has high continuity
(G2) to confirm the continuity of the surface.
For future research, more process will be
considered to be executed on the GPU. Further
analysis of the performance will also be studied to
ensure the quality of the developed method.
ACKNOWLEDGMENTS
This study is supported by Ministry of Education,
Malaysia, Universiti Teknologi MARA and Universiti
Teknologi Malaysia. Normi Abdul Hadi is a
researcher of Universiti Teknologi Malaysia under the
Postdoctoral Fellowship Scheme.
REFERENCES
Abdul Hadi, N. (2019). Digital khat calligraphy using
beta-spline curve on cpu- gpu platform.
Abdul Hadi, N., Ibrahim, A., Yahya, F., and Ali, J.
(2013). Composite contour generation for beta-spline
surface reconstruction. AIP Conference Proceedings,
1522:435–440.
Alias, N. and Kamal, M. H. A. (2017). Integration
of a big data emerging on large sparse simulation
and its application on green computing platform.
ARPN Journal of Engineering and Applied Sciences,
12(12):3817–3826.
Alias, N., Mohsin, H. M., Nadirah, M., Mustaffa, S.,
and Reyaz, R. (2016). Parallel artificial neural
network approaches for detecting the behaviour of
eye movement using cuda software on heterogeneous
cpu-gpu systems. Jurnal Teknologi, 78(12-2).
Barsky, B. A. (1981). The beta-spline: A local
representation based on shape parameters and
fundamental geometric measures.
Computer Graphics at Stanford University. (n.d.), .
Ge, T. (2018). Optimization of gpu-accelerated iterative ct
reconstruction algorithm for clinical use.
Hadi, N. (2018). Big data simulation for surface
reconstruction on cpu-gpu platform. In Journal
of Physics: Conference Series, volume 1192, page
012006. IOP Publishing.
Hadi, N. A. and Alias, N. (2019a). 3-Dimensional Human
Head Reconstruction using Cubic spline surface
on CPU-GPU Platform. International Conference
Proceedings Series by ACM.
Hadi, N. A. and Alias, N. (2019b). 3-Dimensional Human
Head Reconstruction Using Cubic Spline Surface on
CPU-GPU Platform. Proceedings of the 2019 4th
International Conference on Intelligent Information
Technology, 16–20. ACM.
The Performance of 3D Multi-slice Branched Surface Reconstruction on CPU-GPU Platform
53