DocumentCode :
671053
Title :
A parallel root-finding method for omnidirectional image unwrapping
Author :
Nguan Soon Chong ; Wong, M.L.D. ; Yau Hee Kho
Author_Institution :
Swinburne Univ. of Technol. (Sarawak Campus), Kuching, Malaysia
fYear :
2013
fDate :
17-20 Nov. 2013
Firstpage :
1
Lastpage :
6
Abstract :
The panoramic unwrapping of catadioptric omnidirectional view (COV) sensors have mostly relied on a precomputed mapping look-up table due to an expensive computational load that generally has its bottleneck occur at solving a sextic polynomial. However, this approach causes a limitation to the viewpoint dynamics as runtime modifications to the mapping values are not allowed in the implementation. In this paper, a parallel root-finding technique using Compute Unified Device Architecture (CUDA) platform is proposed. The proposed method enables on-the-fly computation of the mapping look-up table thus facilitate in a real-time viewpoint adjustable panoramic unwrapping. Experimental results showed that the proposed implementation incurred minimum computational load, and performed at 10.3 times and 2.3 times the speed of a current generation central processing unit (CPU) respectively on a single-core and multi-core environment.
Keywords :
image processing; multiprocessing systems; parallel architectures; polynomials; table lookup; COV sensors; CPU; CUDA platform; catadioptric omnidirectional view sensors; central processing unit; compute unified device architecture; multicore environment; omnidirectional image unwrapping; on-the-fly computation; panoramic unwrapping; parallel root-finding method; precomputed mapping look-up table; sextic polynomial; single-core environment; Central Processing Unit; Graphics processing units; Instruction sets; Kernel; Mirrors; Multicore processing; Real-time systems; CPU; CUDA; GPU; Image unwrapping; Multi-core; Omnidirectional Sensor; Parallelisation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Visual Communications and Image Processing (VCIP), 2013
Conference_Location :
Kuching
Print_ISBN :
978-1-4799-0288-0
Type :
conf
DOI :
10.1109/VCIP.2013.6706340
Filename :
6706340
Link To Document :
بازگشت