Research on Parallel Algorithm of Sea Surface Oil Spill Identification Based on GPU

Abstract:

To overcome the drawback that spectral matching is time consuming because of the large volume of remote sensing data, a new parallel algorithm of sea surface oil spill identification based on Graphic Process Unit (GPU) is presented in this paper by taking advantage of HJ-1 CCD remote sensing data. Taking minimum distance classification as an example, we reorganized the pre-processed image data, selected reference spectral data and stored in constant memory, and then the algorithm core code of the host was transplanted to the device so as to realize parallelization on NVIDIA GeForce GTX 550Ti device. Results showed that, the maximum speed-up ratio is up to 102. Finally, the advantage of GPU parallel computing for computationally intensive problems has also been validated through experiments. When the image size is 2048x2048, speed-up ratio is up to 89.

• Periodical: Advanced Materials Research (Volume 823)
• Pages: 631-635
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.823.631
• Online since: October 2013
• Authors: Yan Zhou, Bin Zhou, Ying Ying Gai
• Keywords: Graphic Processing Unit (GPU), Oil Spill, Parallel Computing, Speed-Up Ratio

References

[1] Guillem Chust, Yolanda Sagarminaga. The multi-angle view of MISR detects oil slicks under sun glitter conditions. Remote Sensing of Environment, 107(2007) : 232-239.

DOI: 10.1016/j.rse.2006.09.024

[2] Hu C, Weisberg R H, Liu Y, et al. Did the northeastern Gulf of Mexico become greener after the Deepwater Horizon oil spill?. Geophysical Research Letters, 2011, 38, L09601, doi: 10. 1029/2011GL047184.

DOI: 10.1029/2011gl047184

[3] Jingyu Yang. The technology and realize method of RS image parallel process base on GPU. The PLA Information Engineering University, (2008).

[4] M. Blom,P. Follo. VHF SAR image formation implemented on a GPU. Geoscience and Remote Sensing Sylllposium 2005, (2005).

DOI: 10.1109/igarss.2005.1526560

[5] J. Setoain,C. Tenllado,M. Prieto,D. Valeneia,A. Plaza,J. Plaza. Parallel Hyperspectral Image Processing on Commodity Graphics Hardware. Proceedings of the 2006 International Conference on Parallel processing Workshops (ICPPW'06), (2006).

DOI: 10.1109/icppw.2006.60

[6] Rulin Xu, Haifang Zhou, Jingfei Jiang. Design and implementation of a parallel algorithm of the IHS- and wavelet- based image fusion for remote sensing based on GPU. Computer engineering & science, 2012, 34(8): 135-141.

DOI: 10.2991/icmt-13.2013.195

[7] Kirk David B, Wen-mei W. Hwu. Programming massively parallel processors: a hands-on approach. Morgan Kaufmann, (2010).

[8] NVIDIA TESLA on http: /www. nvidia. cn/object/tesla-servers-cn. html.

[9] List of Intel Xeon microprocessors on http: /en. wikipedia. org/wiki/List_of_Intel_Xeon_microprocessors#. 22Westmere-EX. 22_. 2832_nm. 29_Expandable.