空間頻率擷取使用類梯度運算子

Spatial Frequency Extraction using Gradient-liked Operator

林以鎬(Yi-Hao Lin)

指導教授 ： 蕭子健

國立交通大學/資訊學院/資訊科學與工程研究所/碩士(2018年)

摘要

多維度集合經驗模態拆解 (MEEMD) 常用於拆解影像，但其時間複雜度較高，在這裡介紹一個新方法：「空間頻率擷取使用類梯度運算子」，將運算時間減少十倍以上，且很適合平行化運算，在GPGPU版本中，將近有500 倍的效能提升！我們的新穎方法使用類梯度運算子評估不同半徑下的空間頻率，並且將梯度運算的結果積分到空間幀，其結果很類似於MEEMD 的成果。

關鍵字

二維經驗模態拆解 ； 多維度集合經驗模態拆解 ； 空間頻率

並列摘要

The multi-dimensional ensemble empirical mode decomposition (MEEMD) is usually used for temporal-spatial data decomposition. One of the major issue is its high time complexity. A new gradient-liked approach to mimic similar spatial data decomposition results with more than 10x speedup. The GPGPU version of our approach can reach 500x speedup. Our novel approach use gradient-liked operator to evaluate the spatial frequency on different radius and integral the gradient result to spatial frame which is similar to BIMFs of MEEMD.

並列關鍵字

Bi-dimensional Empirical Mode Decomposition ； Multi-dimensional Ensemble Empirical Mode Decomposition ； Spatial Frequency

參考文獻

[1] Huang, Norden E., et al. "The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis." Proceedings of the Royal Society of London A: mathematical, physical and engineering sciences. Vol. 454. No. 1971. The Royal Society, 1998.

[2] Wu, Zhaohua, and Norden E. Huang. "Ensemble empirical mode decomposition: a noiseassisted data analysis method." Advances in adaptive data analysis 1.01 (2009): 1-41.

[3] Nunes, Jean Claude, et al. "Image analysis by bidimensional empirical mode decomposition." Image and vision computing21.12 (2003): 1019-1026.

[4] Wu, Zhaohua, Norden E. Huang, and Xianyao Chen. "The multi-dimensional ensemble empirical mode decomposition method." Advances in Adaptive Data Analysis 1.03 (2009):339-372.

[5] Waskito, Pulung, et al. "Parallelizing Hilbert-Huang transform on a GPU." Networking and Computing (ICNC), 2010 First International Conference on. IEEE, 2010.