Optimal Threshold Estimation using Grey Wolf Optimization for EMD-DTCWT Based ECG Denoising
Deepak H. A.1, T. Vijayakumar2
1Deepak H. A., Research Scholar, Department of E&C, SJBIT, Bangalore, Affiliated to VTU, Belagavi, India.
2Dr. T. Vijayakumar, Professor, Department of E&C, SJBIT, Bangalore, Affiliated to VTU, Belagavi, India.
Manuscript received on March 16, 2020. | Revised Manuscript received on March 24, 2020. | Manuscript published on March 30, 2020. | PP: 2589-2596 | Volume-8 Issue-6, March 2020. | Retrieval Number: F8556038620/2020©BEIESP | DOI: 10.35940/ijrte.F8556.038620
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The noise reduction in the ECG has been focused for research in recent years, since desired reduction allows a better signal pre-processing, and allows to extract from it the maximum amount of efficient and meaningful information. This paper proposes an adaptive threshold technique using Empirical Mode Decomposition (EMD) and Dual-Tree Complex Wavelet Transform (DTCWT) for ECG signal denoising. Initially the data driven EMD technique is applied to get the IMFS and these IMFS further passed though DTCWT for filtration. To accomplish the better adaptive filtering process the optimal threshold is further calculated based on Grey Wolf Optimization (GWO). The performance evaluation is achieved on MIT-BIH database.
Keywords: Cross Correlation, DTCWT, DWT, ECG, EEG, EMG, EMD, GWO, IMF, MAE, MSE, PSNR, SNR
Scope of the Article: Discrete Optimization.