Wavelet and Empirical Mode Decomposition Denoising for GLAS Full Waveform Data

Zhenhua Wang; Xiaodan Liu; Xiangfeng Liu

doi:10.3788/LOP202158.2328001

Journals >Laser & Optoelectronics Progress >Volume 58 >Issue 23 >Page 2328001 > Article

Laser & Optoelectronics Progress
Vol. 58, Issue 23, 2328001 (2021)

Wavelet and Empirical Mode Decomposition Denoising for GLAS Full Waveform Data

Zhenhua Wang¹, Xiaodan Liu¹, and Xiangfeng Liu^2、*

Author Affiliations

¹Department of Information, Shanghai Ocean University, Shanghai 201306, China

²Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 200083, China

show less

DOI: 10.3788/LOP202158.2328001 Cite this Article Set citation alerts

Zhenhua Wang, Xiaodan Liu, Xiangfeng Liu. Wavelet and Empirical Mode Decomposition Denoising for GLAS Full Waveform Data[J]. Laser & Optoelectronics Progress, 2021, 58(23): 2328001 Copy Citation Text

show less

Typical ground feature map and echo waveform of GLAS data. (a) Woodland without slope; (b) woodland with slope; (c) flat land without slope; (d) flat land with slope;(e) buildings without slope; (f) buildings with slope

Fig. 1. Typical ground feature map and echo waveform of GLAS data. (a) Woodland without slope; (b) woodland with slope; (c) flat land without slope; (d) flat land with slope;(e) buildings without slope; (f) buildings with slope

Download full size

Denoising results of 5 methods on GLAS data. (a) Woodland without slope;(b) woodland with slope; (c) flat land without slope; (d) flat land with slope; (e) buildings without slope; (f) buildings with slope

Fig. 2. Denoising results of 5 methods on GLAS data. (a) Woodland without slope;(b) woodland with slope; (c) flat land without slope; (d) flat land with slope; (e) buildings without slope; (f) buildings with slope

Download full size

Fig. 3. Noise reduction accuracy of different methods. (a) SNR; (b) RMSE

Download full size

Method	Precision index	Woodland without slope	Woodland with slope	Flat land without slope	Flat land with slope	Buildings without slope	Buildings with slope
Gaussian	SNR	30.5210	27.1948	51.2415	45.5369	18.4426	22.3167
Gaussian	RMSE	0.0143	0.0155	0.0223	0.0174	0.0057	0.0096
WTD-Soft	SNR	37.6608	33.4901	63.8431	55.9938	32.6993	28.0127
WTD-Soft	RMSE	0.0100	0.0113	0.0119	0.0103	0.0028	0.0072
WTD-Improved	SNR	47.4859	45.7251	70.7825	65.9482	36.1995	40.8214
WTD-Improved	RMSE	0.0066	0.0066	0.0029	0.0070	0.0018	0.0013
EMD-WTD	SNR	27.4859	25.7251	50.7825	45.9482	12.1995	20.8214
EMD-WTD	RMSE	0.0166	0.0166	0.0229	0.0170	0.0078	0.0103
EMD-Hurst	SNR	34.7496	29.4438	54.8620	48.8775	20.2148	28.1858
EMD-Hurst	RMSE	0.0116	0.0138	0.0186	0.0147	0.0052	0.0072

Table 1. Noise reduction accuracy of 5 methods on GLAS data

Zhenhua Wang, Xiaodan Liu, Xiangfeng Liu. Wavelet and Empirical Mode Decomposition Denoising for GLAS Full Waveform Data[J]. Laser & Optoelectronics Progress, 2021, 58(23): 2328001

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information