Shallow-Water LiDAR Echo Signal Denoising Based on Improved EWT

Tingting Wei; Jiazhi Yang; Guoqing Zhou; Xiang Zhou; Xueqin Nong

doi:10.3788/CJL202148.1109001

Journals >Chinese Journal of Lasers >Volume 48 >Issue 11 >Page 1109001 > Article

Chinese Journal of Lasers
Vol. 48, Issue 11, 1109001 (2021)

Shallow-Water LiDAR Echo Signal Denoising Based on Improved EWT

Tingting Wei¹, Jiazhi Yang^1、*, Guoqing Zhou², Xiang Zhou^2、3, and Xueqin Nong⁴

Author Affiliations

¹College of Information Science and Engineering, Guilin University of Technology, Guilin, Guangxi 541006, China

²Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin University of Technology, Guilin, Guangxi 541006, China

³School of Microelectronics, Tianjin University, Tianjin 300072, China

⁴The 34th Research Institute of China Electronics Technology Group Corporation, Guilin, Guangxi 541004, China

show less

DOI: 10.3788/CJL202148.1109001 Cite this Article Set citation alerts

Tingting Wei, Jiazhi Yang, Guoqing Zhou, Xiang Zhou, Xueqin Nong. Shallow-Water LiDAR Echo Signal Denoising Based on Improved EWT[J]. Chinese Journal of Lasers, 2021, 48(11): 1109001 Copy Citation Text

show less

Fig. 1. Flow chart of improved EWT algorithm

Download full size

Fig. 2. Shallow-water LiDAR echo signal model. (a) Echo signal without noise; (b) echo signal with Gaussian white noise (SNR is 20 dB)

Download full size

Fig. 3. Processing results of signal model using improved EWT. (a) Signal with noise and its EMFs (SNR is 20 dB); (b) comparison of signal with noise before and after denosing

Download full size

Fig. 4. Shallow-water LiDAR experimental system. (a) Principle diagram; (b) practical map

Download full size

Fig. 5. Measured waveforms under different water depths. (a) Δd_L1=2.127 m; (b) Δd_L1=1.621 m; (c) Δd_L1=1.434 m; (d) Δd_L1= 0.774 m

Download full size

Fig. 6. Denoised waveforms under different water depths. (a) Δd_L1=2.127 m; (b) Δd_L1=1.621 m;(c) Δd_L1=1.434 m; (d) Δd_L1= 0.774 m

Download full size

Fig. 7. Spectral segmentation under different boundary detect methods. (a) LocalMax; (b) LocalMaxMin; (c) Scalespace; (d) improved EWT

Download full size

Fig. 8. EMFs under different boundary detect methods. (a) LocalMax; (b) LocalMaxMin; (c) Scalespace; (d) improved EWT

Download full size

Fig. 9. Bathymetric errors of different denoising methods

Download full size

Type of denoising algorithm	Denoising algorithm	MSE /m	MAE /m	SNR /dB	PSNR /dB	T /s	Crosscorrelation
WT	Soft thresholding	0.397	0.478	19.706	29.773	1.010	0.993
	Hard thresholding	0.363	0.468	20.097	30.164	1.040	0.993
	Fixed thresholding	0.688	0.625	17.321	27.388	1.150	0.987
	EMD	0.306	0.393	20.843	30.909	1.300	0.995
	EEMD	0.321	0.434	20.634	30.701	3.890	0.994
	CEEMDAN	0.216	0.359	22.350	32.417	5.761	0.996
Type of denoising algorithm	Denoising algorithm	MSE /m	MAE /m	SNR /dB	PSNR /dB	T /s	Crosscorrelation
EWT	LocalMax	0.867	0.754	16.315	26.382	0.590	0.988
	LocMaxMin	1.045	0.854	15.504	25.571	0.510	0.981
	Scalespace	0.288	0.438	21.093	31.159	0.770	0.994
	Improved EWT	0.183	0.326	23.071	33.138	0.560	0.997

Table 1. Comparison of denoising performance parameters

Denoising method	Group	Surface echo			Bottom echo			Δd_L2 /m	Δd_L1 /m
Denoising method	Group	Position /ns	Height /V	Width /ns	Position /ns	Height /V	Width /ns	Δd_L2 /m	Δd_L1 /m
CEEMDAN	1	19.100	0.322	1.995	38.155	0.490	2.127	2.149	2.127
	2	18.301	0.186	1.647	32.010	0.238	2.375	1.546	1.621
	3	18.136	0.408	3.774	30.342	0.606	4.697	1.376	1.434
	4	15.855	0.258	2.072	22.767	0.329	2.480	0.776	0.774
LocalMax boundary detect method	1	23.672	0.480	6.502	42.001	0.400	5.673	2.067	2.127
	2	24.068	0.098	4.845	33.256	0.168	4.117	1.036	1.621
	3	18.593	0.473	3.841	29.067	0.594	2.749	1.181	1.434
	4	16.048	0.321	1.881	22.585	0.861	2.500	0.737	0.774
Denoising method	Group	Surface echo			Bottom echo			Δd_L2 /m	Δd_L1 /m
Denoising method	Group	Position /ns	Height /V	Width /ns	Position /ns	Height /V	Width /ns	Δd_L2 /m	Δd_L1 /m
Improved EWT method	1	22.708	0.445	5.003	40.684	0.527	5.136	2.027	2.127
	2	19.467	0.151	2.698	33.992	0.231	3.195	1.638	1.621
	3	15.563	0.446	2.258	28.726	0.644	2.379	1.484	1.434
	4	16.084	0.386	2.110	22.284	0.848	2.777	0.699	0.774

Table 2. Echo signal decomposition

Tingting Wei, Jiazhi Yang, Guoqing Zhou, Xiang Zhou, Xueqin Nong. Shallow-Water LiDAR Echo Signal Denoising Based on Improved EWT[J]. Chinese Journal of Lasers, 2021, 48(11): 1109001

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information