Compensation for topographic effect on P-band PolSAR data with a polarimetric decomposition technique

Yin Zhang; Ding-Feng Duan

doi:10.1016/j.jnlest.2024.100292

Journal of Electronic Science and Technology, Volume. 23, Issue 1, 100292(2025)

Compensation for topographic effect on P-band PolSAR data with a polarimetric decomposition technique

Yin Zhang¹ and Ding-Feng Duan^2、*

¹Department of State-owned Assets and Laboratory Management, University of Electronic Science and Technology of China, Chengdu, 611731, China

²School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu, 611731, China

show less

Figures & Tables(12)

Fig. 1. Mt. Shasta area, California: (a) S_HH intensity image of the NASA/JPL AIRSAR P-band PolSAR data, (b) azimuth orientation angle (θ₀) image derived from the P-band PolSAR data, and (c) azimuth slope angle θ (°), image derived from DEM. The pixel size of DEM is 10 m by 10 m. Each image covers an area of about 3.8 km (width) by 5.3 km (height).

Download full size

View in Article

Fig. 2. Relationships of the radar look angle (θ_l), local incidence angle (θ_i), and range slope angle (θ_s): (a) θ_l ≥ θ_s and (b) θ_l < θ_s. The slope (or inclined surface) faces the radar look direction. (c) slope faces away from the radar look direction, and θ_l + θ_s ≤ 90°.

Download full size

View in Article

Fig. 3. Howland forest, Maine: (a) S_HH intensity image of the NASA/JPL UAVSAR P-band PolSAR data. θ₀ images, (b) topographic compensation, before, and (c) topographic compensation, after. The image is 570 columns by 570 rows, covering an area of about 8.8 km by 8.8 km.

Download full size

View in Article

Fig. 4. Mt. Shasta area, California. The θ₀ image after the topographic compensation.

Download full size

View in Article

Fig. 5. Mt. Shasta area, California. Scatterplots of backscatter of (a) |S_HH|², (b) |S_VV|², and (c) |S_HV|² before (x-axis) and after (y-axis) the topographic compensation algorithm.

Download full size

View in Article

Fig. 6. Eno river state park, North Carolina: (a) S_HH intensity images of DukeFr_225 and (b) DukeFr_045. The azimuth slope images calculated from the DEM data with the aspect angles of (c) 225°, (d) 45°, (e) local radar incidence angle images of DukeFr_225, and (f) DukeFr_045.

Download full size

View in Article

Fig. 7. Eno river state park, North Carolina: θ₀ images of (a) DukeFr_225 and (b) DukeFr_045 before the topographic compensation; (c) DukeFr_225, and (d) DukeFr_045 after the compensation.

Download full size

View in Article

Fig. 8. Altamaha wildlife management area, Georgia: (a) S_HH intensity image of the NASA/JPL AIRSAR P-band PolSAR data. One sample site at near range with a radar local incidence angle of about 36.25° and one at far range with an incidence angle of ~54.14°. Each square has a size of 50×50. θ₀ images (b) the topographic compensation, before, and (c) the topographic compensation, after. Each image covers an area of about 10.5 km by 4.8 km.

Download full size

View in Article

Table 1. Statistical results of θ₀ including the value range and the proportion of pixels with values > 5°.
View table
View in Article
Table 1. Statistical results of θ₀ including the value range and the proportion of pixels with values > 5°.
Main range of θ₀ Percentage of θ₀ > 5° (%)
Shasta Before [–20°, 20°] 45.2
After [–10°, 10°] 4.2

Table 2. Mean(μ) and standard deviation (σ) of backscatter components before and after the topographic compensation algorithm. The relative change rate (%) is Δ.

View table

View in Article

Table 2. Mean(μ) and standard deviation (σ) of backscatter components before and after the topographic compensation algorithm. The relative change rate (%) is Δ.

		${\left\| {{S_{{\text{HH}}}}} \right\|^2}$	${\left\| {{S_{{\text{VV}}}}} \right\|^2}$	$\left\| {{S_{{\text{HH}}}}S_{{\text{VV}}}^*} \right\|$	\|2S_HV\|²	\|$\sqrt 2 {S_{{\text{HH}}}}S_{{\text{HV}}}^*$	\|$\sqrt 2 {S_{{\text{VV}}}}S_{{\text{HV}}}^*$
Before	μ	0.401	0.207	0.125	0.169	0.089	0.061
Before	σ	0.391	0.294	0.262	0.129	0.083	0.061
After	μ	0.423	0.217	0.132	0.137	0.071	0.055
After	σ	0.400	0.302	0.263	0.111	0.069	0.056
	Δ	5.201	4.608	5.303	–23.358	–25.352	–10.909

Table 3. Mean value (μ) and standard deviation (σ) of the θ₀ values. The θ₀ < 0° andθ₀ ≥ 0° categories were created using DukeFr_225_mask for DukeFr_225 and DukeFr_0455_mask for DukeFr_045, respectively.
View table
View in Article
Table 3. Mean value (μ) and standard deviation (σ) of the θ₀ values. The θ₀ < 0° andθ₀ ≥ 0° categories were created using DukeFr_225_mask for DukeFr_225 and DukeFr_0455_mask for DukeFr_045, respectively.
θ₀ < 0° category θ₀ ≥ 0° category
Before After Before After
DukeFr_225 μ –2.39 0.47 3.44 –0.70
σ 2.11 1.22 2.40 1.21
DukeFr_045 μ –1.67 0.20 3.35 –0.69
σ 1.51 1.23 2.34 1.18

Table 4. Mean value (μ) and one standard deviation (σ) of the absolute θ₀ values at a sample size at the near range and the other site at the far range.
View table
View in Article
Table 4. Mean value (μ) and one standard deviation (σ) of the absolute θ₀ values at a sample size at the near range and the other site at the far range.
Near range Far range
¹Ratio = (mean before the compensation) / (mean after the compensation).
Before μ 0.88 1.02
σ 0.64 0.76
After μ 0.14 0.20
σ 0.11 0.18
¹Ratio 6.29 5.10

Tools

Get Citation

Copy Citation Text

Yin Zhang, Ding-Feng Duan. Compensation for topographic effect on P-band PolSAR data with a polarimetric decomposition technique[J]. Journal of Electronic Science and Technology, 2025, 23(1): 100292

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category:

Received: Mar. 1, 2024

Accepted: Nov. 14, 2024

Published Online: Apr. 7, 2025

The Author Email: Ding-Feng Duan (dingfengduan@uestc.edu.cn)

DOI:10.1016/j.jnlest.2024.100292

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Statistical results of θ0 including the value range and the proportion of pixels with values > 5°.

Table 1. Statistical results of θ0 including the value range and the proportion of pixels with values > 5°.

Table 2. Mean(μ) and standard deviation (σ) of backscatter components before and after the topographic compensation algorithm. The relative change rate (%) is Δ.

Table 2. Mean(μ) and standard deviation (σ) of backscatter components before and after the topographic compensation algorithm. The relative change rate (%) is Δ.

Table 3. Mean value (μ) and standard deviation (σ) of the θ0 values. The θ0 < 0° andθ0 ≥ 0° categories were created using DukeFr_225_mask for DukeFr_225 and DukeFr_0455_mask for DukeFr_045, respectively.

Table 3. Mean value (μ) and standard deviation (σ) of the θ0 values. The θ0 < 0° andθ0 ≥ 0° categories were created using DukeFr_225_mask for DukeFr_225 and DukeFr_0455_mask for DukeFr_045, respectively.

Table 4. Mean value (μ) and one standard deviation (σ) of the absolute θ0 values at a sample size at the near range and the other site at the far range.

Table 4. Mean value (μ) and one standard deviation (σ) of the absolute θ0 values at a sample size at the near range and the other site at the far range.

Table 1. Statistical results of θ₀ including the value range and the proportion of pixels with values > 5°.

Table 1. Statistical results of θ₀ including the value range and the proportion of pixels with values > 5°.

Table 3. Mean value (μ) and standard deviation (σ) of the θ₀ values. The θ₀ < 0° andθ₀ ≥ 0° categories were created using DukeFr_225_mask for DukeFr_225 and DukeFr_0455_mask for DukeFr_045, respectively.

Table 3. Mean value (μ) and standard deviation (σ) of the θ₀ values. The θ₀ < 0° andθ₀ ≥ 0° categories were created using DukeFr_225_mask for DukeFr_225 and DukeFr_0455_mask for DukeFr_045, respectively.

Table 4. Mean value (μ) and one standard deviation (σ) of the absolute θ₀ values at a sample size at the near range and the other site at the far range.

Table 4. Mean value (μ) and one standard deviation (σ) of the absolute θ₀ values at a sample size at the near range and the other site at the far range.