Cloud Shadow Compensation Method of Remote Sensing Images Based on Adaptive Gamma Transformation

Jinhui Yang; Xianjun Gao; Yuanwei Yang; Yuanyuan Feng

doi:10.3788/LOP202259.1028004

Journals >Laser & Optoelectronics Progress >Volume 59 >Issue 10 >Page 1028004 > Article

Laser & Optoelectronics Progress
Vol. 59, Issue 10, 1028004 (2022)

Cloud Shadow Compensation Method of Remote Sensing Images Based on Adaptive Gamma Transformation

Jinhui Yang¹, Xianjun Gao^1、*, Yuanwei Yang^1、2, and Yuanyuan Feng¹

Author Affiliations

¹School of Geosciences, Yangtze University, Wuhan 430100, Hubei , China

²Beijing Key Laboratory of Urban Spatial Information Engineering, Beijing 100045, China

show less

DOI: 10.3788/LOP202259.1028004 Cite this Article Set citation alerts

Jinhui Yang, Xianjun Gao, Yuanwei Yang, Yuanyuan Feng. Cloud Shadow Compensation Method of Remote Sensing Images Based on Adaptive Gamma Transformation[J]. Laser & Optoelectronics Progress, 2022, 59(10): 1028004 Copy Citation Text

show less

Compensation effects for different images under γ=1.12

Fig. 1. Compensation effects for different images under

γ

=1.12

Download full size

Fig. 2. Compensation effects of logarithmic transformation adaptive γ compensation method for different images

Download full size

Fig. 3. Schematic of the shaded pixel matrix window

Download full size

Fig. 4. Influence of the window size on the compensation results. (a) Impact on compensation quality; (b) impact on compensation time

Download full size

Fig. 5. Comparison of compensation effects under different window sizes. (a) n=2; (b) n=10; (c) n=17

Download full size

Fig. 6. Flow chart of adaptive Gamma transform compensation

Download full size

Fig. 7. Compensation result comparison of image 1. (a) Original image; (b) shadow detection; (c) classical Gamma compensation; (d) Gamma correction; (e) logarithmic transform adaptive Gamma compensation; (f) proposed method

Download full size

Fig. 8. Compensation result comparison of image 2. (a) Original image; (b) shadow detection; (c) classical Gamma compensation; (d) Gamma correction; (e) logarithmic transform adaptive Gamma compensation; (f) proposed method

Download full size

Fig. 9. Compensation result comparison of image 3. (a) Original image; (b) shadow detection; (c) classical Gamma compensation; (d) Gamma correction; (e) logarithmic transform adaptive Gamma compensation; (f) proposed method

Download full size

Image	Method	$Δ B^{2} + Δ T^{2}$	Compensation result		Shaded area		Non-shaded area
Image	Method	$Δ B^{2} + Δ T^{2}$	$B$	$T$	$B$	$T$	$B$	$T$
1	Proposed method	0.040	78.73	22.81	48.92	7.59	98.16	23.47
	Classical Gamma	0.298	61.48	14.14
	Gamma correction	0.365	52.07	14.56
	Log-Gamma	0.295	85.44	11.1
2	Proposed method	0.060	82.92	19.55	48.64	9.39	96.36	20.75
	Classical Gamma	0.283	62.15	14.81
	Gamma correction	0.347	53.08	15.18
	Log-Gamma	0.235	68.31	15.01
3	Proposed method	0.085	78.43	19.03	38.62	7.86	90.89	18.32
	Classical Gamma	0.437	60.2	11.19
	Gamma correction	0.547	47.5	11.32
	Log-Gamma	0.330	73.56	11.67

Table 1. Comparison of shadow compensation quality results using different methods

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information