Researching | PolSAR image classification based on object-oriented technology

Journal of Infrared and Millimeter Waves, Volume. 39, Issue 4, 505(2020)

PolSAR image classification based on object-oriented technology

Yan XIAO¹ and Bin WANG²

Author Affiliations

¹College of Exploration and Surveying Engineering, Changchun Institute of Technology, Changchun3002, China

²Changchun Institute of Surveying and Mapping, Changchun13001, China

show less

Abstract Get PDF(in Chinese)

Figures&Tables (8)

References (12)

Paper Information

Figures & Tables(8)

Fig. 1. Location map of study area

Download full size

View in Article

Fig. 2. RADARSAT-2 image (Pauli RGB composition).

Download full size

View in Article

Fig. 3. Distribution map of samples for each class

Download full size

View in Article

Fig. 4. Comparison of accuracies for classifications

Download full size

View in Article

Fig. 5. Land-use maps based on different classification methods

Download full size

View in Article

Table 1. 极化分解方法及相应的极化参数

View in Article

Table 1. 极化分解方法及相应的极化参数

分解方法	极化参数
Pauli	Pauli_ a	Pauli_ b	Pauli_ c
Krogager	Krogager_ KS	Krogager_KH	Krogager_KD
Huynen	Huynen_T₁₁	Huynen_T₂₂	Huynen_T₃₃
Barnes1	Barnes1_T₁₁	Barnes1_T₂₂	Barnes1_T₃₃
Barnes2	Barnes2_T₁₁	Barnes2_T₂₂	Barnes2_T₃₃
Cloude	Cloude_T₁₁	Cloude_T₂₂	Cloude_T₃₃
H/A/α	H/A/α_T₁₁	H/A/α_T₂₂	H/A/α_T₃₃
	Entropy(H)	SERD	RVI
	DERD	PolarizationAsymmetry(PA)	ShannonEntropy (SE)
	PedestalHeight (PH)	PolarizationFraction (PF)	Anisotropy(A)
Freeman2	Freeman2_Vol	Freeman2_Ground
Freeman3	Freeman_Vol	Freeman_Odd	Freeman_Dbl
Yamaguchi3	Yamaguchi3_Vol	Yamaguchi3_Odd	Yamaguchi3_Dbl
Yamaguchi4	Yamaguchi4_Vol	Yamaguchi4_Odd	Yamaguchi4_Dbl
	Yamaguchi4_Hlx
Neumann	Neumann_delta_mod	Neumann_delta_pha
Touzi	TSVM_alpha_s	TSVM_alpha_s1	TSVM_alpha_s2
	TSVM_alpha_s3	TSVM_tau_m	TSVM_tau_m1
	TSVM_tau_m2	TSVM_tau_m3
Holm1	Holm1_T₁₁	Holm1_T₂₂	Holm1_T₃₃
Holm2	Holm2_T₁₁	Holm2_T₂₂	Holm2_T₃₃
Van Zyl	VanZyl3_Vol	VanZyl3_Odd	VanZyl3_Dbl

Table 2. 分类结果的混淆矩阵

View in Article

Table 2. 分类结果的混淆矩阵

	水体	道路	居民地	耕地	林地	草地	总和	UA/(%)
水体	693295	5500	50127	0	721	2931	752574	92.12
道路	610	938013	80331	50388	60134	51158	1180634	79.45
居民地	1142	113522	2874381	28776	51145	5961	3074927	93.48
耕地	0	113578	169628	3908101	827790	1241	5020338	77.85
林地	491	125447	73537	180194	3124430	3524	3507353	89.08
草地	0	41578	6670	2464	2797	143533	197042	72.84
总和	695538	1337638	3254674	4169923	4067017	208078
PA/(%)	99.68	70.12	88.32	93.72	76.82	68.98
OA/(%)	85.06
Kappa	0.8006

Table 3. 不同分类方法的分类精度对比
View table
View in Article
Table 3. 不同分类方法的分类精度对比
总体精度/(%) Kappa 耗时
本文方法 85.06 0.8006 17 min
Pauli+ReliefF-PSO_SVM+集成 62.52 0.5002 14 min
极化分解+FSO+集成 65.37 0.5376 14h13min
极化分解+ReliefF-PSO_SVM 83.10 0.7747 9 min

Tools

Get Citation

Copy Citation Text

Yan XIAO, Bin WANG. PolSAR image classification based on object-oriented technology[J]. Journal of Infrared and Millimeter Waves, 2020, 39(4): 505

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Remote Sensing Technology and Application

Received: Nov. 13, 2019

Accepted: --

Published Online: Sep. 17, 2020

The Author Email:

DOI:10.11972/j.issn.1001-9014.2020.04.015

Topics

laser devices and laser physics

Lasers and Laser Optics

laser manufacturing

Instrumentation, Measurement and Metrology