LAND COVER CHANGE DETECTION USING RADARSAT2 POLARIMETRIC SAR IMAGES
Zhixin Qi and Anthony Gar-On Yeh
The University of Hon...
Outline
1
Introduction
2
Study area and data
3
Methodology
4
Results and discussion
5
Conclusions
Background
There are many illegal land developments in some of China’s
rapidly developing regions, such as the Pearl Rive...
RADAR vs. Optical remote sensing
Radar remote sensing, which is not affected by
cloud conditions, is promising for monito...
Multi-polarization vs. single-polarization
Polarimetric SAR (PolSAR)
HH
Single-polarization SAR
VV
HH
HV
VH
polarization
The polarization information contained in the waves backscattered from
a given medium is highly related to:
•...
Study Objective
Research questions
• Unsupervised methods
– They cannot determine types of changes.
• Post-classification ...
Study Area
Land Cover Classes in the Study Area
Study Data
•
•
•
RADARSAT-2 Fine Quad-Pol images (Single Look Complex).
Full polarization: HH, HV, VH and VV.
Incidence ...
Field Work
Field Work
 Class
Sub-class 
Plots
Pixels
Change
Barren land to crop/natural vegetation
68
28,995
 
 
 
 
 
 
 
 
...
Methodology
Land cover change detection using two PolSAR images acquired
over the same area at different times
Polarimetric Decomposition
•
•
Polarimetric decomposition can be used to support the classification of
PolSAR data. It is...
Polarimetric Decomposition
Decomposition Methods
•
•
•
•
•
•
•
•
•
•
•
•
•
Pauli (Cloude & Pottier, 1996)
Barnes (Barnes,...
Image Segmentation
Determining the optimal scale for the segmentation of the Pauli RGB
composition image of RADARSAT-2 Po...
Separate Segmentation
Hierarchical Segmentation
Hierarchical segmentation for
delineating image objects from
two successive RADARSAT-2
PolSAR im...
Change Vector Analysis (CVA)
Two images, image (t1) and image (t2), are acquired over the same area at different
times t1 ...
Change Vector Analysis (CVA)
March 21, 2009
September 29,2009
Change magnitude
Changed areas
PolSAR image classification
Methodology of land cover classification using RADARSAT-2 PolSAR images
Proposed Method Vs. WSC
Classification Results
Land Cover Change Results
e
Land cover change detection results
(a)Proposed method (CVA, PCC, and
OOIA)
(b)WSC-based PCC
...
Accuracy Assessment
Change type
Accuracy
statistics
All the types
DA (%)
FAR (%)
OER (%)
DA (%)
FAR (%)
OER (%)
DA (%)
...
Conclusions
•
The proposed method performs much better than WSC-based PCC in term
of land cover change detection using RA...
Conclusions
•
The proposed method performs much better than WSC-based PCC in term
of land cover change detection using RA...
POLSAR CHANGE DETECTION
of 28

POLSAR CHANGE DETECTION

polsar change detection by Qi 2013/11/14
Published on: Mar 4, 2016
Published in: Technology      Business      
Source: www.slideshare.net


Transcripts - POLSAR CHANGE DETECTION

  • 1. LAND COVER CHANGE DETECTION USING RADARSAT2 POLARIMETRIC SAR IMAGES Zhixin Qi and Anthony Gar-On Yeh The University of Hong Kong, Hong Kong, China
  • 2. Outline 1 Introduction 2 Study area and data 3 Methodology 4 Results and discussion 5 Conclusions
  • 3. Background There are many illegal land developments in some of China’s rapidly developing regions, such as the Pearl River Delta (PRD).
  • 4. RADAR vs. Optical remote sensing Radar remote sensing, which is not affected by cloud conditions, is promising for monitoring short-term land cover changes.
  • 5. Multi-polarization vs. single-polarization Polarimetric SAR (PolSAR) HH Single-polarization SAR VV HH HV VH
  • 6. polarization The polarization information contained in the waves backscattered from a given medium is highly related to: • its geometrical structure reflectivity, shape and orientation • its geophysical properties such as humidity, roughness, … V or H H H H Completely Polarised Scattering Partially Polarised Scattering
  • 7. Study Objective Research questions • Unsupervised methods – They cannot determine types of changes. • Post-classification comparison (PCC) – Poor accuracy of PolSAR image classification caused by the limited spectral information and speckle noise • Pixel-based methods – They may cause false alarms due to the speckle effect. Study objective • This study aims to develop a new method that integrates change vector analysis (CVA) and post-classification comparison (PCC) with objectoriented image analysis (OOIA) to detect land cover changes from RADARSAT-2 polarimetric SAR (PolSAR) images.
  • 8. Study Area
  • 9. Land Cover Classes in the Study Area
  • 10. Study Data • • • RADARSAT-2 Fine Quad-Pol images (Single Look Complex). Full polarization: HH, HV, VH and VV. Incidence angle: 31.50°.
  • 11. Field Work
  • 12. Field Work  Class Sub-class  Plots Pixels Change Barren land to crop/natural vegetation 68 28,995                 Water to crop/natural vegetation 47 19,089 Crop/natural vegetation to water 75 17,738 Barren land to built-up areas 51 14,257 Barren land to water 41 9,417 Water to lawns 7 4,089 Crop/natural vegetation to barren land 10 3,392 Water to barren land 9 3,380 Total 308 100,357 No change Banana 107 41,656               Barren land 84 23,743 Forests 118 36,437 Lawns 98 34,159 Crop/natural vegetation 202 93,196 Built-up areas 224 65,939 Water 130 66,130 Total 963 361,260 Number of change and no-change samples selected for the verification of land cover change detection results
  • 13. Methodology Land cover change detection using two PolSAR images acquired over the same area at different times
  • 14. Polarimetric Decomposition • • Polarimetric decomposition can be used to support the classification of PolSAR data. It is aimed at extracting polarimetric parameters related to the physical scattering mechanisms of the observed objects. In this study, PolSARPro_4.03 software package was used to implement polarimetric decomposition. Polarimetric Decomposition
  • 15. Polarimetric Decomposition Decomposition Methods • • • • • • • • • • • • • Pauli (Cloude & Pottier, 1996) Barnes (Barnes, 1988) Huynen (Huynen, 1970) Cloude (Cloude, 1985) Holm (Holm & Barnes, 1988) H/A/Alpha (Cloude & Pottier, 1997) Freeman 2 Components (Freeman, 2007) Freeman 3 Components (Freeman & Durden, 1998) Van Zyl (Van Zyl, 1993) Neumann (Neumann et al., 2009) Krogager (Krogager, 1990) Yamaguchi (Yamaguchi et al., 2005) Touzi (Touzi, 2007) methods
  • 16. Image Segmentation Determining the optimal scale for the segmentation of the Pauli RGB composition image of RADARSAT-2 PolSAR data.
  • 17. Separate Segmentation
  • 18. Hierarchical Segmentation Hierarchical segmentation for delineating image objects from two successive RADARSAT-2 PolSAR images A feature is an attribute that represents certain information concerning objects of interest, such as color, shape, and texture.
  • 19. Change Vector Analysis (CVA) Two images, image (t1) and image (t2), are acquired over the same area at different times t1 and t2. If k features are extracted from an image object, the feature vectors of the image object in the two images are given by X = (x1, x2, …, xk)T and Y = (y1, y2, …, yk)T respectively, the feature change vectors are defined as  x1 − y1     x − y2  ∆G = X − Y =  2     x − y  k   k (1) where ∆G includes all the change information between the two images for a given image object, and the change magnitude is computed with ∆G (2) ∆G = ( x1 − y1 ) 2 + ( x 2 − y 2 ) 2 +  + ( x k − y k ) 2 The higher the is, the more likely that changes take place. Unsupervised ∆G classifiers or threshold methods are commonly applied on the change magnitude to identify changes.
  • 20. Change Vector Analysis (CVA) March 21, 2009 September 29,2009 Change magnitude Changed areas
  • 21. PolSAR image classification Methodology of land cover classification using RADARSAT-2 PolSAR images
  • 22. Proposed Method Vs. WSC
  • 23. Classification Results
  • 24. Land Cover Change Results e Land cover change detection results (a)Proposed method (CVA, PCC, and OOIA) (b)WSC-based PCC (c)PCC and OOIA (without CVA) (d)CVA and PCC (without OOIA) (e)CVA and OOIA (without PCC)
  • 25. Accuracy Assessment Change type Accuracy statistics All the types DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) DA (%) FAR (%) OER (%) BL-CN BL-BU BL-W CN-BL CN-W W-BL W-L W-CN Proposed method (CVA, PCC, and OOIA) 86.71 3.35 5.51 46.50 0.11 3.46 46.59 0.09 1.73 47.97 0.09 1.15 73.88 0.85 1.04 52.06 0.05 1.89 68.76 0.86 1.09 89.90 0.51 0.59 63.87 0.11 1.60 WSC-based PCC 94.94 38.10 30.92 33.23 0.50 4.66 32.98 0.33 2.39 41.89 0.79 1.96 34.08 0.85 1.58 37.38 0.27 2.67 51.30 1.33 1.68 71.14 1.05 1.29 39.22 0.20 2.70 CVA and OOIA (without PCC) 90.51 7.57 7.99 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA PCC and OOIA (without CVA) 93.15 17.87 15.48 58.08 0.18 2.80 48.93 0.09 1.66 48.32 0.20 1.25 76.62 0.94 1.11 52.06 0.05 1.89 68.91 1.26 1.48 89.90 0.51 0.59 63.87 0.11 1.60 CVA and PCC (without OOIA) 88.13 10.25 10.60 31.23 0.30 4.59 45.46 0.87 2.53 44.88 0.40 1.52 38.41 0.60 1.05 40.27 0.31 2.59 68.05 1.62 1.85 47.79 0.60 1.05 38.40 0.16 2.70
  • 26. Conclusions • The proposed method performs much better than WSC-based PCC in term of land cover change detection using RADARSAT-2 PolSAR images. • The use of CVA before PCC can significantly reduce false alarms caused by the error of the classification of PolSAR images. • Using PCC after CVA can reduce false alarms caused by environmental changes, such as seasonal vegetation growth and moisture variation. PCC that is based on the proposed classification approach, which integrates polarimetric decomposition, decision tree algorithms, and SVMs, achieves much higher accuracy than WSC-based PCC. • OOIA reduces false alarms caused by speckles in PolSAR images and improves the accuracy of change type determination. • Further investigation will be conducted to examine the effect of seasonal vegetation growth on the monitoring of human-induced land cover changes as well as how to distinguish between human-induced land cover changes and changes caused by seasonal vegetation growth.
  • 27. Conclusions • The proposed method performs much better than WSC-based PCC in term of land cover change detection using RADARSAT-2 PolSAR images. • The use of CVA before PCC can significantly reduce false alarms caused by the error of the classification of PolSAR images. • Using PCC after CVA can reduce false alarms caused by environmental changes, such as seasonal vegetation growth and moisture variation. PCC that is based on the proposed classification approach, which integrates polarimetric decomposition, decision tree algorithms, and SVMs, achieves much higher accuracy than WSC-based PCC. • OOIA reduces false alarms caused by speckles in PolSAR images and improves the accuracy of change type determination. • Further investigation will be conducted to examine the effect of seasonal vegetation growth on the monitoring of human-induced land cover changes as well as how to distinguish between human-induced land cover changes and changes caused by seasonal vegetation growth.

Related Documents