| 11. |
|
|
| 12. |
- Hu, Hongtao, et al.
(author)
-
Urban Land-use Mapping and Change Detection with RADARSAT Fine-Beam SAR Data Using Neural Network and Rule-based Classifiers
- 2008
-
In: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Beijing 2008. ; , s. 1549-1554
-
Conference paper (other academic/artistic)abstract
- This paper presents a new approach to extract urban landuse/land-cover information from high-resolution radar satellite data. Five-date RADARSAT fine-beam SAR images over the rural-urban fringe of the Greater Toronto Area were acquired during May to August in 2002. One scene of Landsat TM imagery was acquired in 1988 for change detection. The major landuse/land-cover classes were high-density built-up areas, low-density built-up areas, roads, forests, parks, golf courses, water and four types of agricultural crops (soybeans, corn, winter wheat/rye and pasture). The proposed approach to classify SAR images consisted of three steps: 1) image segmentation, 2) feature selection and object-based neural network classification, 3) rule set development to improve classification accuracy. Post-classification change detections were then performed using the final classification result of RADARSAT SAR images and the classification result of Landsat TM imagery. The results showed that the proposed approach achieved very good classification accuracy (overall: 87.9%; kappa: 0.867). The change detection procedure was able to identify the areas of significant changes, for example, new built-up areas, even though the overall accuracy of the change detection was not high.
|
|
| 13. |
- Hu, Hongtao, et al.
(author)
-
Urban-landuse/land-cover mapping with high-resolution SAR imagery by integrating support vector machines into object-based analysis
- 2008
-
In: <em>SPIE</em> <em>Europe Remote Sensing Conference</em>, 2008. - : SPIE.
-
Conference paper (peer-reviewed)abstract
- Thispaper investigates the capability of high-resolution SAR data for urbanlanduse/land-cover mapping by integrating support vector machines (SVMs) into object-basedanalysis. Five-date RADARSAT fine-beam C-HH SAR images with a pixelspacing of 6.25 meter were acquired over the rural-urban fringeof the Great Toronto Area (GTA) during May to Augustin 2002. First, the SAR images were segmented using multi-resolutionsegmentation algorithm and two segmentation levels were created. Next, arange of spectral, shape and texture features were selected andcalculated for all image objects on both levels. The objectson the lower level then inherited features of their superobjects. In this way, the objects on the lower levelreceived detailed descriptions about their neighbours and contexts. Finally, SVMclassifiers were used to classify the image objects on thelower level based on the selected features. For training theSVM, sample image objects on the lower level were used.One-against-one approach was chosen to apply SVM to multiclass classificationof SAR images in this research. The results show thatthe proposed method can achieve a high accuracy for theclassification of high-resolution SAR images over urban areas.
|
|
| 14. |
- Hu, Yunfeng, et al.
(author)
-
ALOS PALSAR differential interferometry for mapping co-seismic deformation of the 2008 Wenchuan earthquake, China
- 2009
-
In: Second International Conference on Earth Observation for Global Changes. - : SPIE - International Society for Optical Engineering. - 9780819477743 ; , s. 74711A-
-
Conference paper (peer-reviewed)abstract
- Differential interferometric synthetic aperture radar (DInSAR) technique has been widely accepted as a powerful tool to map surface deformation. To quantitatively evaluate the surface displacement caused by Wenchuan Earthquake on 12 May 2008 in Sichuan Province, China, a series of interferograms were generated from 25 ALOS/PALSAR image pairs, and the surface displacement was then mapped. According to the wrapped differential interferogram, the main rupture fault was plotted with an orientation of North-East 47° and a spanning length of approximately 230 km. The serious affected region with area of 5,000 km2 and the affected region with area of 250,000 km2 were also mapped. Along the radar look of sight (LOS), it is estimated that the ground surface displaced approximated a maximum of 57 cm and 119 cm away from and towards the satellite respectively, i.e. the vertical displacement was a maximum of 73 cm and 150 cm down lift and uplift respectively. The capability of DInSAR technique and ALOS PALSAR data for co-seismic deformation mapping has been demonstrated and proved to be useful in the surface deformation applications. In addition, some limitations were discussed including the topographic, atmospheric, and orbital errors.
|
|
| 15. |
|
|
| 16. |
- Hu, Yunfeng, et al.
(author)
-
Spatial-Temporal Pattern of GIMMS NDVI and Its Dynamics in Mongolian Plateau
- 2008
-
In: International Workshop on Earth Observation and Remote Sensing Applications, EORSA2008. - NEW YORK : IEEE. - 9781424423934 ; , s. 119-124
-
Conference paper (peer-reviewed)abstract
- The physical geography of Mongolian Plateau plays an important role in the East Asian climate ecology system. In this research, GIMMS NDVI, the third generation of NDVI dataset, was processed using the MVC method first, then the spatial-temporal patterns of GIMMS NDVI in Mongolian Plateau during 1982-2003 was investigated, and the transect from Tariat to Xilin Gol was also selected to analyze the NDVI dynamic processes in detail. The results demonstrated that: 1) The general spatial distribution pattern of NDVI showed a clear spatial differentiation. The high value pixels were mainly distributed in the east and north of Mongolian Plateau with forest and meadow steppe land cover, while the low value pixels were mainly distributed in the west and centre part of Mongolian Plateau with desert and Gobi land cover. However, the annual NDVI variability was relative small either in the high-covered regions (i.e. forest, forest steppe, and meadow steppe) or in low-covered regions (i.e. steppe desert, desert and Gobi), while the region with typical steppe normally had higher annual NDVI variability; 2) During 1982-2003, the dynamic evolution process of NDVI in Mongolian Plateau also showed an evident spatial differentiation. About 12.4% of total area featured a significant increase, 4.8% of total area featured an increase but without significance, and 9.3% of total area featured decrease without significance. The other part, about 73.5% of total area, had no obvious change. The NDVI increased significantly in the South-East, South and of Mongolian Plateau, while it decreased in the North-East and North of Mongolian Plateau. Further, the NDVI-increased regions were those typical steppe and farming-pastoral regions before, while the NDVI-decreased regions were those well-covered forest, forest steppe and meadow steppe regions before.
|
|
| 17. |
- Hu, Yunfeng, et al.
(author)
-
The Trajectory of Urbanization Process in the Yangtze River Delta During 1990 to 2005
- 2009
-
In: 2009 JOINT URBAN REMOTE SENSING EVENT. - NEW YORK : IEEE. - 9781424434602 ; , s. 407-414
-
Conference paper (peer-reviewed)abstract
- Rapid economic development, combined with tremendous population growth and migration from rural areas, has resulted in a dramatic urbanization in the Yangtze River Delta over the last three decades. In order to develop a more quantitative understanding of urbanization trajectory in the Yangtze Delta from the late 1980s to 2005, the time-series national land cover/land-use database (NLCD), interpreted from remote sensing imagery (including Landsat TM, ETM+ and CBERS), was examined from the aspects of spatial pattern and temporal process. The results showed that the urban land in the Yangtze River Delta has increased from 2.29x10(3) km(2) in the late 1980s to 4.19x10(3) km(2) in 2005. The percentages of total land area in the Yangtze Delta occupied by urban environments in the late 1980s, 1995, 2000 and 2005 were 2.09%, 3.03%, 3.28%, and 3.82% respectively. During the late 1980s - 2005, the total urban expansion in this region was 1.92x10(3) km(2), resulting in an annual urban expansion area (UXa) of about 118.72 km(2) per year, with an urban erpansion intensity (UXt)of 83.03%. There were significant spatial trends and differences in urbanization level and urban expansion. Most urban land and urban expansion have occurred along the Huning Railway (Shanghai-Nanjing) and the Hanzhou Bay. This study also revealed substantial temporal differences in the rate at which urban areas have expanded. There was a great deal of urban growth in the late 1980s - 1995, and also in 2000 - 2005. The 1995 - 2000 period saw a slow expansion speed due to government initiatives designed to hall the loss of agricultural land. In this study, both patterns and dynamics of urban distribution and urban expansion were examined on temporal and spatial scales, and integrated to produce a comprehensive evaluation of recent urbanization trajectory in the Yangtze River Delta.
|
|
| 18. |
- Kolehmainen, Karoliina, et al.
(author)
-
MULTITEMPORAL SPOT IMAGES FOR URBAN LAND-COVER CHANGE DETECTION OVER STOCKHOLM BETWEEN 1986 AND 2004
- 2008
-
In: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences.
-
Conference paper (other academic/artistic)abstract
- The overall objective of this research is to detect new urban areas over Stockholm Region between 1986 and 2004 using multitemporal remote sensing. Two SPOT images acquired on 13th of June 1986 and 29th of July 2004 were used for changed detection. Three change detection methods were tested for this purpose: image differencing, principal component analysis and change vector analysis using normalised difference vegetation index and brightness index. The results showed that image differencing using the red bands and second principle components performed better in detecting new urban features than change vector analysis (the overall accuracies: 89%, 87% and 83% respectively & kappa: 0.77, 0.74 and 0.67). Even though overall accuracies are all above 80%, the kappa coefficients were much lower indicating substantial amount of omission and commission errors presented in the change maps.
|
|
| 19. |
|
|
| 20. |
|
|
