Yong Wang

Using SAR Images to Detect Ships From Sea Clutter

An innovative constant false alarm rate (CFAR) algorithm was studied for ship detection using synthetic aperture radar (SAR) images of the sea. Two advances were achieved. An alpha-stable distribution rather than a traditional Weibull or -distribution was used to model the distribution of sea clutter. The distribution of sea clutter in a SAR image was typically heterogeneous, caused mainly by variable wind and current conditions. Image segmentation was carried out to improve the homogeneity of the distribution in each subimage or region. In comparison with ship detection using the CFAR algorithms based on the Weibull or K -distribution, our algorithm detected the most number of ships with the smallest number of false alarms.

Using Landsat 7 TM data acquired days after a flood event to delineate the maximum flood extent on a coastal floodplain

In response to Hurricane Floyd, the Tar River crested at a record height of 4.30u2009m above the flood stage at the river gauge station of Greenville (North Carolina, USA) on 21 September 1999. This resulted in a massive flooding in the area. To delineate the maximum flood extent, an area of 238.4u2009km2 along the Tar/Pamlico River, North Carolina, and within the overlapped area of Landsat 7 Thematic Mapper (TM) path 14/row 35 and path 15/row 35 scenes was studied. Three TM datasets of 28 July 1999 (path 15/row 35), 23 September 1999 (path 14/row 35) and 30 September 1999 (path 15/row 35) were analysed as pre-flood data, near peak data, and nine days after the peak data, respectively. The 23 and 30 September flood extent maps were derived by change detection and then verified by 85 nonflooded and flooded sites within the study area. The overall accuracies at the sites were between 82.5–99.3% on both inundation extent maps. Although the recorded river surface level fell 2.62u2009m from 23 to 30 September at the river gauge station of Greenville, comparison of the two flood extent maps on a pixel-by-pixel basis showed an agreement of 90.7% in terms of regular river channels and waterbodies, flooded areas and nonflooded areas. The 30 September map captured over 90% of the flood extent as identified on the 23 September map. These results suggest that it is possible to use remotely sensed data acquired days after a rivers crest to capture most of the maximum extent of a flood occurring on a coastal floodplain, and should somewhat reduce the requirement to have concurrently remotely sensed data in mapping a flood extent on a coastal floodplain.

A Novel Vehicle Detection Method With High Resolution Highway Aerial Image

A robust and efficient vehicle detection method from high resolution aerial image is still challenging. In this paper, a novel and robust method for automatic vehicle detection using aerial images over highway was presented. In the method, a GIS road vector map was used to constrain the vehicle detection system to the highway networks. After the morphological structure element was identified, we utilized the grayscale opening transformation and grayscale top-hat transformation to identify hypothesis vehicles in the light or white background, and used the grayscale closing transformation and grayscale bot-hat transformation to identify the hypothesis vehicles in the black or dark background. Then, targets with large size or covering a large area were sieved from the hypothesis vehicles using an area threshold that is much larger than a typical vehicle. Targets, whose width is narrower than the diameter of structure element utilized in the grayscale morphological transformation, were smoothed out from the hypothesis vehicles using binary morphological opening transformation. Finally, the hypothesis vehicles detected in both cases were overlaid. It should be noted that in the detection system, a vehicle could be detected twice by the two approaches. The two identical hypothesis vehicles should be amalgamated into a single one for accuracy assessment subsequently. We tested our system on seventeen highway scenes of aerial images with a spatial resolution of 0.15 × 0.15 m. The experimental results showed that the correctness, completeness, and quality rates of the proposed vehicle detection method were about 98%, 93%, and 92%, respectively. Thus, our proposed approach is robust and efficient to detect vehicles of highway using high resolution aerial images.

Compensation for the NsRCM and Phase Error After Polar Format Resampling for Airborne Spotlight SAR Raw Data of High Resolution

When the range migration caused by motion error exceeds the range cell resolution, the performance of a conventional phase autofocus approach degrades. In this paper, a new adaptive motion compensation (MoCo) algorithm with the removal of the migration that is nonsystematic has been developed for airborne spotlight synthetic aperture radar (SAR) imagery with high resolution. In the algorithm, the relationship between nonsystematic range cell migration (NsRCM) and phase error was first explicitly revealed after the polar format algorithm resampling. The NsRCM could be readily calculated by coarse but reliable phase error estimation. Subsequently, the NsRCM and the bulk of the azimuth phase error were corrected. After the removal of the NsRCM, degradation of the conventional phase autofocus resulting from sidelobe increase as well as mainlobe broadening was avoided. Finally, a fine MoCo procedure was performed to remove the residual azimuth phase error satisfactorily. Through the analysis of the airborne spotlight SAR raw data with high-resolution and wide-swath illumination, a well-focused imagery was obtained. Quantitative assessment of the image quality was satisfactory. The MoCo algorithm was validated.

A 2-D Space-Variant Chirp Scaling Algorithm Based on the RCM Equalization and Subband Synthesis to Process Geosynchronous SAR Data

A space-variant chirp scaling algorithm based on the range cell migration (RCM) equalization and azimuth subband synthesis has been studied to process simulated geosynchronous synthetic aperture radar (GEO-SAR) data. The acceptable order of terms in polynomials for the slant range models in the RCM correction and phase error compensation, division of subband, and suppression of grating lobes of the subbands was investigated. Qualitatively and quantitatively, the method was able to focus simulated GEO-SAR signals well. Finally, the constraint on the spatial extent of azimuth and range dimensions using the algorithm was assessed.

Analysis of the water volume, length, total area and inundated area of the Three Gorges Reservoir, China using the SRTM DEM data

Using the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) data covering the region of the Three Gorges Reservoir, Changjiang, China, we have computed the water volume, length, and total and inundated areas of the reservoir, with the assumption that the water surface within the reservoir is flat. When the reservoirs surface water level is 175 m above the mean sea level, the computed values may be comparable to the official data published by the Chinese government.

Seasonal change in the extent of inundation on floodplains detected by JERS-1 Synthetic Aperture Radar data

Two sets of JERS-1 (Japanese Earth Resource Satellite–1) Synthetic Aperture Radar (SAR) data, coupled with ancillary datasets, were analysed in an effort to find a single algorithm to study the extent of inundation and its variation on floodplains at a regional scale. The SAR data were acquired on 14 January, 1993 and 9 August, 1994. The study area was ca 14u2009212u2009km2, covering the lower portions of the Cape Fear, Lumber, Little Pee Dee and Waccamaw river basins within the states of North Carolina and South Carolina, USA. The analysis was based on the decision tree classification that classifies the study area into three aquatic categories, water, marsh and flooded forest, and two upland classes, field and non-flooded forest. From January 1993 to August 1994, the aquatic extent varied from 4872u2009km2 to 3496u2009km2, and upland 9340u2009km2 to 10u2009717u2009km2. The decrease of the water, marsh and flooded forest categories and the increase of the field and non-flooded forest classes were mainly caused by falls in water surface heights and discharges of the rivers and their tributaries from January 1993 to August 1994. The overall classification accuracy was near to 90%. The search for the single algorithm ended with promising results and also prompted additional research.

Sliding Spotlight and TOPS SAR Data Processing Without Subaperture

During the data acquisition of a sliding spotlight or terrain observation by progressive scan (TOPS) synthetic aperture radar (SAR), the steering of the antenna main beam increases the azimuth bandwidth but could result in the azimuth signal aliasing in the Doppler domain. To remove the aliasing, one has used a subaperture method. In this letter, we show a focusing scheme without the use of the subaperture for both sliding spotlight and TOPS SARs. In doing so, we eliminated the obvious increase in data volume or the subaperture division by choosing the pulse repetition frequency that is only 20% greater than the instantaneous bandwidth. The method was incorporated with an available imaging algorithm and then used to process simulated and collected data of the sliding spotlight and TOPS SARs. Well-focused results without aliasing were obtained.

Removal of azimuth ambiguities and detection of a ship: using polarimetric airborne C-band SAR images

Synthetic aperture radar (SAR) imagery from the sea can contain ships and their ambiguities. The ambiguities are visually identifiable due to their high intensities in the low radar backscatter background of sea environments and can be mistaken as ships, resulting in false alarms in ship detection. Analysing polarimetric characteristics of ships and ambiguities, we found that (a) backscattering from a ship consisted of a mixture of single-bounced, double-bounced and depolarized or diffused scattering types due to its complex physical structure; (b) that only a strong single- or double-bounce scatterer produced ambiguities in azimuth that look like relatively strong double- or single-bounce scatterers, respectively; and (c) that eigenvalues corresponding to the single- or double-bounce scattering mechanisms of the ambiguities were high but the eigenvalue corresponding to the depolarized scattering mechanisms of the ambiguities was low. With these findings, we proposed a ship detection method that applies the eigenvalue to differentiate the ship target and azimuth ambiguities. One set of C-band JPL AIRSAR (Jet Propulsion Laboratory Airborne Synthetic Aperture Radar) polarimetric data from the sea have been chosen to evaluate the method that can effectively delineate ships from their azimuth ambiguities.

Fusion of high-resolution DEMs derived from COSMO-SkyMed and TerraSAR-X InSAR datasets

Voids caused by shadow, layover, and decorrelation usually occur in digital elevation models (DEMs) of mountainous areas that are derived from interferometric synthetic aperture radar (InSAR) datasets. The presence of voids degrades the quality and usability of the DEMs. Thus, void removal is considered as an integral part of the DEM production using InSAR data. The fusion of multiple DEMs has been widely recognized as a promising way for the void removal. Because the vertical accuracy of multiple DEMs can be different, the selection of optimum weights becomes a key problem in the fusion and is studied in this article. As a showcase, two high-resolution InSAR DEMs near Mt. Qilian in northwest China are created and then merged. The two pairs of InSAR data were acquired by TerraSAR-X from an ascending orbit and COSMO-SkyMed from a descending orbit. A maximum likelihood fusion scheme with the weights optimally determined by the height of ambiguity and the variance of phase noise is adopted to syncretize the two DEMs in our study. The fused DEM has a fine spatial resolution of 10xa0m and depicts the landform of the study area well. The percentage of void cells in the fused DEM is only 0.13xa0%, while 6.9 and 5.7xa0% of the cells in the COSMO-SkyMed DEM and the TerraSAR-X DEM are originally voids. Using the ICESat/GLAS elevation data and the Chinese national DEM of scale 1:50,000 as references, we evaluate vertical accuracy levels of the fused DEM as well as the original InSAR DEMs. The results show that substantial improvements could be achieved by DEM fusion after atmospheric phase screen removal. The quality of fused DEM can even meet the high-resolution terrain information (HRTI) standard.