Centerline detection and line width estimation are important for many computer vision applications, e.g., road network extraction from high resolution remotely sensed imagery. Radon transform-based linear feature detection has many advantages over other approaches: for example, its robustness in noisy images. However, it usually fails to detect the centerline of a thick line due to the peak selection problem. In this paper, several key issues that affect the centerline detection using the radon transform are investigated. A mean filter is proposed to locate the true peak in the radon image and a profile analysis technique is used to further refine the line parameters. The thetas-boundary problem of the radon transform is also discussed and the erroneous line parameters are corrected. Intensive experiments have shown that the proposed methodology is effective in finding the centerline and estimating the line width of thick lines

Accurate Centerline Detection and Line Width Estimation of Thick Lines Using the Radon Transform

This paper evaluates the potentialities of polarimetric ship scattering for basing classification methods that provide reasonable performance within cluttered scenes. Both simulated and airborne polarimetric synthetic aperture radar (SAR) images have been used to validate the conclusions of a previous phenomenological study. Numerical simulations have been carried out with GRECOSAR, a polarimetric interferometric SAR simulation tool that is able to process highly complex targets with a fast and accurate radar-cross-section prediction module. A representative set of scenarios has been defined, which includes various realistic ship models, sea states, and imaging geometries. In all of them, a two-scale sea surface model precisely accounting for sea-ship interaction and sea clutter has been added. The analysis of different images has shown that, with an adequate spatial resolution, ships may be characterized by a particular spatial arrangement and polarization state distribution of dominant scattering centers. This feature has allowed one to propose a new classification algorithm, which shows a promising behavior after various preliminary tests. In this paper, the performance of this technique is further evaluated with realistic clutter. The results show that robust classification is possible even in highly cluttered scenes if quad-pol imagery is available. On the contrary, in low clutter conditions, the usage of less restrictive solutions, like circular dual-pol schemes, is feasible and may still get an acceptable performance.

Exploitation of Ship Scattering in Polarimetric SAR for an Improved Classification Under High Clutter Conditions

Synthetic aperture radar (SAR) images of ship wakes frequently exhibit long streaks. Some of the streaks are attributed to the Kelvin V-wave wake, and others are attributed to the turbulent and bright narrow V-wake. Ocean waves reduce the detectability of Kelvin wakes in SAR images. The SAR imaging of Kelvin wakes of two ships with different dimensions moving in rough sea is studied theoretically. The Kelvin wakes are simulated using the Michell theory. The random sea surface is simulated using the Pierson-Moskowitz type wave spectrum. The elevations of the points of the water surface and their velocities and accelerations are considered as a superposition of a ship's coherent waves and wind waves. A two-scale composite model and the velocity-bunching integral are used to simulate the intensity of SAR images of a disturbed sea surface with an embedded Kelvin wake. The fast discrete Radon transform is applied to detect the boundaries of the simulated Kelvin wakes. The probabilities of missing detections (PMD) and false alarms (PFA) are estimated numerically. The influence of ships, wind, and SAR parameters on PMD and PFA is discussed. It is illustrated that PMD/PFA strongly depend on the sea state.

On Detectability of a Ship's Kelvin Wake in Simulated SAR Images of Rough Sea Surface

This paper presents a novel method for vessel classification based on single-pass polarimetric synthetic aperture radar (SAR) interferometry. It has been developed according to recent ship scattering studies that show that the polarimetric response of many types of vessels can be described by trihedral- and dihedral-like mechanisms. The adopted methodology is quite simple. The input interferometric data are decomposed in terms of the Pauli basis, and hence, one height image is derived for each simple mechanism. Then, the local maxima of these images are isolated, and a 3-D map of scatters is generated. The correlation of this map with the scattering distribution expected for a set of reference ships provides the final classification decision. The performance of the proposed method has been tested with the orbital SAR simulator developed at Universitat Politecnica de Catalunya. Different vessel models have been processed with a sensor configuration similar to the incoming TanDEM-X system. The analysis of diverse vessel bearings, vessel speeds, and sea states shows that the map of scatters matches reasonably the geometry of ships allowing a correct identification even for adverse environmental conditions.

/pdf/single-pass-polarimetric-sar-interferometry-for-vessel-26bc1gw3lx.pdf

Single-Pass Polarimetric SAR Interferometry for Vessel Classification

The problem in obtaining stable motion estimation of maritime targets is that sea clutter makes wake structure detection and reconnaissance difficult. This letter presents a complete procedure for the automatic estimation of maritime target motion parameters by evaluating the generated Kelvin waves detected in synthetic aperture radar (SAR) images. The algorithm consists in evaluating a dual-stage low-rank plus sparse decomposition (LRSD) assisted by Radon transform (RT) for clutter reduction, sparse object detection, precise wake inclination estimation, and Kelvin wave spectral analysis. The algorithm is based on the robust principal component analysis (RPCA) implemented by convex programming. The LRSD algorithm permits the extrapolation of sparse objects of interest consisting of the maritime targets and the Kelvin pattern from the unchanging low-rank background. This dual-stage RPCA and RT applied to SAR surveillance permits fast detection and enhanced motion parameter estimation of maritime targets.

Low-Rank Plus Sparse Decomposition and Localized Radon Transform for Ship-Wake Detection in Synthetic Aperture Radar Images

Synthetic aperture radar (SAR) images of ships and their wakes frequently exhibit long dark and bright streaks. Some of them can be attributed to the Kelvin wavewake and others to the turbulent and bright narrow V-wake. The wakes contain information about the ship. The present work deals with estimates of the ship beam and its velocity by processing SAR images of the Kelvin and turbulent wakes. It is assumed that the ship moves along a straight path with constant speed. For the detection of the linear features of the ship wake, the fast discrete Radon transform is employed. Once the turbulent wake is detected, the ship beam is estimated by a novel method that exploits the expansion of the turbulent wake aft a ship. A semiempirical relation between the ship beam and the width of its turbulent wake is derived and analyzed. An algorithm for estimating the width of the turbulent wake in SAR images and the ship beam is developed. The spectrum of ship-generated waves along the Kelvin cusp-lines is discussed. Processing of the lines, pertaining to the Kelvin wake bounds, and analysis of the spectral peaks enables to estimate the ship speed. Numerical examples of processing of airborne SAR images are provided.

M. Marom

Papers

The speed and beam of a ship from its wake's SAR images