In polarimetric synthetic aperture radar (POLSAR), it is of great significance for civil and military applications to find novel model-based decomposition methods suitable for ship detection in different detection backgrounds. Based on the physical interpretation of polarimetric decomposition theory and the Lasso rule for sparse features, we propose a four-component decomposition model, which is composed of surface scattering (Odd), double-bounce scattering (Dbl), volume scattering (Vol), and ±45° oriented dipole (Od). In principle, the Od component can describe the compounded scattering structure of a ship consisting of odd-bounce and even-bounce reflectors. Moreover, the pocket perceptron learning algorithm (PPLA) and support vector machine (SVM) are utilized to solve the linear inseparable problems in this study. Using large amounts of RADARSAT-2 (RS-2) fully polarized SAR data and AIRSAR data, our experimental results show that the Od component can make a great contribution to ship detection. Compared with other conventional decomposition methods used in the experiments, the proposed four-component decomposition method has better performance and is more effective and feasible to detect ships.

Azimuth ambiguity is a common issue in polarimetric synthetic aperture radar (PolSAR) imagery, particularly on calm and windless maritime surfaces, which causes numerous false alarms in ship detection. Numerous methods have been applied in Single Look Complex (SLC) PolSAR imagery to suppress ambiguities. Nevertheless, identifying and removing azimuth ambiguities in Multilook Complex (MLC) PolSAR imagery remains an open problem. This paper proposes an azimuth ambiguity identification method for ship detection in multilook PolSAR imagery. The process is divided into two steps: potential target detection and ambiguity identification. Firstly, the four-component scattering model (Y4O) proposed by Yamaguchi is utilized to decompose the multilook PolSAR image into four dominant scattering categories. Then, the constant false alarm rate (CFAR) detection is conducted based on the total scattering power to detect all potential targets. Azimuth ambiguities are identified according to the correlation coefficient between the measured and standard scattering power vectors. Eventually, the detection map is formed by removing azimuth ambiguities from the CFAR detection result. The proposed method is validated on RadarSAT-2 and AIRSAR images.

Constant false alarm rate (CFAR) detector is a common method for ship detection in polarimetric synthetic aperture radar (PolSAR) image. CFAR detectors greatly depend on the clutter modeling which can be easily affected by the contamination caused by both lower- and higher-intensity outliers such as spilled oil, intensive targets and so on. Traditional CFAR detectors perform detection in a pixel-by-pixel manner which ignores the spatial information. Both the bias in clutter modeling and the absence of spatial information can degrade the ship target detection performance. In this study, a superpixel-level polarimetric bilateral truncated statistics (BTS) CFAR detector is proposed to promote the ship target detection performance in complex ocean scenarios. As the preprocessing of the PolSAR image, the superpixel segmentation is conducted based on the multilook polarimetric whitening filter (MPWF) result to select candidate ship target superpixels for bilateral truncation and background clutter modeling. The elliptical truncation is expanded to complex situation and the relationship between the second moments before and after truncation is derived. The maximum likelihood estimation (MLE) estimator of the equivalent number of looks (ENL) based on the bilateral truncation distribution is derived and compared with other parameter estimators. The influence of the truncation depth on estimator performance is analyzed, according to which the adaptive bilateral truncation method is determined. The Gaussian mixture model (GMM) and the Parzen window kernel method are compared with the model-based method and utilized for data fitting. The proposed method performs bilateral truncation based on the superpixel segmentation result to provide the pure clutter samples for the accurate parameter estimation and clutter distribution modeling, reducing time consumption and false alarm. The method is validated efficient on both simulated and measured data from RADARSAT-2.

This paper describes an optimised scattering power decomposition method suitable for ship detection in polarimetric synthetic aperture radar imagery. Based on the principal component analysis technique for dimension reduction and the scattering power decompositions with strong physical interpretation, a novel optimised scattering power decomposition model is proposed which comprises surface scattering, double‐bounce scattering, ±45° oriented dipole, and volume scattering components. Wherein, the oriented dipole component can be used to characterise the compounded scattering of the target which is composed of even‐bounce and odd‐bounce reflectors. Furthermore, pocket algorithm and support vector machine are adopted to solve linear non‐separable problems under complex experimental conditions in this study. Extensive experiments carried out on RADARSAT‐2 data and NASA/JPL AIRSAR data show that the oriented dipole component significantly contributes to the optimised four‐component decomposition method and can play a vital role in ship detection. Compared to other scattering power decompositions, the optimised decomposition method not only achieves better performance but is more in accordance with the real scattering mechanisms, which is more applicable to ship detection.