In this letter, we propose a novel ship detection method in high-resolution synthetic aperture radar (SAR) images. In the proposed method, the superpixel is utilized as the basic processing cell. We propose a superpixel-level Fisher vector to characterize the difference between the target and clutter superpixel due to the strong discriminating ability of the Fisher vector. Besides feature extraction, the threshold determination is another key step in the detection scheme. We determine the threshold in a semisupervised way, which combines the advantages of both supervised and unsupervised methods. Based on the threshold, the target superpixels are discriminated and we obtain the final detection result. Compared with existing methods, the proposed method shows robust performance under low signal-to-clutter ratio (SCR) conditions. Experiments performed on both synthetic data and a TerraSAR-X image demonstrate the effectiveness of the proposed method.

In this paper, we present a novel method for ship classification in synthetic aperture radar (SAR) images. The proposed method consists of feature extraction and classifier training. Inspired by SAR-HOG feature in automatic target recognition, we first design a novel feature named MSHOG by improving SAR-HOG, adapting it to ship classification, and employing manifold learning to achieve dimensionality reduction. Then, we train the classifier and dictionary jointly in task-driven dictionary learning (TDDL) framework. To further improve the performance of TDDL, we enforce structured incoherent constraints on it and develop an efficient algorithm for solving corresponding optimization problem. Extensive experiments performed on two datasets with TerraSAR-X images demonstrate that the proposed method, MSHOG feature and TDDL with structured incoherent constraints, outperforms other existing methods and achieves state-of-art performance.

The comprehensive utilization of images from various satellite sensors can significantly increase the performance of remote sensing applications and has therefore attracted extensive research attention. One of the essential challenges that researches encounter comes from multisource image registration. This paper proposes a novel region-based image registration method for multisource images. The proposed method exploits the region features of input images which provide more consistent and common information of the multisource data. The image region features are extracted based on image semantic segmentation using the deep convolution neural network (CNN) approach. The final registration result is a pixel-level output corresponding to the input images. The proposed registration scheme overcomes the limits of traditional feature extraction methods (e.g., point feature) adopted in previous registration schemes. Results indicate that the proposed method has good performance for the multisource remote sensing image registration and can serve as a building block for the fusion of multisource images.

Speckle reduction is a longstanding topic for polarimetric synthetic aperture radar (PolSAR) images. In this paper, we propose a novel end-to-end PolSAR image despeckling framework for the first time, which predicts the weight matrices of neighboring pixels instead of the target pixel itself nor the nor the noise, to achieve image despeckling. It hardly relies on any assumptions on the speckle noise distribution. Within this framework, residual in residual scaling network (RIRSN) is developed by combining the advantages of residual connections and residual scaling. To reduce network redundancy further, a dynamic version of RIRSN (DRIRSN) is also proposed by adjusting the network structure dynamically based on noise level and image content. Specifically, in DRIRSN, we introduce a lightweight network called picture2vector to estimate noise level, and a well-designed loss function to estimate image information level and measure image denoising quality simultaneously. The proposed picture2vector and loss function guide DRIRSN to focus on image areas with rich content and information, enhancing the adaptability of the network. DRIRSN inherits the properties of RIRSN for adaptively selecting and weighting the pixels of the neighborhood, and dynamically adjusts the network structure according to the estimated noise level and image content. We compare the proposed networks with reference methods on both simulated images and real images. Experimental results demonstrate that the proposed networks can effectively reduce speckle noise with low time consumption and, meanwhile, better preserve the details and the repetitive structures such as textures and edges, and the polarimetric scattering characteristics, compared with the other methods.

Target recognition for synthetic aperture radar (SAR) images has been a longstanding hotspot. However, using polarimetric information to enhance recognition performance is under-researched. In this paper, we develop a visualization approach to analyze and highlight the contributions of polarimetric elements, and propose a simple polarimetric correlation feature for target recognition. In the visualization method, a channel-wise convolutional structure is well developed, which serves as a proxy of the polarimetric elements. By using the gradients of the target class flowing into each developed convolutional channel with normalization, we obtain activation maps indicating the contribution of each polarimetric element. Then a comprehensive quantitative evaluation of polarimetric element contribution is also performed. We demonstrate again that polarimetric information maintains significant advantages over single-polarization intensity, and the correlations between cross- and co-polarization emerge as key components for target recognition. Exactly inspired by these insights, the straightforward feature is defined as the correlation between the target and typical scatterers, seamlessly fusing intensities and correlations. This polarimetric correlation feature aptly encapsulates the physical scattering of the target, shows a clear mapping relationship with the geometric structure, and captures the differences between categories, achieving intra-category compactness and inter-category separation. Experimental results on both synthetic and real datasets underscore the effectiveness and superiority of this proposed feature.

In this letter, we present a novel ship detection method for polarimetric synthetic aperture radar (PolSAR) images. Generalized polarization relative entropy (GPRE) is proposed to measure the differences between the target and clutter in scattering mechanism, randomness, and intensity. Since it is difficult to derive a theoretical closed-form of the GPRE, we employ the kernel density estimation to model the distribution of the GPRE in ocean regions. Then, a constant false alarm rate (CFAR) ship detection method is proposed based on the estimated distribution. Experiments performed on both synthetic and real scene images demonstrate the effectiveness of the proposed method.