Abstract:Synthetic aperture radar (SAR) images have been one of the important tools to support earth observations and topographic measurements. It means that SAR images are essentially rich in structures. However, the single spatial relationship is difficult to deal with the heterogeneous structures of the SAR images. In this paper, we propose an adaptive hierarchical multinomial latent model with hybrid kernel function for SAR image semantic segmentation. In the proposed approach, we design a hybrid kernel function combing Gaussian radial basis function (GRBF) and ridgelet kernel function to adaptively describe the spatial relationships between the central pixel and the surrounding pixels. Then, based on the hybrid kernel function, adaptive methods are proposed for semantic segmentation. Specifically, an SAR image is divided into different characteristics subspaces, homogeneous, structural, and aggregated subspaces, by SAR hierarchical semantic model. For the homogeneous subspace, GRBF is used to describe the isotropic spatial relationships. Then, multilayer multinomial latent model with GRBF is used for segmentation to improve the labeling consistency and reduce the wrong segmentation. For the structural subspace, the ridgelet kernel function is used to describe the anisotropic spatial relationships. Then, we adopt the singlelayer multinomial latent model with ridgelet kernel function for segmentation to preserve the details (such as edge, lines, and small objects). For aggregated subspace, bag-of-words model is used to extract the features of the aggregated portions, and then affinity propagation cluster is used for segmentation. Finally, the segmentation results of different subspaces are integrated together to obtain the final segmentation result. Comprehensive experiments on both synthetic and real SAR images demonstrate that the segmentation results by our proposed approach achieve the semantic consistency, labeling consistency, and detail preservation simultaneously.

Robust Image Similarity Metric Method for SAR Images

Robust Local Feature Descriptor for Multisource Remote Sensing Image Registration

Robust Optical and SAR Image Registration Based on Phase Congruency Scale Space

Rank-Based Local Self-Similarity Descriptor for Optical-to-SAR Image Matching

Automatic Registration of Optical and SAR Images Via Improved Phase Congruency Model.

A Multistatic ISAR Imaging Method Based on Similarity Prior With Overlaps Among Observation Angles

A Robust Registration Method for Multi-view SAR Images based on Best Buddy Similarity

Using Similarity Parameters for Supervised Polarimetric Sar Image Classification

SAR Image Change Detection via Spatial Metric Learning With an Improved Mahalanobis Distance

A Robust Contour Model for Matching Synthetic Aperture Radar (sar) Images with Maps

SAR image matching based on rotation-invariant description

A parameterized evaluation model of similarity on multi-target retrieval for remote sensing image

A Similarity Metric in Image Searching.

Robust Registration of Multimodal Remote Sensing Images Based on Structural Similarity

Similarity Ratio Based Adaptive Mahalanobis Distance Algorithm to Generate SAR Superpixels

Adaptive Hierarchical Multinomial Latent Model with Hybrid Kernel Function for SAR Image Semantic Segmentation

MOTIF: MULTI-ORIENTATION TENSOR INDEX FEATURE DESCRIPTOR FOR SAR-OPTICAL IMAGE REGISTRATION

A High Performance Edge Point Feature Match Method of SAR Images

Robust Optical and SAR Image Matching Using Attention-Enhanced Structural Features

SAR image matching algorithm based on multi-class features

Optical and SAR Image Registration Based on Multi-Scale Orientated Map of Phase Congruency