Characterising spatial information as reinforcement of spectral signatures can largely assist the performance in hyperspectral image (HSI) classification. Markov random fields (MRFs) are probabilistic image texture models, and capable of encoding contextual dependencies through charactering local conditional probabilities. As a representative standardised measure of dispersion of image probability distributions, coefficient of variation (CoV) can be a useful tool for characterising spatial heterogeneity. Their parameter derivation processes also share strong compatibility with convolutional neural networks that specifies spatial correlations in local neighbourhoods. In this work, we propose an MRF and CoV based spectral-spatial convolutional network (MRF-CoV-CNN) for HSI classification. MRF models and CoVs are characterised as measures of spatial distributions and further combined with spectral information. Then the proposed MRF-CoV-CNN takes the fused features as input and produces reliable classification results. Comprehensive experiments have been conducted on the Pavia university dataset and the Salinas dataset to evaluate the proposed method both visually and quantitatively.

With advancements in synthetic aperture radar (SAR) imaging geodesy and geodetic stereo SAR, the geometric accuracy of point scatterers (PSs), such as corner reflectors, has reached decimeter or even centimeter levels, making them viable as ground control points (GCPs). Compared with global navigation satellite system (GNSS)-based methods, SAR-based GCP extraction approaches are more efficient and cost-effective. This article investigates the extraction, positioning, and quality control of lamp-pole GCPs from spaceborne stereo SAR imagery. We first reformulate the positioning problem as a Gauss-Markov model (GMM) and derive the relationship between precision and the difference of incidence angles. We then analyze the concept of lamp-pole GCPs, which, due to their visibility in SAR images from multiple orbital perspectives, exhibit high precision. Subsequently, we propose an automatic method for extracting GCPs from stereo SAR images, along with a robust positioning technique that adjusts observations into distance units. Finally, we analyze three quality indicators of the extracted GCPs: precision, internal reliability, and external reliability. We validate the effectiveness of the proposed methods through simulation and real-life Gaofen-3 data.

Several researchers have shown that the Capon algorithm can be applied to reprocess SAR images, resulting in super-resolution reconstructed scenes with lower sidelobe levels. Thus by employing the Capon-based reprocessed images in Persistent Scatterer Interferometry (PSI), the persistent scatterer (PS) density can be increased. In this letter, we exploit the Capon-based PS densification method for Sentinel-1 (S-1)Terrain Observation by Progressive Scans(TOPS) data. We propose a revised robust approach of the Capon algorithm, which applies the automatic diagonal loading (DL) method when the condition number of the covariance matrix is big enough. The proposed approach is robust and can avoid spurious persistent scatterer candidate (PSC) points introduced by DL approaches. We also consider and analyze the spectral property caused by the scanning mode of TOPS in the reprocessing. We applied the revised-robust-Capon-based reprocessing algorithm to a stack of real-life S-1 data and selected PSCs from them. The final result shows that the number of PSs increases by approximately 30% with respect to the original stack.

The Capon algorithm can be applied to reprocess SAR images, resulting in super-resolution super-resolution reconstructed scenes with lower sidelobe levels. Base on this idea, we have proposed a processing chain of Super-Resolution Persistent Scatterer Interferometry (SR-PSI), to increase PS density. In this paper, we review the main aspect of SR-PSI. We also propose a revised robust Capon algorithm. The result of real-life Sentinel-1 (S-1) data is shown.