Airborne synthetic aperture radar (SAR) imagery is susceptible to non-systematic motion errors, which will definitely defocus the target image and degrade the recognition accordingly. Representations learned from conventional convolutional neural networks (CNNs) do not emphasize much about the intrinsic features of the data distribution, which collapses the purposefulness of feature extraction. To this end, a deep Merged Auto-Focusing Network (MAFNet) is proposed for adaptively removing the defocusing effect from the input SAR data used for target recognition. Specifically, we propose an end-to-end architecture consisting of one focusing module and one recognition module. The focusing module contains a U-shaped sub-network based on the cross convolution to ensure the phase history coherence of the extracted features, and a deep unfolding method is devised to improve the mathematical generalization of focusing and sparse features so as to enhance the purposefulness of feature extraction. In particular, a compact surrogate function is designed for the non-convex feature quantization problem in the focusing module, which leads to closed-form solutions. The recognition module simply consists of a classifier. By building a multivariate loss function consisting of a cross-entropy loss function and a novel focusing loss function, MAFNet achieves accurate target recognition even when the input data is contaminated by non-systematic phase errors and additive noises. MSTAR data set is utilized to validate the effectiveness of MAFNet, and comparisons with conventional algorithms are performed to demonstrate the superiority of the proposed network.
Airborne aircrafts are dependent on the Global Navigation Satellite System (GNSS), which is susceptible to interference due to the satellite base-station and cooperative communication. Synthetic aperture radar altimeter (SARAL) provides the ability to measure the topographic terrain for matching with Digital Elevation Model (DEM) to achieve positioning without relying on GNSS. However, due to the near-vertical coupling in the delay-Doppler map (DDM), the similarity of DDMs of adjacent apertures is high, and the probability of successful matching is low. To this end, a novel neural network of terrain matching and aircraft positioning is proposed based on the airborne SARAL imagery. The model-driven terrain matching and aircraft positioning network (TMP-Net) is capable of realizing aircraft positioning by utilizing the real-time DDMs to match with the DEM-based DDM references, which are generated by a point-by-point coupling mechanism between the airborne routine and ground terrain DEM. Specifically, the training dataset is established by a numerical simulation method based on a semi-analytical model. Therefore, DEM-based DDM references can be generated by forward deduction when only regional DEM can be obtained. In addition to the model-based DDM generation, feature extraction, and similarity measurement, an aircraft positioning module is added. Three different positioning methods are designed to achieve the aircraft positioning, where three-point weighting exhibits the best performance in terms of positioning accuracy. Due to the fact that both the weighted triplet loss and softmax loss are employed in a cooperative manner, the matching accuracy can be improved and the positioning error can be reduced. Finally, both simulated and measured airborne datasets are used to validate the effectiveness of the proposed algorithm. Quantitative and qualitative evaluations show the superiority.
31 July 2019 in The Journal of Korean Institute of Information Technology
For conventional structure-guaranteed Synthetic Aperture Radar (SAR) imagery problems, some of them are modeled as ℓ 1 /ℓ 2 mixed norm regularization, and some are modeled as a group of least absolute shrinkage and selection operator (GLasso). However, due to the inflexible prior based on the Euclidean distance blocking, the target of interests and the background noise are easily attributed into the same block, which will consequently be retained or removed simultaneously. Therefore, it is difficult to distinguish the target with rich structures in a complicated environment. In this paper, to realize the elaborated structure feature enhancement of interested targets in the echoed SAR data, a novel Morphological AutoBlocking-Alternating Direction Method of Multipliers (MAB-ADMM) SAR imaging algorithm is proposed with a hybrid Frobenius and morphology ℓ F /ℓ M norm regularization. To realize the flexible blocking, an auto-blocking mechanism is constructed by Spatially-Variant Morphological Amoeba (SVMA) in the ℓ F /ℓ M mixed norm. In the auto-blocking mechanism, the geodesic distance of SVMA is utilized to accurately and flexibly fit the target structure and guarantee the elaborated structure. Moreover, in the proposed MAB-ADMM algorithm, the flexibly blocking structure feature and intra-block sparse feature are jointly optimized by the ℓ F /ℓ M and ℓ 1 norm. In the experiments, both simulated and raw SAR data collected by an airborne SAR are applied to qualitatively evaluate the proposed algorithm. Comparisons with conventional methods are carried out to reveal the superiority of the proposed algorithm. Moreover, the phase transition diagram (PTD) is shown to verify the proposed algorithm in terms of quantitative aspects.
29 May 2024 in IEEE Journal on Miniaturization for Air and Space Systems
Low-rank matrix decomposition is effective for sparse recovery. However, the conventions are limited in accuracy for high-resolution synthetic aperture radar (SAR) imagery due to the shrinkage effect in the cost function, which leads to biased estimates. To this end, an enhanced-low rank matrix decomposition (E-LRMD) SAR imaging algorithm is proposed, which employs a factor group-sparse regularization (FGSR) to approximate the intended cost function, so that the low-rank features can be represented. Since, the constructed regularization function is factorized, the singular value decomposition is avoided, and the computational burden can be reduced accordingly. Furthermore, $\ell _{1}$ -norm is incorporated to encode the sparse feature. To incorporate with the enhancement of multiple features, the alternating direction method of multipliers (ADMM) framework is utilized. Therefore, both the low-rank and sparse features can be accurately recovered and enhanced, cooperatively, where the error propagation between the enhancement of multiple features is minimized. In the experiments, the effectiveness and robustness of the algorithm are verified by the simulated data and practical UAV-SAR data, respectively. Also, a phase transition diagram (PTD) experiment is carried out to analyse the advantages of the proposed algorithm in terms of quantitative aspects compared with the conventional methods.