Airborne SAR tomography (TomoSAR) 3D image reconstruction can be realized with combination of 2D imaging algorithms and compressed sensing (CS) algorithms. However, most typical CS algorithms cannot achieve a balance between algorithm efficiency and 3D reconstruction accuracy. Due to difficulties in flight path control of airborne SAR, it is hard to realize registration of SAR images with frequency-domain imaging algorithms because of time-varying baseline. To address these problems, an efficient 3D image reconstruction method for airborne TomoSAR based on back projection (BP) algorithm and improved adaptive iterative shrinkage-thresholding algorithm (IA-ISTA) is proposed. First, 2D images are achieved with BP algorithm on the ground plane. After registration of SAR images, 3D image reconstruction results in the elevation direction are realized with IA-ISTA. Selection criterion of IA-ISTA parameters are given in this paper. At last, final 3D image reconstruction results are achieved after geometrical transformation based on geometric relationship. Both simulated data and measured data of a P-band airborne TomoSAR system are used. 3D image reconstruction results show that the proposed method outperforms traditional methods regarding efficiency and accuracy, which proves the validity and practicality of the proposed method.

Recent research has shown that generative adversarial networks (GANs) have been successfully applied in aligning features for domain adaptation. However, the extracted features might lose category-specific information, because they are distinguished as either a source or a target during the adversarial training. To address this issue, a two-stage training framework consisting of two sets of GANs and a dedicated classifier is proposed in this study. In the pretraining stage, we use an encoder–decoder–classifier structure to obtain discriminative and representative features of the source domain and use it as a reference in the subsequent training. In the adversarial training stage, two sets of GANs are used to align target-domain features with those of the source domain and transfer samples of the target domain to the source domain simultaneously. A dedicated classifier is trained along with the adversarial loss to force the generated features of the target domain to carry category-specific information, which significantly improves classification performance. The features of the source domain stay intact in the adversarial training stage. Thus, our approach can alleviate the training burden of GANs. The proposed method has been validated on digital datasets and office31 datasets. Experimental results demonstrate that an average accuracy of 96.5% and 95.1% is achieved, which leads to superior or comparable performance to state-of-the-art results.

Multichannel SAR systems have grown rapidly over the past decade due to their powerful high-resolution and wide-swath (HRWS) capabilities. Because spatially separated channels also have the potential to suppress jamming, dual-channel cancellation is a general method that is effective regardless of the type of jamming signal. In this paper, the principle of dual-channel cancellation (DCC) is introduced, and several practical problems using DCC are also discussed. Moreover, this paper emphasizes interferometric phase estimation, which is the key to DCC. If the jamming-to-signal ratio (JSR) is high, the interferometric phase can be estimated accurately from the interferometry of two channel signals, but estimation becomes rather difficult when the JSR decreases. To solve the problem of interferometric phase estimation under a low JSR, a novel interferometric phase estimation method using cosine similarity is proposed in this paper. L-band airborne dual-channel SAR is performed to investigate the applicability of the method. The results not only prove that cosine similarity is an effective method for interferometric phase estimation, but also demonstrate the potential of DCC in the SAR anti-jamming processing.

Intrusion detection of abnormal objects is critical to avoid traffic accidents and ensure the safety of train operations. Computer-vision based approaches using RGB images have been intensively investigated for intrusion detection at daytime. However, the abnormal object detection using infrared images at nighttime remains more challenging because training samples of infrared images are limited to address this issue, we propose a data augmentation strategy motivated by image style transfer using CycleGAN. First, the synthetic images are generated which conditioned on railway scene images at daytime and non-railway scene images at nighttime. Then, a SSD object detection model is trained using the generated synthetic samples. Finally, the trained SSD model is used to detect abnormal objects for infrared images at nighttime. Experimental results demonstrate that the proposed data augmentation strategy and the object detection method for nighttime scene is effective.