Supervised learning models have achieved remarkable success in the field of remote sensing, but their applicability is limited by the significant requirement of high-quality labeled data. This article presents the Triplet Adversarial Contrastive Learning (TACL) model, as a self-supervised feature extractor. Considering the potential contrastive semantic conflicts, which may occur due to the different descriptive abilities of various modalities, TACL constructs a triplet contrastive learning framework aligning the two original modalities and a fused modality. To augment the acquired representations of the model, TACL introduces an Adversarial Hard-negative Sample Generation strategy, aiming to boost the resemblance between the feature vectors of negative samples and anchors. Additionally, a ConvNeXt-based lightweight encoder is designed as the foundational backbone of the model, specifically enriching of the expression of central features. A series of few-shot classification experiments substantiate the exceptional performance of the features extracted by TACL, with the simplistic classifier SVM. As a label-free pre-training approach, TACL holds great potential for enhancing the performance of various multimodal remote sensing tasks in scenarios with limited label availability.

Maritime thermal infrared (IR) imaging is susceptible to various types of mixed noise interference, such as Gaussian, stripe, and internal stray radiation noise, due to atmospheric radiation and complex dynamic backgrounds. These factors directly hinder subsequent image processing applications such as target detection and tracking. In this article, we propose a hybrid denoising model based on a model reconstruction framework to jointly address nonuniformity issues in single-frame thermal IR images. Stripe noise, due to its directional nature, exhibits distinct physical characteristics in both the spatial and frequency domains. Most existing methods rely on designing prior constraints based on features from only one domain. To adapt to diverse stripe distribution patterns, we introduce spatial priors for stripes in the hybrid denoising model, concurrently pioneering the introduction of frequency-domain priors in a plug-and-play technology for the first time. This approach facilitates mutual enhancement of multidomain characteristic constraints in stripe removal. Furthermore, we use polynomial fitting to characterize the distribution of internal stray radiation noise, which possesses global smoothness characteristics. The hybrid model effectively mitigates the impact of stripe distribution on the smoothing correction. Experiments demonstrate the effectiveness and superior performance of the proposed hybrid model in addressing mixed noise. Furthermore, we degraded the hybrid model into a destriping model and demonstrated its outstanding performance in the standalone destriping task. The proposed method can provide valuable references for the restoration of thermal IR or hyperspectral remote sensing images and offer additional information for IR target detection and tracking tasks.

Polarimetric hyperspectral images (PHSI) can provide complementary representations of a scene from the perspectives of images, spectra and polarization at the same time, and are expected to improve the quality of scene description. In this paper, the clustering for PHSI is deemed to be a multi-view clustering task, and a tensorial polarimetric-spectral multi-view subspace clustering (TPS-MSC) algorithm for PHSI is proposed. It constructs a small size dictionary, instead of a large self-representative dictionary, by pre-clustering each view independently to give a sparse representation of all the data. Then the view-specific representation matrices are tensorized to explore the low-rank structure among different views, and the consistency of all views in pre-clustering is incorporated into the representation learning framework to strengthen the inter-view correlations. The proposed model is efficiently optimized by the alternative direction minimization of multipliers (ADMM) algorithm. Some experiments are carried out to validate the capacity of PHSI for target identification, and to demonstrate the accuracy and efficiency of the proposed TPS-MSC algorithm.

Small target detection is the primary technology for infrared search and tracking (IRST) systems and plays a vital role in practical applications. Existing algorithms have the following challenges: 1) insufficient local and nonlocal feature extraction and 2) imbalance between accuracy and real-time detection performance. In this study, a novel model for fast detection based on interpatch correlation enhancement (IPCE) and joint local visual saliency prior is proposed to overcome such issues. Regarding the correlation in interpatch dimension, the improved tensor nuclear norm is used to further extract the low-rank structure of the background tensor, which fully exploits the low-rank component and reduces the iteration times. Furthermore, with the hypothesis that the target is locally saliency, a prior model based on the visual saliency mechanism is proposed as the constraint of the target tensor. It effectively reduces the false detection of the sparse edge structure. In general, the proposed IPCE jointly exploits both local and nonlocal correlation of the original image, achieving robustness in different scenarios. Finally, the proposed model is solved by the alternating direction method of multipliers (ADMM). Experiments on seven datasets demonstrate that IPCE outperforms the state of the arts in terms of the balance between detection efficiency and accuracy.

The infrared small-target-detection algorithm has theoretical significance and military value. Meanwhile, it is also a challenging task, especially to enhance greatly the true targets from the intricate background clutters at a low signal-to-noise ratio. In this letter, a multiscale local gray dynamic range (MLGDR) method is presented based on the assumption that the target and the background have different gray dynamic ranges in the local areas. Consequently, the final MLGDR map is innovatively achieved by the two proposed properties, including the local multiscale differences in the gray distribution changes and in the gray values. The results of the experiments indicate that the proposed method is capable of enhancing the target and suppressing the background clutter simultaneously. In particular, compared with the baseline methods, our method achieved a high signal-to-noise ratio, a high detection rate, and a low false-alarm rate under various scenes.