Hypersonic targets have complex motion states and high maneuverability. The traditional interactive multi-model (IMM) has low tracking accuracy and a slow convergence speed. Therefore, this paper proposes a strong tracking cubature Kalman filter (CKF) adaptive interactive multi-model (AIMM) based on multiple fading factors. Firstly, this paper analyzes the structure of the CKF algorithm, introduces the fading factor of the strong tracking algorithm into the covariance matrix of the time update and measurement update, and adjusts the filter gain online and in real time, which can reduce the decline infilter accuracy caused by model mismatch. Secondly, Singer model, “current” statistical (CS) model, and Jerk model are selected in the model set of IMM and introduced singular value decomposition (SVD) decomposition to solve the problem that Cholesky decomposition cannot be performed in the CKF due to the model dimension expansion. Last, an adaptive algorithm for the Markov matrix in the IMM is proposed. The transition probability was adaptively modified by the value of the model likelihood function to enhance the proportion of matching models. The simulation results show that the proposed algorithm enhanced the proportion of matching models in the IMM and improved the tracking accuracy by 16.51% and the convergence speed by 37.5%.

Rotating object detection in remote sensing image has made substantial progress. However, the widely used feature extraction and fusion modules cannot accurately represent the spatial location and angle information of remote sensing objects in arbitrary orientation. A detection scheme that accurately expresses spatial angle characteristics is necessary when remote sensing objects are densely distributed. This paper proposes an effective network for rotating object detection in optical remote sensing image (BDR-Net). Firstly, a feature aggregation module (FAM) is designed to refine the feature extraction of rotating objects, which can generate refined feature maps. Secondly, two-dimensional (2-D) Gaussian modeling simulates the rotation angle information of the object. Thirdly, given that the Bhattacharyya distance is mathematically symmetric and rigorous, we design a Bhattacharyya distance based learning module (BDLM). We improve the loss function to more accurately guide the regression of the relative location information of a rotating box. The effectiveness of the proposed method is validated by the results of experiments on three public datasets: DOTA, DIOR, and HRSC2016. The results of our research prove that a Bhattacharyya distance with a Gaussian distribution could be used as an approximate regression loss of the object detection model. This paper offers an original approach to combining object detection with metric learning to apply rotating object detection to instrumentation in a deployable and rapid manner.

Weakly supervised visible-to-infrared image translation in remote sensing scenes is a challenging task that attempts to generate corresponding infrared images from visible images with cross-model manner, thus tackling the acute scarcity of infrared data in certain crucial task scenes. Existing methods frequently concentrate solely on style transfer, resulting in stylistic similarity but physical inconsistency, which limits the authenticity of cross-modal translation. Infrared thermal features stemming from brightness and darkness contrasts among distinct regions due to temperature disparities, are associated with visual representations in foreground objects in visible images. Therefore, this paper proposes ThermalMask, a framework for infrared image generation guided by thermal features, aiming to make the generated images more closely with infrared features in remote sensing. Specifically, a saliency mask generation network (SMGN) and a semantic attention generation network (SAGN) are designed in the generator. The SMGN is used to generate background and foreground masks to preserve the background and most prominent features of the input remote sensing image. Meanwhile, the SAGN generates attention maps through pyramid non-local attention mechanism, guiding the network to adaptively focus on salient regions within both visible and infrared feature maps. In addition, to refine the generated infrared features, a pixel-spectrum multi-spatial constraint (PSC) targeting high-frequency components is designed, enabling the generative network to focus more on expressing infrared features. Extensive experiment results have shown that our method has good performance in visible-to-infrared image translation within remote sensing scenes, surpassing existing state-of-the-art image translation methods. Moreover, it holds promise for application in downstream tasks where infrared data is insufficient.

Tiny objects in remote sensing images only have a few pixels, and the detection difficulty is much higher than that of regular objects. General object detectors lack effective extraction of tiny object features, and are sensitive to the Intersection-over-Union (IoU) calculation and the threshold setting in the prediction stage. Therefore, it is particularly important to design a tiny-object-specific detector that can avoid the above problems. This article proposes the network JSDNet by learning the geometric Jensen-Shannon (JS) divergence representation between Gaussian distributions. First, the Swin Transformer model is integrated into the feature extraction stage as the backbone to improve the feature extraction capability of JSDNet for tiny objects. Second, the anchor box and ground-truth are modeled as two two-dimensional (2D) Gaussian distributions, so that the tiny object is represented as a statistical distribution model. Then, in view of the sensitivity problem faced by the IoU calculation for tiny objects, the JSDM module is designed as a regression sub-network, and the geometric JS divergence between two Gaussian distributions is derived from the perspective of information geometry to guide the regression prediction of anchor boxes. Experiments on the AI-TOD and DOTA datasets show that JSDNet can achieve superior detection performance for tiny objects compared to state-of-the-art general object detectors.

Due to the limitations of infrared image acquisition conditions, many essential tasks currently rely on visible images as the main source of training data. However, single-modal data makes it difficult for downstream networks to show optimal performance. Therefore, converting the more easily obtainable visible images into infrared images emerges as an effective remedy to alleviate the critical shortage of infrared data. Yet current methods typically focus solely on transferring visible images to infrared style, while overlooking the crucial infrared thermal feature during cross-modal translation. To elevate the authenticity of cross-model translation at the feature level, this paper introduces a translation network based on frequency feature mapping and dual patches contrast, MappingFormer, which can achieve cross-modal image generation from visible to infrared. Specifically, the generator incorporates two branches: low-frequency feature mapping (LFM) and high-frequency feature refinement (HFR), both have embedded the Swin Transformer blocks. The LFM branch captures the fuzzy structural from visible images, while the HFR focuses on mapping edge and texture features. The extracted dual-branch frequency features undergo refinement and fusion through cross-attention mechanisms. Additionally, a dual contrast learning mechanism based on feature patch (DFPC) is designed to infer effective mappings between unaligned cross-modal data. Numerous experimental results prove the effectiveness of this method in cross-modal feature mapping and image generation from visible to infrared. This method holds significant potential for downstream tasks where infrared data is limited.