Few-shot object detection (FSOD) in remote sensing faces a fundamental challenge in balancing contextual feature learning with representation stability. Current approaches either excel at capturing rich contextual relationships through multimodal architectures but struggle with catastrophic forgetting, or maintain stable feature representations through incremental learning frameworks at the cost of contextual understanding. This limitation is particularly acute in remote sensing, where complex spatial-contextual dependencies form distinctive co-occurrence patterns important for accurate object detection. We present MemDeT, a memory-augmented detection transformer that bridges this gap through three key modules: (1) a contextual layer-wise fusion (CLF) module that progressively integrates visual and textual information across transformer layers through an adaptive attention mechanism and unified cross-modal fusion, enabling both fine-grained object feature extraction at lower layers and abstract contextual relationship learning at higher layers; (2) a unified episodic memory (UEM) that serves as a dynamic knowledge repository, employing similarity-surprise based update mechanisms with structured key-value memory organization to strategically retrieve and update relevant past experiences while preserving base knowledge through contextual scoring and dynamic retrieval; and (3) a memory-augmented decoder (MAD) that generates context-aware queries by combining current visual observations with accumulated contextual knowledge through memory-prioritized attention and progressive query refinement. Extensive experiments on remote sensing datasets, NWPU VHR-10, DIOR and iSAID, demonstrate that MemDeT significantly outperforms state-of-the-art methods in both contextual understanding and knowledge retention. Cross-dataset evaluation further validates MemDeT’s robust generalization capabilities, successfully transferring knowledge from DIOR to iSAID’s urban scenes, particularly in scenarios requiring strong contextual understanding with limited training examples.

Few-shot object detection (FSOD) addresses the challenge of limited labeled data by enabling detectors to learn from minimal annotations. Recent work on image–text fusion has shown promise in overcoming data scarcity issues. However, these models suffer from catastrophic forgetting due to two distinct challenges during few-shot adaptation. First, feature space distortion disrupts established relationships between image–text modalities and blurs decision boundaries, leading to inaccuracies in object classification. Second, attention drift impairs the precise visual–textual alignments learned during base training, limiting the model’s ability to focus on relevant regions for accurate object localization. To overcome these limitations, we propose Protoalign, a novel prototype-guided multilayer alignment network that maintains robust cross-modal relationships across multiple network layers while mitigating catastrophic forgetting. Specifically, cross-modal prototype guidance (CPG) enables stable feature learning through class-specific prototype fusion to mitigate feature space distortion. Multimodal feature aggregation (MFA) strengthens feature relationships through channel-level interactions to overcome attention drift. Moreover, we propose an integrator that facilitates consistent information flow between network layers. Protoalign progressively refines multimodal features, enabling effective novel class adaptation while preserving crucial base knowledge. The refined representations are then processed by a detection transformer (DETR) decoder for accurate object detection. Extensive experiments on iSAID, DIOR, FAIR1M-Airplane, and NWPU VHR-10 datasets demonstrate that Protoalign achieves superior performance while significantly reducing catastrophic forgetting compared to existing methods.

Despite DETR-like methods having improved end-to-end detection capabilities, they fundamentally struggle with the uniform processing of entire flattened feature maps, causing queries to attend to irrelevant regions. This results in redundant attention patterns, leading to computational burdens and inefficiencies in the detection of tiny objects in remote sensing imagery across dense and sparse settings. To address these limitations, we present a lightweight Transformer-based encoder-decoder architecture called Dynamic Adaptive Region Transformer (DART). Specifically, Density Adaptive Transformer (DAT) employs an adaptive-region attention (ARA) mechanism that dynamically generates content-aware spatial regions rooted in feature density and semantic relevance. This strategy prioritizes computational resources on semantically rich areas while minimizing focus on irrelevant background regions. Region Aware Decoder (RAD) incorporates a masked region-aware cross-attention (MRA) mechanism, where queries interact exclusively with the adaptive masked regions generated by the encoder, thereby reducing redundant focus on overlapping or irrelevant areas. Meanwhile, a query diversity loss is introduced to penalize overlapping attention patterns among queries, encouraging each query to focus on distinct and complementary regions. By adapting to data density and directing queries to essential areas within the image, DART enhances feature extraction and object localization for various-sized objects in dense and sparse settings. Experimental results demonstrate that DART achieves state-of-the-art performance on AI-TOD, DOTA-v2.0 and LEVIR-Ship benchmarks and exhibits strong generalization capabilities on DIOR, while using only 13 million parameters and a computational cost of 68 GFLOPs.

Accurately and robustly detecting infrared (IR) small targets plays an essential role in IR search and tracking (IRST) systems against complex backgrounds. However, the low signal-to-clutter ratio (SCR) feature target, also known as the dim target, is more similar to the local background region, making it prone to being misdetected as background or buried in clutter interferences, ultimately leading to detection failure. This has become a common and unavoidable challenge in the field. To alleviate this issue, an interpretation weighted sparse (IWS) method is proposed to detect IR small and dim targets while maintaining clutter suppression capability. The interpretability operation of the proposed method stems from an analysis of IR target characteristics, including dim target feature extraction and clutter suppression. Based on the multisubspaces model, the relevant feature of the dim target is mined by a designed structure tensor, improving the target’s sparsity. Then, the contrast characteristic is introduced to eliminate the cost of dim target mine operation, namely residual clutter interference. Finally, the IWS map is obtained by fusing the target detection module and clutter suppression module, after which an adaptive threshold segmentation operation is adopted to extract the small targets. A series of experimental results and application discussion cases demonstrate that IWS outperforms other existing methods regarding detection performance while also exhibiting background suppression ability.

Infrared (IR) small target detection is a critical technology in both military and civil applications. Nowadays, the optimization-based methods are widely concerned for their good performance under complex backgrounds. However, these optimization models are constructed simply, without sufficient information to fully characterize the features of an image. Moreover, the true target is unprotected in the target weight, and appropriate structural constraints on background are also deficient. Consequently, an IR small target detection algorithm based on target-and-background-weighted two-layer nested model is proposed in this letter. Initially, a two-layer nested model is designed to characterize the global and the local structure of an IR image simultaneously. Then, the smallest eigenvalue of structure tensor (SEoST) is introduced to reweight the overlapping edge information (OEI) map as a target weight, in order to reduce the false alarm rate without missed detection. Furthermore, a weight matrix of background is elaborated as a structural constraint, in order to get a more accurate background estimation result. Finally, the alternating direction multiplier method (ADMM) with a soft-threshold solver is designed to solve the proposed optimization model. Experiments are carried out to evaluate the proposed method by comparing with eight baseline methods, and the results show that the proposed method achieves superior detection performance under complex scenes.

The accurate detection of small infrared (IR) targets in complex backgrounds is crucial for the effective operation of IR search and track (IRST) systems. In this letter, we propose a novel multidirectional diffusion difference measure (MDDDM) approach for accurate and low-complexity IR small target detection. MDDDM consists of three stages. First, a gray intensity measure (GIM) is generated to highlight the important features and suppress high-intensity background regions by using a computationally efficient difference of Gaussian (DoG) filter. Second, the diffusion difference measure (DDM) kernel is computed by fusing an improved Gaussian kernel (IGK) and a directional filter kernel (DFK), utilizing feature minimization instead of omnidirectional gradient judgment to suppress background edge clutters. Finally, threshold segmentation operation is used for precise localization and extraction of the target from the background. Experimental results demonstrate that MDDDM outperforms other state-of-the-art algorithms in terms of detection rate and false alarm rate while maintaining the fastest running speed.

The strong edge of bright clutter would produce a large number of false alarms, and it seriously debases the detection performance of infrared (IR) small targets. Existing IR small target detection algorithms often lack adaptability in complex scenes, and their performance heavily relies on initial parameter configurations, including target size, which is closely related to the characteristics of the background and target. This article proposes an IR small target detection algorithm based on adaptive size estimation by a multidirectional gradient filter to address these limitations. First, based on the Gaussian-like distribution of a small target, the multidirectional gradient filter is constructed to enhance the target in order to get the target characteristic map (TCM). Second, the size of the small target is estimated from the enhanced TCMs by the criteria of sinusoidal curve best fitting. Subsequently, a multidirectional morphological filter with the optimal estimated size is proposed to suppress the strong background clutter to obtain the local strength difference map (LSDM). Finally, the corresponding enhanced TCM with the optimal estimated target size is fused with the LSDM for threshold segmentation to further suppress the background and detect the small targets. A large number of experimental results show that the proposed method can not only accurately estimate target size but also effectively detect small targets in diverse, complex scenes.

Existing infrared small target detection algorithms often lack adaptability in complex scenes and heavily rely on parameter configurations. To address this limitation, we propose a novel infrared small target detection method based on adaptive region growing algorithm with iterative threshold analysis that leverages the homogenous compactness of the small target and discontinuity with its surroundings. Initially, the image undergoes adaptive splitting into multiple regions using an automatic seeded region growing (ASRG) algorithm, eliminating the need for preassigned seed points. Next, the segmentation results at each threshold are utilized to calculate the relative residual map (RRM) and local dissimilarity map (LDM), contributing to the selection of the optimal threshold. Finally, RRM and LDM corresponding to the optimal threshold are integrated to accurately characterize the small target signal while effectively removing background clutter. Experimental results show that the proposed method is effective in clutter removal and small target detection in diverse complex scenes, and is robust to the shape and size of targets.

This letter proposes an efficient infrared small target detection method based on the 1-D difference of guided filtering (DoGF). First, the 1-D DoGF is constructed by measuring the difference of image structure fidelity between primitive guided filtering (GF) and local variance weighted GF from the perspective of 1-D signal analysis, which can effectively filter out 1-D noise components and protrude pulse signals. Second, the 1-D row and column DoGF are applied to process the infrared image along horizontal and vertical directions, respectively, and then the row–column crossed DoGF (rcDoGF) and column–row crossed DoGF (crDoGF) are calculated and integrated, which can greatly highlight the pulse-like small target signal and eliminate the background clutter. Finally, the small targets can be extracted with a simple adaptive threshold. Experimental results show that the proposed algorithm has high detection accuracy for small infrared targets under heavy noise interference, as well as for targets with different sizes and shapes.

Convolutional neural network (CNN) model has made a great breakthrough in target detection in remote sensing image due to the excellent feature extraction capability. However, diverse scenes and complex contextual information of remote sensing image make these CNN models face big challenges. For example, the distinctiveness between the target and the context would be reduced greatly. This letter proposes an object-and-scene context constrained CNN method to detect target in remote sensing image. This method has two channels, namely, object context constrained channel and scene context constrained channel. The object context constrained channel uses recurrent neural network (RNN) to explore the contextual relationship between the target and the object, including feature relationship and position relationship. The scene context constrained channel adopts priori scene information and Bayesian criterion to infer the relationship between the scene and the target, and it make full use of the scene information to enhance the target detection performance. The experimental results on two datasets demonstrate the robustness and effectiveness of the proposed method.

This letter proposes a novel method based on multiscale dual morphological reconstruction (MDMR) to detect low-contrast ship target with unknown size and polarity (bright or dark) in infrared image. First, the multiscale morphological reconstruction perceives the structural information of various scales. The target signal map (TSM) is presented to indicate the target polarity and enhance target by background subtraction and logarithmic histogram transform. Next, the boundary constraint entropy threshold selection (BCETS) is designed to extract candidate targets and avoid the over- or under-segmentation problems. Then, the local contour contrast descriptor (LCCD) is constructed and the discriminant rule of low-contrast target is established to identify true target and eliminate false alarms. Finally, the pixel-OR operation fuses the multiscale detection results. Extensive experiments show that the proposed method has a better effectiveness and robustness against compared methods.

Remote sensing image has been widely used in many fields such as military reconnaissance and earthquake relief. However, the complexity and diversity of the scene make the target detection and recognition performance poor. Convolutional neural networks (CNNs) have made breakthrough in remote sensing image processing due to their ability to extract deep features. This letter proposes a remote sensing image scene recognition method based on object relationship reasoning CNN (ORRCNN), which makes use of the relationship between objects to infer the scene information. The method has prior scene-information-based channel and object-detection-based channel to classify the remote sensing image. The prior scene-information-based channel makes use of the feature space to identify the scene, and the object-detection-based channel adopts the relationship between the object and the scene to classify the scene. Afterward, the Bayesian criterion infers the scene more accurately by means of fusing the scene information from the above channels. The experimental results show that the proposed method is excellent especially in the scene where there are iconic objects in the remote image.

Based on the process of generating HDR images from LDR image sequences with different light exposures in the same scene, a new fitting method of camera response curves is proposed to solve the problem that the boundary of the fitting algorithm of camera response curves will be blurred and it is difficult to determine and verify the accuracy of the fitting curves. The optimal response curve is fitted by increasing LDR images step by step through considering the pixel value and texture characteristics. In order to validate the fitting effect of curves, we compare the photographed images and the real images in different time intervals on the basis of HDR images and response curves. We use RGB and gray image experiments to compare the current mainstream algorithms and the accuracy of our proposed algorithm can reach 96%, which has robustness.

In the intricate infrared cloudy-sky background, the edge of cloud might be falsely detected as a target because it is usually similar to a small target in contrast and complexity, which are the criteria that conventional small target detection methods adopt to differentiate between background and target. However, the shape of a small target is isotropic and similar to 2-D Gaussian function, while the strong background edge is anisotropic and tends to spread in a certain direction. In this letter, we propose a method to detect small targets in the intricate infrared cloudy-sky background by a local difference adaptive measure (LDAM). This proposed method uses the local structure tensor to perceive the dominant direction and its uncertainty in the local infrared image and then sets the direction and shape of the filter to calculate the local difference. In this way, the background edge is estimated accurately and suppressed effectively. Extensive experiments show that the proposed method outperforms the baseline methods.