Superpixel segmentation has been demonstrated to be a powerful tool in hyperspectral image (HSI) classification. Each superpixel region can be regarded as a homogeneous region, which is composed of a series of spatial neighboring pixels. However, a superpixel region may contain the pixels from different classes. To further explore the optimal representations of superpixels, a new framework based on two k selection rules is proposed to find the most representative training and test samples. The proposed method consists of the following four steps: first, a superpixel segmentation algorithm is performed on the HSI to cluster the pixels with similar spectral features into the same superpixel. Then, a domain transform recursive filtering is used to extract the spectral–spatial features of the HSI. Next, the $k$ nearest neighbor (KNN) method is utilized to select k $_1$ representative training samples and k $_2$ test pixels for each superpixel, which can effectively overcome the within-class variations and between-class interference, respectively. Finally, the class label of superpixels can be determined by measuring the averaged distances among the selected training and test samples. Experiments conducted on four real hyperspectral datasets show that the proposed method provides competitive classification performances with respect to several recently proposed spectral–spatial classification methods.

Multitask sparse representation method improves the detection performance by constructing multiple associated sub-sparse representation tasks and jointly learning multiple sub-sparse representation tasks, and this method can make use of the spectral information. However, the using of spatial information needs to be improved. This paper designs a hyperspectral image target detection method which can both make use of spectral and spatial information, that is a weighted joint k-nearest neighbor and multitask learning sparse representation method (WJNN-MTL-SR) is proposed. This method mainly consists of the following steps:1) using multitask sparse representation to obtain the representation residuals. 2) weighted joint k-nearest neighbor is used into the joint region of test pixels to obtain the weighted joint Euclidean distance. 3) a decision function, combining the weighted joint Euclidean distance and residuals of the multitask sparse representation, is used to get target detection result. Experimental results demonstrate that the proposed method show better detection performance than state-of-the-art methods.

With the advancement of infrared technology, infrared small target detection plays a crucial role in precise guidance and warning systems, attracting widespread attention from researchers. Due to their characteristics of low contrast and low signal-to-noise ratio, infrared small targets are easily affected by noise and background interference, making accurate detection challenging. To address the problem of infrared small target detection and shape-preserving segmentation in complex backgrounds, we propose a novel network named U2AMFP-Net. Specifically, we improve the residual U-attention-blocks(RUAB) structure, as the original residual U-blocks(RSU) structure tends to retain excessive invalid features of non-target regions when extracting intra-layer and inter-layer information at different scales. To overcome this issue, we employ attention mechanisms to focus the model more on small target features while suppressing irrelevant information. Additionally, we design the multiscale feature pyramid network (MFPN) on the network to avoid the problem of edge information loss caused by excessive skip connections, thereby further improving detection rates. Experimental evaluations demonstrate that our method preserves more complete shapes of infrared weak small targets while ensuring accurate detection. Compared with other state-of-the-art methods, our approach exhibits superior performance across various metrics. Furthermore, we construct a new training dataset containing 10,000 images of infrared small targets using existing datasets.

When light propagates in foggy weather, it is affected and scattered by suspended particles in the air. As a result, images taken in this environment often suffer from blurring, reduced contrast, loss of details, and other issues. The primary challenge in dehazing images is to estimate the transmission coefficient map in the atmospheric degradation model. In this paper, we propose a dehazing algorithm based on the optimization of the “haze-line” prior and non-local self-similarity prior. First, we divided the input haze image into small blocks and used the nearest neighbor classification algorithm to cluster the small patches, which were referred to as “patch-lines”. Based on the characteristics of these “patch-lines”, we could estimate the transmission coefficient map for the image. We then applied the transmission map to a weighted least squares filter to smooth it. Finally, we calculated the clear image using the haze degradation model. The experimental results demonstrate that our algorithm enhanced the image contrast and preserved the fine details, both qualitatively and quantitatively.

Siamese-based trackers have shown excellent performance in unmanned aerial vehicle (UAV) tracking due to their balanced accuracy and speed. However, most existing methods rely heavily on local convolution operations of CNNs for feature extraction, which fails to establish global feature relationships. In addition, the challenge of information imbalance, caused by differences in background and the prevalence of single-object scenarios, remains a significant obstacle in the current domain. This can potentially result in unwanted boundary effects in UAV tracking. To address these challenges, this paper introduces a novel tracker called eccentric loss and adaptive search (ELAS), which is specifically designed for UAV videos and aims to overcome these challenges. The proposed ELAS enhances feature representation capabilities by utilizing multi-input and multi-mapping of information. It also addresses information imbalance through a differentiated loss function. Furthermore, an adaptive search strategy is introduced to accommodate varying search scales across different scenarios. Specifically, an additional multi-layer feature enhancement (MLE) network is first introduced, which absorbs the middle and high layer information in the backbone network and directs the features to focus on global connections using self-attention, avoiding the loss of useful semantic information due to falling into local optima. Then, a novel eccentric loss is designed to guide network training, enhancing the saliency of object features on the feature map by accounting for the volatility of negative samples and estimating classification performance. Finally, an adaptive multi-scale search strategy is proposed to detect the saliency of objects through between-class segmentation effectiveness, and thus adjust the proportional scale parameter. Extensive experiments on challenging UAV benchmarks demonstrate that ELAS achieves state-of-the-art performance with real-time speed, confirming its effectiveness and robustness.

Infrared small target detection (IRSTD) is inherently challenging due to extremely low signal-to-noise ratios (SNR) in complex backgrounds. Effective detection necessitates capturing local context and suppressing noise to enhance target discrimination. To this end, we propose MSSLNet, which significantly improves detection performance by minimizing irrelevant noise interactions and capturing multi-scale local context. Specifically, a Sparse Large-kernel Convolution Block (SLCB) is introduced, utilizing a sparse-dense dual-branch design to bolster target response while preserving critical information flow. Based on SLCB, a Multi-Scale Sparse Large-kernel Block (MSSLB) is developed to model contextual information for effective target-background separation, further enhanced by a Cross-Gated Module (CGM). Additionally, a Bi-directional Guided Fusion (BGF) module is proposed to optimize cross-layer detail-semantic fusion within skip connections. With only 0.293 M parameters and 2.99 GFLOPs, MSSLNet is highly efficient. Extensive experiments on the NUAA-SIRST and IRSTD-1K datasets demonstrate that our method achieves state-of-the-art performance in both accuracy and efficiency. The code is available at https://github.com/xiaodacheng01/msslnet-main.