Lane line detection in complex scenarios is a key part of the intelligent driving environment sensing technology, essential for ensuring traffic safety and improving transportation system efficiency. However, existing models ignore the data imbalance of the far-end, near-end, foreground object and background image, it leads to the inconsistency between the detected curve lane line and reality, and the poor complementarity of different scales of information. To solve the above problems, a high-precision and intelligent lane line detection algorithm based on bow height feature points (BHFP) and cross-scale feature correlation network (CFCNet) is proposed, the sub-module functions are shown below. A key point regression method based on BHFP extraction and box intersection over the union loss function correction was used to enhance the automatic modeling ability of the curve lane line. The CFCNet is used to mine the similarity and different characteristics of backbone network output data at different scales. A global information reflow enhancement model is used to highlight the advantages of deep and shallow features in correcting elongated target locations and sharpening structural details. The experimental results show that this algorithm improves the robustness and positioning efficiency of the lane line detection in complex road scenarios, and reduces the missing detection phenomenon of the model. It has significant advantages in the regression localization of lane line far-end targets.

With the proliferation of cloud computing, outsourcing databases has become a common strategy for reducing on premise storage and computation costs. However, this approach raises serious privacy concerns, as sensitive data and query information may be exposed to the cloud. While existing top-$k$query methods have made progress in performance and privacy protection, they offer limited support for the more advanced requirement of diversity constrained queries. In light of this, we present SecDiv, the first privacy-preserving query system that supports diversity constraints over ciphertext in the cloud. SecDiv is built on a two-server distributed trust model and lightweight additive secret sharing, and hides data contents under an honest-but-curious, non-colluding adversary model to ensure that cloud servers learn no sensitive information. SecDiv comprises three customized secure components: SecDMap maps the structured database of the data owner into two secret-shared tables; SecQMap translates each SQL statement and its diversity constraints into vectors whose lengths match the database attributes, thereby hiding targeted attributes and literal values; and SecCQ performs secure filtering, ordering, and top-$k$selection in the cloud, centered on a secure most significant bit comparison implemented by a parallel-prefix adder. A formal security analysis is conducted to provide theoretical guarantees for the security of SecDiv. SecDiv is evaluated on three real datasets, with diversity constraints configured using top-$k$and category count conditions to emulate practical ranking scenarios. Compared with a plaintext baseline, SecDiv achieves identical results with 100% accuracy. Query latency remains practical, with second-level response times in typical settings, and communication overhead increases as expected. Overall, experimental results demonstrate that SecDiv attains a balanced trade-off among privacy, accuracy, and efficiency in real-world cloud service environments.

The latest remote sensing image saliency detectors primarily rely on RGB information alone. However, spatial and geometric information embedded in depth images is robust to variations in lighting and color. Integrating depth information with RGB images can enhance the spatial structure of objects. In light of this, we innovatively propose a remote sensing image saliency detection model that fuses RGB and depth information, named the multimodal guided transformer architecture (MGTA). Specifically, we first introduce the strong correlated complementary fusion (SCCF) module to explore cross-modal consistency and similarity, maintaining consistency across different modalities while uncovering multidimensional common information. Additionally, the global-local context information interaction (GLCII) module is designed to extract global semantic information and local detail information, effectively utilizing contextual information while reducing the number of parameters. Finally, a cascaded feature-guided decoder (CFGD) is employed to gradually fuse hierarchical decoding features, effectively integrating multi-level data and accurately locating target positions. Extensive experiments demonstrate that our proposed model outperforms 14 state-of-the-art methods. The code and results of our method are available at https://github.com/Zackisliuzao/MGTANet.

Moving object detection (MOD) in satellite video sequences faces persistent challenges including low contrast against complex backgrounds, limited motion modeling, and severe scale variation. To effectively address these, we propose STFDNet, a novel combined model-driven and data-driven framework. STFDNet comprises three core components: the hybrid temporal motion module (HTMM), dynamic frequency learning (DFL), and a progressive cascaded learning strategy (PCLS). Initially, HTMM leverages both explicit and implicit strategies to model multi-frame temporal differences, effectively capturing both short- and long-term motion. Then, DFL integrates a frequency-domain dynamic filter to learn global frequency characteristics, enhancing feature representation beyond the spatial domain and improving distinction from background noise. Finally, PCLS progressively refines detection results by fusing multi-domain features in a step-by-step manner, transitioning from coarse-grained to fine-grained representations. Experiments on the Jilin-1 satellite video dataset demonstrate that the proposed method effectively mitigates background noise, limited motion modeling, and object scale variations, significantly improving detection accuracy and robustness, achieving an F1-score of 85.4%.

Achieving accurate 3-D environment perception is a key task in the field of remote sensing. Multispectral point cloud has rich integrated 3-D spatial–spectral information, which provides a data basis for realizing more detailed scene understanding and perception. However, the diversity of land covers and the complexity of its features pose challenges to classification. In addition, the current methods mechanically pool and fuse local features to obtain global information, which has limited the utility for multispectral point cloud classification. In this article, we propose a masking graph cross-convolution network (MGC2N), which aims to address these problems by utilizing spectral features to construct point-to-point relationships independent of spatial distance. A self-attention masking (SAM) module and a spatial–spectral cross-convolution (S2C2) module are innovatively designed into the proposed MGC2N. The former is used to adaptively adjust the nodes and edges of the adjacency matrix to dynamically extract effective features for different land covers; the latter is used to extract spatial distribution features and local spectral features of the land covers in the scene to enhance the discriminative ability of the learned features. Our method achieves the best-in-class performance on two real multispectral point cloud datasets, demonstrating its effectiveness in improving classification accuracy and robustness.

Recently, the hypergraph convolutional network (HGCN) has attracted increasing attention in hyperspectral image (HSI) classification. Compared to graph convolutional networks, HGCN has a stronger ability to mine nonlinear high-order correlations. However, the problems of intraclass variability and interclass similarity exist due to the effects of light, environment, and sensor bias, resulting in insufficient reliability of hypergraphs constructed by directly utilizing the original spectral features. Motivated by the observation that the land cover in HSI contains the spatial distribution semantic information of community structures, which can be used to extract deeper contextual semantic features, we propose a novel community structure guided network (CSGNet) for HSI classification. Specifically, CSGNet adopts a dual-branch architecture: the HGCN branch focuses on superpixel-level high-order feature extraction, while the convolutional neural network (CNN) branch enhances pixel-level local features. In HGCN branch, a novel reliable hypergraph construction approach is introduced, which strikes a balance between depth-first search (DFS) and breadth-first search (BFS), effectively representing different community structure features and improving the ability of edge detection. Meanwhile, kernel function mapping is used to achieve more accurate node connections and enhances classification within classes. Finally, to achieve balanced training of the HGCN and CNN branches, we add their cross-entropy loss as an auxiliary component in the backpropagation process. Experimental results demonstrate that CSGNet outperforms the state-of-the-art methods. The code will be released athttps://github.com/KustTeamWQW/CSGNet.

Multispectral LiDAR point clouds are valuable in remote sensing for their spatial-spectral consistency, yet their high acquisition and annotation costs pose significant challenges. To mitigate this, self-supervised learning has emerged as a promising solution, reducing reliance on annotated data while improving model generalization. However, existing self-supervised frameworks for point clouds often overlook the complexity of ground object distribution in large-scale remote sensing scenarios and fail to leverage the spectral information inherent in multispectral point clouds. In this paper, we introduce the Self-Supervised Spatial-Spectral Distance Reconstruction Network (S4DR-Net), a novel self-supervised pre-training network designed for multispectral point cloud classification. Serving as the key component of the network, the Spatial-Spectral Distance Prediction module (S-SDP) effectively addresses these limitations by reconstructing the distance relationships between voxel blocks in three-dimensional Euclidean as well as spectral spaces. By jointly considering spatial and spectral distances, S-SDP enables the network to learn a unified representation that captures the intrinsic spatial-spectral consistency of multispectral point clouds. This design allows S4DR-Net to generate low-dimensional feature representations in a self-supervised manner, without reliance on manual annotations. We conducted experiments and evaluated on two real-world multispectral point cloud datasets. The results demonstrate that S4DR-Net consistently outperforms existing self-supervised pre-training methods, achieving superior accuracy and generalization compared with current state-of-the-art approaches. The code will be released at https://github.com/KustTeamWQW/S4DR-Net.

Hyperspectral imagery (HSI) delivers detailed spectral information critical for remote sensing applications such as high-accuracy land cover classification, quantitative parameter retrieval, and environmental monitoring. However, satellite-borne HSI often suffers from limited spatial resolution owing to inherent sensor constraints, whereas airborne HSI is constrained by restricted spatial coverage. Reconstructing high-resolution HSI from multispectral imagery emerges as a promising strategy to address these challenges. In this study, we propose the Frequency-Enhanced Spatial-Spectral Network (FESSN), a computationally efficient architecture that innovatively integrates multi-domain fusion across spatial, spectral, and frequency domains to achieve superior reconstruction performance. A key innovation is the neural network-driven frequency enhanced modulation (FEM), which adaptively refines spectral amplitudes and phases via fast Fourier transform, providing interpretable, parameter-efficient enhancements to bridge spatial-spectral modeling gaps. Meanwhile, a Mamba-based Multi-Scale Spatial Fusion module (MMSAF) that seamlessly integrates local features with long-range dependencies, and a U-shaped Spectral Module (USEM) that integrates Mamba and attention mechanisms to model inter-group and intra-group relationships, while adhering to spectral sparsity priors. The experimental results demonstrate FESSN outperforming six state-of-the-art methods in metrics like RMSE (up to 5.31% improvement), PSNR, SAM, ERGAS, and SSIM. Downstream tasks in land cover classification further validate its utility, positioning FESSN as a breakthrough in efficient, high-fidelity HSI reconstruction. To facilitate reproducibility and further research, the code will be publicly available at https://github.com/KustAIRS/TGRS-FESSN.

The heart, responsible for circulating blood throughout our body, contains four chambers. Existing analysis methods primarily focus on one single ventricle. Transthoracic echocardiography provides real-time estimations of cardiac function and enables comprehensive observations of the entire heart, especially through the apical 4-chamber view. However, no current clinical indices evaluate cardiac function considering all four chambers simultaneously. Manual estimation of the four chambers is laborious, inefficient, and complicated by anatomical complexity and variable image quality, including motion artifacts and unclear borders. There is a significant need for a high-performance segmentation tool that can assess all four chambers concurrently. To address this, we collected a clinically representative dataset of 2D apical 4-chamber view echocardiograms, with annotated 4-chamber regions serving as the basis for automatic 4-chamber synergy analysis. We then proposed a boundary-enhanced network, denoted as $BeU^{2}$-Net, tailored for transthoracic echocardiography 4-chamber segmentation using our private dataset. Specifically, our network employs a two-level nested encoder-decoder architecture, utilizing a segmentation-specific residual U-block with a mixture of receptive fields at each stage to capture multi-level and multi-scale features. A dedicated boundary prediction branch, incorporating edge details, is integrated to enhance boundary segmentation performance. Experiments on both private and public datasets demonstrate that our $BeU^{2}$-Net possesses superior boundary detection capabilities and achieves high segmentation performance for echocardiographic images.

Knowledge graph construction is aimed at storing and representing the knowledge of the objective world in a structured form. Existing methods for automatic construction of knowledge graphs have problems such as difficulty in understanding potential semantics and low precision. The emergence of Large Language Models (LLMs) provides an effective way for automatic knowledge graph construction. However, using LLMs as automatic knowledge graph construction engines relies on the embedding of schema layers, which brings challenges to the input length of LLMs. In this paper, we present a framework for Adaptive Construction of Knowledge Graph by leveraging the exceptional generation capabilities of LLMs and the latent relational semantic information of triples, named ACKG-LLM. Our proposed framework divides the knowledge graph construction task into three subtasks within a unified pipeline: triple extraction of open information, additional relational semantic information embedding and knowledge graph normalization based on schema-level embedding. The framework can construct knowledge graphs in different domains, making up for the defects of existing frameworks that need to retrain and fine-tune the internal model. Extensive experiments demonstrate that our proposed ACKG-LLM performs favorably against representative methods on the REBEL and WiKi-NRE datasets. The code is available athttps://github.com/KustTeamWQW/ACKG-LLM

Due to the inherent visual similarity between the camouflaged object and background, camouflaged object detection (COD) is widely recognized as a challenging task in the field of computer vision, and traditional object detection networks often struggle to extract features and accurately identify camouflaged objects. In this paper, we propose an edge-guided pixel level connected component assisted network for COD. Specifically, the edge prior is used to guide object feature extraction and the pixel level connected component obtained from the extracted feature is used to refine the bounding box of the object. We selectively employ a gray-polarization COD dataset to showcase the ability of feature extraction from backgrounds where camouflaged objects may blend in or be occluded. Numerous experiments demonstrate the superiority of our method compared to state-of-the-arts in the case of limited information.

Multispectral LiDAR contributes to the rapid acquisition of 3D spatial and spectral information of land covers, providing more comprehensive features for classification. Despite the impressive performance of existing Graph Neural Networks (GNNs) in point cloud classification, extracting local features with discriminative ability remains challenging in multispectral LiDAR scenes due to the uneven distribution of geometric and spectral information. To enhance the local representation of spectral features, we propose a novel Graph Convolutional Network with Local Topology and Spectral Feature Representation (GCN-LTSFR). The network constructs optimal local topological graphs of corresponding scales based on the feature distribution density of the point cloud to enhance local spectral features. Experimental results demonstrate that the proposed GCN-LTSFR outperforms several state-of-the-art methods on a real multispectral point cloud.

With the development of remote sensing technology for earth observation, the collaborative utilization of Hyperspectral images (HSI) and multispectral images (MSI) has received increasing attention in terrestrial observation. HSI and MSI represent two typical types of optical remote sensing data and can provide rich complementary information. However, the paradoxical problem of high spatial resolution and high spectral resolution leads to difficulties in extracting complementary information. In this paper, we propose an Adaptive Feature Exchange network with Complementary Advantages (AFECAnet). Specifically, we enhance the discriminative feature extraction of HSI-MSI by introducing a spectral-spatial feature enhancement module based on a dual attention mechanism. Subsequently, in order to reduce redundant information, we design an adaptive feature interaction strategy based on batch normalization privatization factors. This strategy helps to accurately replace redundant information and reduces the computational burden on the network. Experimental results demonstrate that the proposed AFECAnet has a significant improvement in HSI-MSI collaborative classification.

Recently, some hybrid networks, combining graph convolutional network (GCN) and convolutional neural network (CNN) into a unified framework, have drawn increasing attention in hyperspectral image (HSI) classification. Compared with the single CNN or GCN architecture, hybrid networks can simultaneously perform feature learning on pixel-level and superpixel-level regions and generate complementary spectral-spatial features. However, existing methods primarily employ simple fusion strategies such as linear combination or concatenation, resulting in the extracted complementary features not being fully exploited and utilized. In this work, we propose a differential feature-enhanced fusion network (DFEFN) for HSI classification. Specifically, DFEFN consists of two different convolutional network architectures, (i.e. GCN and CNN), and a differential feature enhancement fusion (DFEF) module. The features extracted by CNN and GCN can be enhanced and fused through the DFEF module. Experiments results on two benchmark HSI datasets demonstrate that DFEFN achieves better classification performance compared with state-of-the-art methods.

In accomplishing the task of semantic segmentation of RGB-T remote sensing images, there is a great challenge due to severe occlusion, long-tailed data distribution, and insufficient pixel representation of certain objects. The effective use of high-level semantic contexts among various ground object categories is crucial, yet presents considerable diffi-culties. Traditional RGB-T remote sensing image semantic segmentation methods often fail to capture and utilize complex relationships and interdependencies among these categories, leading to reduced semantic segmentation accuracy. This paper proposes a novel, knowledge-enhancement module, empowering the model to utilize human-like commonsense knowledge. Specifically, we firstly employ the self-attention and cross-attention mechanisms to fuse RGB and Thermal features. Subsequently, we collect the weights of the classification layer to get a high-level semantic pool based on all the categories. Alongside this, a prior knowledge graph is developed to enable information propagation among all categories. We applied this knowledge-enhancement module to enhance the Mask R-CNN, named KEMask R-CNN. Experiments results on the RS RGB-T dataset demonstrate the progressiveness of the proposed method.

Recently, the hypergraph neural network (HGNN) has drawn increasing attention in modeling complex high-order correlations. Compared to simple graph neural networks, HGNNs exhibit more powerful representational ability. There are two limitations in the application of hypergraph theory to hyperspectral image (HSI) classification. One is the inadequate explicit representation of semantic information contained in HSI. Another is the loss of pixel-level spectral-spatial information. Thus, an enhanced hypergraph neural network (EHGNN) is proposed to promote the application of hypergraph theory to HSI classification. Specifically, two important enhancements are introduced: 1) the concept of key hypergraph, providing more rich semantic information and improving the interpretability for complex distribution structures, and 2) the integration of convolutional neural network (CNN) and HGNN architectures into an end-to-end framework, the loss of spectral-spatial information at the pixel-level is effectively reduced. Through these two enhancements, EHGNN exhibits a 4% improvement in overall accuracy (OA) on the Pavia University dataset and a 2% improvement in OA on the Xuzhou dataset compared to HGNN. Furthermore, the test results on two HSI datasets demonstrate that our EHGNN achieves competitive performance compared to other state-of-the-art methods.

In recent years, numerous outstanding technologies have been proposed for salient object detection (SOD) in remote sensing images (RSIs), but most of them focus solely on improving performance while disregarding computational, thereby lacking portability and mobility. This article introduces a novel lightweight progressive multilevel feature collaborative network, termed LPMFCNet. This framework constructs progressive feature information through multilevel image content extraction and designs a multichannel interactive deep neural network with information fusion and filtering functions. First, a spatial detail enhancement module (SDEM) is devised to acquire distant feature information through intermediate branch expansion of receptive fields while preserving multiscale information extraction. Second, an advanced semantic interaction module (ASIM) is proposed to model distant dependency relationships between deep semantic features to better identify the positional information of salient objects. Finally, a multilevel feature collaboration module (MFCM) is designed to collaboratively utilize target features from a multilevel perspective, which fully mining deep-level semantic positional information while retaining target detail information. Extensive experimental comparisons are conducted on two remote sensing datasets with 17 advanced methods. Results demonstrate that the proposed method exhibits superior detection performance while maintaining lightweightness. The LPMFCNet only contains 3.26M parameters and runs 0.5G FLOPs for a$256\times 256$image.

In this article, we propose a novel mutual prediction self-supervised learning (MPS2L) method for remote sensing (RS) image change detection (CD). Compared with the previous self-supervised CD methods based on contrastive learning (CL), MPS2L employing a pixel-level training strategy based on masked image modeling (MIM) can effectively train the model to interpret the local scene of RS images. Utilizing global and local scenes and temporal change features extracted from masked bitemporal images to achieve cross-temporal mutual prediction makes the model have the ability to understand the overall observation scene and capture the change information. The training of the two abilities is carried out simultaneously, avoiding the problem of multiobjective conflict or mutual inhibition. To better focus on the changing regions in RS scenes, we further introduce a change feature interaction module (CFIM), comprising spatial and channel feature interaction. The channel interaction module (CIM) can facilitate the cross-temporal transmission of global scene information by channel attention, and the spatial interaction module (SIM) can promote the network to capture information on changing regions by spatial attention. The experimental results on three benchmark RS CD datasets demonstrate the effectiveness and priority of our proposed MPS2L compared to some existing state-of-the-art (SOTA) methods. The source code of the proposed MPS2L will be made available publicly athttps://github.com/KustTeamWQW/MPS2L.

Remote sensing cross-scene classification has always been an important research field, especially in the field of 3-D classification, which is of great significance. Considering the diversity of collection conditions, seasons, and regional styles, deep learning networks well-trained on one source domain dataset tend to suffer from severe performance degradation when applied to other target domain datasets. To tackle the issue, in this article, we propose a new cross-scene classification method, which combines pre-alignment and Shannon entropy constraint to accomplish unsupervised domain adaptive classification (PS-UDA). On the one hand, the pre-alignment employs$L_{2}$-paradigm constraint and Laplace matrix to pre-align the features. With the$L_{2}$-paradigm constraint, the originally distant features of the source and target domain are constrained to the same sphere surface, and it is easier to make the distribution alignment on the sphere surface. Further, the Laplace matrix is used to map the source and target domain. In this way, similar features of the source and target domain are further aligned, and dissimilar features become discrete from each other. On the other hand, this article employs the Shannon entropy constraint to motivate the network to obtain more high-confidence target domain pseudo-labels. In addition, to fully utilize the unlabeled target domain information, the target domain features are augmented using the adjacency matrix. Experimental results of two cross-scene multispectral point cloud classifications demonstrate that the proposed PS-UDA can effectively mitigate the spectral shift issue in cross-scene multispectral point clouds, achieving state-of-the-art performance.

Cross-scene multispectral point cloud classification aims to transfer knowledge of labeled source scenes to improve the discriminability of the model on the unlabeled target scenes. From a novel perspective, we argue that the information transfer between the source and target scenes can be used to solve cross-scene multispectral point cloud classification task. Specifically, we propose a Coupled Graph Convolutional Network (Coupled-GCN) to achieve joint alignment of node- and class-level structures within scenes by passing information between different scenes. To reduce the effect of spectral shift between the source and target scenes and seek scene-invariant intrinsic features, we propose a scene adaptive learning module by optimizing three different loss functions, namely, source classifier loss, domain classifier loss, and target classifier loss as a whole. In the cross-scene multispectral point cloud classification task, the proposed Coupled-GCN can alleviate the spectral shift problem compared to the traditional GCN and achieves an overall F_score of 65.04%.

Multispectral point clouds provide the data basis for finer land cover classification due to the simultaneous spatial and spectral information. How to jointly utilize spatial-spectral information becomes a hot research direction. Benefiting from the excellent performance of graph neural networks (GNNs) on non-Euclidean data, it is well suited to modelling multispectral point clouds to achieve higher classification accuracy. This paper proposes a novel graph convolutional networks with multi-kernel learning (GCN-MKL) for adaptively constructing a graph of multispectral point cloud for finer classification. Specifically, we use multiple base kernels to map the multispectral point cloud into a high-dimensional feature space and learn a linear combination of base kernels through a multi-kernel learning mechanism embedded in the network. The learned multi-kernel graph can effectively measure the high-dimensional similarity between multispectral points. Experimental results demonstrate that the proposed GCN-MKL outperforms several state-of-the-art methods on a real multispectral point cloud.

In real-world scenarios, the information quality provided by RGB and thermal (RGB-T) sensors often varies across samples. This variation will negatively impact the performance of semantic segmentation models in utilizing complementary information from RGB-T modalities, resulting in a decrease in accuracy and fusion credibility. Dynamically estimating the uncertainty of each modality for different samples could help the model perceive such information quality variation and then provide guidance for a reliable fusion. With this in mind, we propose a novel uncertainty-guided trustworthy fusion network (UTFNet) for RGB-T semantic segmentation. Specifically, we design an uncertainty estimation and evidential fusion (UEEF) module to quantify the uncertainty of each modality and then utilize the uncertainty to guide the information fusion. In the UEEF module, we introduce the Dirichlet distribution to model the distribution of the predicted probabilities, parameterized with evidence from each modality and then integrate them with the Dempster-Shafer theory (DST). Moreover, illumination evidence gathering (IEG) and multi-scale evidence gathering (MEG) modules by considering illumination and target multi-scale information respectively are designed to gather more reliable evidence. In the IEG module, we calculate the illumination probability and model it as the illumination evidence. The MEG module can collect evidence for each modality across multiple scales. Both qualitative and quantitative results demonstrate the effectiveness of our proposed model in accuracy, robustness and trustworthiness. The code will be accessible at https://github.com/KustTeamWQW/UTFNet.

Existing RGB-Infrared detection models do not explicitly encourage RGB and infrared to achieve effective multimodal learning. We find that when fusing RGB and infrared images, cross-modal redundant information weakens the degree of complementary information fusion. Inspired by this observation, we propose Redundant Information Suppression Network (RISNet) which suppresses cross-modal redundant information and facilitates the fusion of RGB-Infrared complementary information. Specifically, we design a novel mutual information minimization module to reduce the redundancy between appearance features from RGB images and infrared radiation features from infrared images, which enables the network to take full advantage of the complementary advantages of multimodality and improve the detection performance. Experimental results demonstrate that RISNet outperforms the best competitive algorithm for RGB-Infrared pedestrian detection.

Multispectral point cloud, as a new type of data containing both spectrum and spatial geometry, opens the door to three-dimensional (3D) land cover classification at a finer scale. In this paper, we model the multispectral point cloud as a spatial-spectral graph and propose a smooth graph convolutional network for multispectral point cloud classification, abbreviated 3SGCN. We construct the spectral graph and spatial graph respectively to mine patterns in spectral and spatial geometric domains. Then, the multispectral point cloud graph is generated by combining the spatial and spectral graphs. For remote sensing scene classification tasks, it is usually desirable to make the classification map relatively smooth and avoid salt and pepper noise. Heat operator is introduced to enhance the low- frequency filters and enforce the smoothness in the graph signal. Further, a graph -based smoothness prior is deployed in our loss function. Experiments are conducted on real multispectral point cloud. The experimental results demonstrate that 3 SGCN can achieve significant improvements in comparison with several state-of-the art algori thms.

Multispectral light detection and ranging (MS-LiDAR) systems open the door to the possibility in the 3-D land cover classification at a finer scale using only point cloud data. This article proposes a model based on the tensor representation for multispectral point cloud classification. The proposed method combines the 3-D local spatial structure of each multispectral point by characterizing the point with a second-order tensor. The first mode of the tensor indicates the spatial location and spectral information of each point (i.e., the row of the second-order tensor) and the second mode denotes the neighborhood geometric and spectral structures (i.e., the column of the second-order tensor). Then we develop a novel tensor manifold discriminant embedding (TMDE) algorithm to extract the geometric-spectral features for multispectral point clouds classification. TMDE solves the mapping matrices of each mode by preserving the intraclass samples' distribution further making it more compact and maximizing the distance of different classes. Finally, the support vector machine classifier with the extracted features as input is used to implement the classification of multispectral point clouds. Experiments are conducted on two real multispectral point cloud data sets. The experimental results demonstrate that the proposed method can achieve significant improvements in classification accuracies in comparison with several state-of-the-art algorithms.

A novel discriminative graph-based fusion (DGF) method is proposed for urban area classification to fuse heterogeneous features from two data sources, i.e., hyperspectral image (HSI) and light detecting and ranging (LiDAR) data. The features include spectral characteristics in HSI, height in LiDAR data, and geometry in image processing technologies like morphological profiles (MPs). Our proposed DGF method couples dimension reduction and heterogeneous feature fusion. The core idea of the proposed method is to search for a projection matrix by minimizing the similarity term that preserves the local geometry of each class and maximizing the dissimilarity term that contains the relation of between-class distance. As a result, the proposed method can pull close together samples of the same class while pushing those of different classes apart in the projected space by fusing graphs constructed by different groups of heterogeneous features. The edges of the graphs are measured by kernel. Furthermore, the multiscale DGF (MS-DGF) is introduced to utilize the capability of similarity measure of different scales of kernel and avoid finding the optimal scale simultaneously. Experiments are conducted on real HSI along with LiDAR data. The corresponding results demonstrate that the proposed method can make an effective fusion of heterogeneous features to make full use of the complementary information of HSI and LiDAR, which facilitates fine classification task of urban area, compared with several state-of-the-art algorithms.

To effectively learn heterogeneous features extracted from raw LiDAR point cloud data for landcover classification, a multiple kernel sparse representation classification (MKSRC) framework is proposed in this paper. In the MKSRC, multiple kernel learning (MKL) is embedded into sparse representation classification (SRC). The heterogeneous features are first extracted from the raw LiDAR point cloud data before classification. These features contain useful information from different dimensions, including single point features and neighbor features. Based on feature extraction, on the one hand, MKL is reasonably integrated into the SRC, namely, different base kernels that are constructed with each heterogeneous feature separately are utilized in the process of sparse representation. Furthermore, joint sparsity model is also introduced into the MKSRC framework and multiple kernel joint SRC (MKJSRC) is then proposed. On the other hand, improved kernel alignment (IKA) methods are proposed to more effectively determine the weights of base kernels in both of MKSRC and MKJSRC. Experiments are conducted on three real airborne LiDAR data sets. The experimental results demonstrate that MKSRC and MKJSRC frameworks can effectively learn the heterogeneous features for LiDAR point cloud classification and outperforms the other state-of-the-art sparse representation-based classifiers and the recent MKL algorithm. Moreover, the proposed IKA is helpful to better determine the “optimal” weights of the base kernels in both MKSRC and MKJSRC than in the existing kernel alignment method.

Resent years, sparse representation theory has been widely used in signal processing field. Researchers introduce this theory into the application of pattern recognition and classification and get the sparse representation classifier (SRC). In this paper, we use the SRC to achieve the classification of LiDAR (Light Detection and Ranging) points. To get a better performance, we introduce the kernel method into SRC, for the advancement of kernel in solving nonlinear problem. Also, a joint sparse representation is used for the category similarity of neighboring LiDAR points.

In this paper, we propose a discriminative multiple kernel learning (DMKL) method for spectral image classification. The core idea of the proposed method is to learn an optimal combined kernel from predefined basic kernels by maximizing separability in reproduction kernel Hilbert space. DMKL achieves the maximum separability via finding an optimal projective direction according to statistical significance, which leads to the minimum within-class scatter and maximum between-class scatter instead of a time-consuming search for the optimal kernel combination. Fisher criterion (FC) and maximum margin criterion (MMC) are used to find the optimal projective direction, thus leading to two variants of the proposed method, DMKL-FC and DMKL-MMC, respectively. After learning the projective direction, all basic kernels are projected to generate a discriminative combined kernel. Three merits are realized by DMKL. First, DMKL can achieve a substantial improvement in classification performance without strict limitation for selection of basic kernels. Second, the discriminating scales of a Gaussian kernel, the useful bands for classification, and the competitive sizes of spatial filters can be selected by ranking the corresponding weights, where the large weights correspond to the most relevant. Third, DMKL reduces the computational burden by requiring fewer support vectors. Experiments are conducted on two hyperspectral data sets and one multispectral data set. The corresponding experimental results demonstrate that the proposed algorithms can achieve the best performance with satisfactory computational efficiency for spectral image classification, compared with several state-of-the-art algorithms.

A novel multiple-kernel learning (MKL) model is proposed for urban classification to integrate heterogeneous features (HF-MKL) from two data sources, i.e., spectral images and LiDAR data. The features include spectral, spatial, and elevation attributes of urban objects from the two data sources. With these heterogeneous features (HFs), the new MKL model is designed to carry out feature fusion that is embedded in classification. First, Gaussian kernels with different bandwidths are used to measure the similarity of samples on each feature at different scales. Then, these multiscale kernels with different features are integrated using a linear combination. In the combination, the weights of the kernels with different features are determined by finding a projection based on the maximum variance. This way, the discriminative ability of the HFs is exploited at different scales and is also integrated to generate an optimal combined kernel. Finally, the optimization of the conventional support vector machine with this kernel is performed to construct a more effective classifier. Experiments are conducted on two real data sets, and the experimental results show that the HF-MKL model achieves the best performance in terms of classification accuracies in integrating the HFs for classification when compared with several state-of-the-art algorithms.

In the past decade, with the development of kernel-based machine learning, many different multiple kernel learning (MKL) methods were proposed which focus on selecting the pivotal kernel to be preserved and confirming the optimal kernel combination. In this paper, we address the question mentioned above by using subspace projection method and put forward a linear discriminant based MKL (LDMKL) algorithm. LDMKL algorithm reduces the computational burden and keeps the excellent property of MKL in terms of good classification accuracy by finding the optimal projective direction which makes the intraclass scatter minimum and interclass scatter maximum instead of the time-consuming search for optimal kernel combination. Experimental results indicate that LDMKL algorithm provides the best performances among several the state-of-the-art algorithms while demonstrating satisfactory computational efficiency.

Multiple kernel learning (MKL) is becoming more and more popular in machine learning. Traditional MKL methods usually learn the optimal combinations of both kernels and classifiers as the optimization task which is difficult to be solved. In this paper, we study a Boosting framework of MKL for classification in hyperspectral images. The multiple kernel Boosting (MKBoost) is proposed to solve the MKL problem, which apply the idea of Boosting to the multiple kernel classifiers based on the SVM. Experiments are conducted on different real hyperspectral data sets, and the corresponding experimental results show that MKBoost algorithm provides the best performances compared with the state-of-the-art kernel methods.

In this paper, we present an efficient approach based on multiple kernel learning (MKL) for mapping cloud thickness with MODIS and CloudSat data. In order to adapt the characteristics of radar data, we generalize a signal model from the gas imaging model, and the signal model provides a way for transforming the mapping of cloud thickness into a linear estimate problem. Then, considering the disadvantage of complexity and nonlinearity of the MODIS data, the MKL method which has been shown to improve the performance of many learning tasks is qualified for the mapping of cloud thickness. The MODIS data in a real scenarios is used to test the performance of the develop method and the experimental results indicate that the proposed MKL method outperforms single kernel method for the research of mapping cloud thickness.