Diling Liao

Multimodal image matching (MMIM) poses persistent challenges in remote sensing due to substantial variations in radiometric properties, scale, and rotation across different modalities. Traditional approaches typically employ pyramid scale space constructions to handle scale differences. However, the processing of multi-scale features often incurs a high computational cost and reduced efficiency. To address this issue, this paper proposes a novel MMIM algorithm based on log-polar sampling (LPS), which achieves both scale and rotation invariance while significantly reducing redundant computations. The proposed method begins with a feature detection strategy based on a single-scale edge thinning map (ETM), which is generated by combining Gaussian steerable filtering with an iterative edge thinning process to extract edge points accurately. To mitigate the impact of radiometric discrepancies between modalities, a quantized orientation map (QOM) is constructed using radiometrically invariant feature orientations. LPS is subsequently performed within neighborhoods of keypoints to generate logpolar patches (LPPs) that are inherently invariant to scale and rotation. These LPPs are then encoded into feature descriptors via orientation distribution histograms. A two-stage matching strategy is further introduced, incorporating keypoint elimination and descriptor reconstruction to enhance robustness. Extensive experiments conducted on six types of typical multimodal images demonstrate that the proposed algorithm consistently outperforms state-of-the-art MMIM methods in terms of both accuracy and computational efficiency. The source code and datasets will be publicly available at https://github.com/liujy325/LPSD.

The joint classification of hyperspectral images (HSI) and LiDAR data is an important research direction in the field of remote sensing image processing. However, existing methods are mostly limited to inadequate feature extraction or simple serial fusion strategies, failing to fully exploit the deep cross-modal relationships between LiDAR and HSI data. Meanwhile, global modeling based on Transformers, despite capturing long-range dependencies, faces difficulties in handling high-dimensional remote sensing data due to quadratic computational complexity, leading to constrained global-local feature collaboration. To address these issues, this paper proposes a novel hierarchical multi-scale Mamba network (HMMamba) for joint classification of HSI and LiDAR data. First, the method adopts a hierarchical design to obtain rich multi-level feature representations through multi-scale feature extraction. Second, it utilizes adaptive weight allocation to achieve dynamic cross-modal feature integration. Finally, it constructs a Fused Feature enhancement Mamba Block (FFMB), which integrates a dual-branch attention mechanism and selective state space modeling to achieve efficient global context modeling under linear computational complexity Extensive experiments on four public datasets (Houston2013, MUUFL, Trento, and Augsburg) demonstrate that HMMamba significantly outperforms existing state-of-the-art methods in terms of classification performance. Specifically, on Augsburg dataset, the OA improvement is 3.24% compared to the suboptimal method, fully validating the superior performance of the proposed method. The codelink of the proposed method is https://github.com/leiyeqi/HMMamba.

When hyperspectral image classification encounters high-dimensional spectral channels, the feature utilization rate is often low due to significant redundancy between channels and uneven discriminatory power. Furthermore, different land cover types exhibit notable differences in scale and morphological changes across space. This structural diversity poses significant challenges to spatial feature modeling. To address this problem, this paper proposes a novel framework based on the Mamba model-PS2Mamba for hyperspectral image classification. This framework integrates three strategies: spectral fine modeling, global perceptual scale adaptation, and multi-scale spatial structure modeling. Firstly, this paper designs a statistically enhanced normalized band refinement (SENBR) module, which dynamically enhances or suppresses channel features based on channel correlation, variability, and importance, effectively suppressing redundant and noisy bands. Secondly, a global-aware scale adaptation (GASA) module is proposed. By incorporating a scale scoring network and a multi-directional modeling mechanism, this module enables adaptive perception and directional enhancement modeling of spatial structures at different scales. Finally, a lightweight multi-scale spatial structure extraction module, LightPyramid, is constructed. By enriching spatial semantic information through multi-branch parallel convolutions, it enhances the model’s ability to represent complex surface structures, while maintaining spatial resolution. Experimental results on four representative hyperspectral datasets, including Pavia University, Salinas, and two UAV-based datasets (HongHu and HanChuan), demonstrate that PS2Mamba achieves overall accuracies of 98.77%, 99.61%, 96.20%, and 96.01%, respectively. Compared with existing CNN, GCN, Transformer, and Mamba based models, PS2Mamba achieves up to 12.91% improvement in accuracy, showing superior generalization and robustness, particularly under small-sample conditions.

In recent years, convolutional neural networks (CNNs) have been impressive due to their excellent feature representation abilities, but it is difficult to learn long-distance spatial structures information. Unlike CNN, graph convolutional networks (GCNs) can well handle the intrinsic manifold structures of hyperspectral images (HSIs). However, the existing GCN-based classification methods do not fully utilize the edge relationship, which makes their performance is limited. In addition, a small number of training samples is also a reason for hindering high-performance hyperspectral image classification. Therefore, this paper proposes a hybrid CNN-GCN network (HCGN) for hyperspectral image classification. Firstly, a graph edge enhanced module (GEEM) is designed to enhance the superpixel-level features of graph edge nodes and improve the spatial discrimination ability of ground objects. In particular, considering multiscale information is complementary, a multiscale graph edge enhanced module (MS-GEEM) based on GEEM is proposed to fully utilize texture structures of different sizes. Then, in order to enhance the pixel-level multi hierarchical fine feature representation of images, a multiscale cross fusion module (MS-CFM) based on the CNN framework is proposed. Finally, the extracted pixel-level features and superpixel-level features are cascaded. Through a series of experiments, it has been proved that compared with some state-of-the-art methods, HCGN combines the advantages of CNN and GCN frameworks, can provide superior classification performance under limited training samples, and demonstrates the advantages and great potential of HCGN. The source code of the proposed method will be available publicly at https://github.com/Dilingliao/HCGN.

With the increasing number of remote sensing (RS) data sources, the joint utilization of multimodal data in Earth observation tasks has become a crucial research topic. As a typical representative of remote sensing data, Hyperspectral images (HSI) provide accurate spectral information, while rich elevation information can be obtained from Light Detection and Ranging (LiDAR) data. However, due to the significant differences in multimodal heterogeneous features, how to efficiently fuse HSI and LiDAR data, which remains one of the challenges faced by existing research. In addition, the edge contour information of images is not fully considered by existing methods, which can easily lead to performance bottlenecks. Thus, a joint classification network of HSI and LiDAR data based on Mamba (HLMamba), is proposed. Specifically, a gradient joint calculation (GJA) is first performed on LiDAR data to obtain the edge contour data of the land distribution. Subsequently, a multimodal feature extraction module (MFEM) was proposed to capture the semantic features of HSI, LiDAR, and edge contour data. Then, to efficiently fuse multimodal features, a novel deep learning (DL) framework called Mamba, was introduced, and a multimodal Mamba fusion module (MMFM) was constructed. By efficiently modeling the long-distance dependencies of multimodal sequences, the MMFM can better explore the internal features of multimodal data and the interrelationships between modalities, thereby enhancing fusion performance. Finally, to validate the effectiveness of HLMamba, a series of experiments were conducted on three common HSI and LiDAR datasets. The results indicate that HLMamba has superior classification performance compared to other the state-of-the-art DL methods. The source code of the proposed method will be available publicly at https://github.com/Dilingliao/HLMamba.

In the past, deep learning (DL) technologies have been widely used in hyperspectral image classification tasks. Among them, convolutional neural networks (CNNs) use fixed size receptive field (RF) to obtain spectral and spatial features of hyperspectral images (HSIs), showing great feature extraction capabilities, which are one of the most popular DL frameworks. However, the convolution using local extraction and global parameter sharing mechanism pays more attention to spatial content information, which changes the spectral sequence information in the learned features. In addition, CNN is difficult to describe the long-distance correlation between HSI pixels and bands. To solve these problems, a spectral-spatial fusion Transformer network (S ² FTNet) is proposed for the classification of hyperspectral images. Specifically, S ² FTNet adopts the Transformer framework to build a spatial Transformer module (SpaFormer) and a spectral Transformer module (SpeFormer) to capture image spatial and spectral long-distance dependencies. In addition, an adaptive spectral-spatial fusion mechanism (AS ² FM) is proposed to effectively fuse the obtained advanced high-level semantic features. Finally, a large number of experiments were carried out on four datasets, Indian Pines, Pavia, Salinas and WHU-Hi-LongKou, which verified that the proposed S ² FTNet can provide better classification performance than other the state-of-the-art networks.

In recent years, due to its powerful feature extraction ability, the deep learning method has been widely used in hyperspectral image classification tasks. However, the features extracted by classical deep learning methods have limited discrimination ability, resulting in unsatisfactory classification performance. In addition, due to the limited data samples of hyperspectral images (HSIs), how to achieve high classification performance under limited samples is also a research hotspot. In order to solve the above problems, this paper proposes a deep learning network framework named the three-dimensional coordination attention mechanism network (3DCAMNet). In this paper, a three-dimensional coordination attention mechanism (3DCAM) is designed. This attention mechanism can not only obtain the long-distance dependence of the spatial position of HSIs in the vertical and horizontal directions, but also obtain the difference of importance between different spectral bands. In order to extract the spectral and spatial information of HSIs more fully, a convolution module based on convolutional neural network (CNN) is adopted in this paper. In addition, the linear module is introduced after the convolution module, which can extract more fine advanced features. In order to verify the effectiveness of 3DCAMNet, a series of experiments were carried out on five datasets, namely, Indian Pines (IP), Pavia University (UP), Kennedy Space Center (KSC), Salinas Valley (SV), and University of Houston (HT). The OAs obtained by the proposed method on the five datasets were 95.81%, 97.01%, 99.01%, 97.48%, and 97.69% respectively, 3.71%, 9.56%, 0.67%, 2.89% and 0.11% higher than the most advanced A2S2K-ResNet. Experimental results show that, compared with some state-of-the-art methods, 3DCAMNet not only has higher classification performance, but also has stronger robustness.