Hyperspectral image classification faces significant challenges due to high-dimensional spectral redundancy and complex spatial-spectral dependencies. While existing MambaHSI models leverage state-space modeling to enhance representation learning, their unidirectional formulation fails to fully capture bidirectional spatial interactions and cross-band contextual dependencies. Moreover, the uniform projection mechanism struggles to effectively distinguish spectral variations across different wavelengths. To address these limitations, we propose MambaHSI+, a novel framework that integrates bidirectional state-space modeling with spectral trajectory learning. The proposed architecture introduces three key innovations: (1) A bidirectional context modeling module, enhanced by reverse-order modeling, enables multi-directional spatial information propagation through recursive state transitions, facilitating comprehensive local-global feature aggregation while maintaining linear computational efficiency; (2) a spectral trajectory learning paradigm that formulates spectral evolution as continuous state-space process with bidirectional propagation, effectively encoding cross-band relationships; and (3) a Mamba-enhanced channel attention mechanism that adaptively emphasizes discriminative spectral features via selective state-space transformations. Extensive experiments on four benchmark datasets demonstrate state-of-the-art performance, achieving an average accuracy improvement of 2.51% over MambaHSI. By integrating state-space systems with advanced Mamba mechanisms, MambaHSI+ establishes a new paradigm for spectral-spatial representation learning, significantly advancing classification accuracy while ensuring computational efficiency. The latest logs are now available at https://github.com/RockAilab/MambaHSI_Plus.

In the fields of 3D modeling, analysis of discontinuities and engineering calculation, surface extraction is of great importance. The rapid development of photogrammetry and Light Detection and Ranging (LiDAR) technology facilitates the study of surface extraction. Automatic extraction of rock surfaces from 3D rock-mass point clouds also becomes the basis of 3D modeling and engineering calculation of rock mass. This paper presents an automated and effective method for extracting rock surfaces from unorganized rock-mass point clouds. This method consists of three stages: (i) clustering based on voxels; (ii) estimating major orientations based on Gaussian Kernel and (iii) rock surface extraction. Firstly, the two-level spatial grid is used for fast voxelization and segmenting the point cloud into three types of voxels, including coplanar, non-coplanar and sparse voxels. Secondly, the coplanar voxels, rather than the scattered points, are employed to estimate major orientations by using a bivariate Gaussian Kernel. Finally, the seed voxels are selected on the basis of major orientations and the region growing method based on voxels is applied to extract rock surfaces, resulting in sets of surface clusters. The sub-surfaces of each cluster are coplanar or parallel. In this paper, artificial icosahedron point cloud and natural rock-mass point clouds are used for testing the proposed method, respectively. The experimental results show that, the proposed method can effectively and accurately extract rock surfaces in unorganized rock-mass point clouds.

Impact craters refer to the most salient features on the moon surface. They are of huge significance for analyzing the moon topography, selecting the lunar landing site and other lunar exploration missions, etc. However, existing methods of impact crater detection have been largely implemented on the optical image data, thereby causing them to be sensitive to the sunlight. Thus, these methods can easily achieve unsatisfactory detection results. In this study, an original two-stage small crater detection method is proposed, which is sufficiently effective in addressing the sunlight effects. At the first stage of the proposed method, a semantic segmentation is conducted to detect small impact craters by fully exploiting the elevation information in the digital elevation map (DEM) data. Subsequently, at the second stage, the detection accuracy is improved under the special post-processing. As opposed to other methods based on DEM images, the proposed method, respectively, increases the new crusher percentage, recall and crusher level $F_1$ by 4.89%, 5.42% and 0.67%.

China’s Chang’e lunar exploration project obtains digital orthophoto image (DOM) and digital elevation model (DEM) data covering the whole Moon, which are critical to lunar research. The DOM data have three resolutions (i.e., 7, 20 and 50 m), while the DEM has two resolutions (i.e., 20 and 50 m). Analysis and research on these image data effectively help humans to understand the Moon. In addition, impact craters are considered the most basic feature of the Moon’s surface. Statistics regarding the size and distribution of impact craters are essential for lunar geology. In existing works, however, the lunar surface has been reconstructed less accurately, and there is insufficient semantic information regarding the craters. In order to build a three-dimensional (3D) model of the Moon with crater information using Chang‘e data in the Chang‘e reference frame, we propose a four-step framework. First, software is implemented to annotate the lunar impact craters from Chang’e data by complying with our existing study on an auxiliary annotation method and open-source software LabelMe. Second, auxiliary annotation software is adopted to annotate six segments in the Chang’e data for an overall 25,250 impact crater targets. The existing but inaccurate craters are combined with our labeled data to generate a larger dataset of craters. This data set is analyzed and compared with the common detection data. Third, deep learning detection methods are employed to detect impact craters. To address the problem attributed to the resolution of Chang’e data being too high, a quadtree decomposition is conducted. Lastly, a geographic information system is used to map the DEM data to 3D space and annotate the semantic information of the impact craters. In brief, a 3D model of the Moon with crater information is implemented based on Chang’e data in the Chang‘e reference frame, which is of high significance.

Reconstructing 3D structures from non-overlapping multi-view images is a crucial task in the field of 3D computer vision, since it is difficult to establish feature correspondences and infer depth from overlapping parts of views. Previous methods, whether generating the surface mesh or volume of an object, face challenges in simultaneously ensuring the accuracy of detailed topology and the integrity of the overall structure. In this paper, we introduce a novel coarse-to-fine Transformer-based reconstruction network to generate precise point clouds from multiple input images at sparse and non-overlapping viewpoints. Specifically, we firstly employ a general point cloud generation architecture enhanced by the concept of adaptive centroid constraint for the coarse point cloud corresponding to the object. Subsequently, a Transformer-based refinement module applies deformation to each point. We design an attention-based encoder to encode both image projection features and point cloud geometric features, along with a decoder to calculate deformation residuals. Experiments on ShapeNet demonstrate that our proposed method outperforms other competing methods.

Existing 3-D object extraction methods on terrestrial laser point clouds are further developed through filtering and labeling. However, such predefined features are heuristically designed to process generic object point clouds. Thus, existing abilities are insufficient to handle specific rock-mass point clouds. Given the complexity and diversity of terrestrial environments, the effective removal of vegetation points from rock-mass point clouds is particularly challenging. To address such problems, this study presents a novel approach for 3-D rock-mass point clouds labeling by using convolutional feature learning based on distribution priors with multiple scales and views. First, to extract discriminative features of each point for classification, we propose novel multiview supporting planes to analyze the spatial distribution and structure of its neighboring points for each category. Second, we define the multiscale spatial distribution matrix on a grid representation (e.g., the number of points projected into each cell). Last, the statistical information of points is nonlinearly combined and hierarchically compressed to generate a compact and effective convolutional feature representation for classification. The effectiveness of the proposed method is evaluated via experiments on rock-mass point clouds from different scenes. Compared with existing extraction approaches, experimental results indicate the superiority of the proposed method in terms of the precision and recall.