Multimodal data fusion, e.g., hyperspectral image (HSI) and light detection and ranging (LiDAR) data fusion, plays an important role in object recognition and classification tasks. However, existing methods pay little attention to the specificity of HSI spectral channels and the complementarity of HSI and LiDAR spatial information. In addition, the utilized feature extraction modules tend to consider the feature transmission processes among different modalities independently. Therefore, a new data fusion network named AM ³ Net is proposed for multimodal data classification; it includes three parts. First, an involution operator slides over the input HSI’s spectral channels, which can independently measure the contribution rate of the spectral channel of each pixel to the spectral feature tensor construction. Furthermore, the spatial information of HSI and LiDAR data is integrated and excavated in an adaptively fused, modality-oriented manner. Second, a spectral-spatial mutual-guided module is designed for the feature collaborative transmission among spectral features and spatial information, which can increase the semantic relatedness connection through adaptive, multiscale, and mutual-learning transmission. Finally, the fused spatial-spectral features are embedded into a classification module to obtain the final results, which determines whether to continue updating the network weights. Experimental evaluations on HSI-LiDAR datasets indicate that AM ³ Net possesses a better feature representation ability than the state-of-the-art methods. Additionally, AM ³ Net still maintains considerable performance when its input is replaced with multispectral and synthetic aperture radar data. The result indicates that the proposed data fusion framework is compatible with diversified data types.

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.

With the extensive application of 3D maps, acquiring high-quality images with unmanned aerial vehicles (UAVs) for precise 3D reconstruction has become a prominent topic of study. In this research, we proposed a coverage path planning method for UAVs to achieve full coverage of a target area and to collect high-resolution images while considering the overlap ratio of the collected images and energy consumption of clustered UAVs. The overlap ratio of the collected image set is guaranteed through a map decomposition method, which can ensure that the reconstruction results will not get affected by model breaking. In consideration of the small battery capacity of common commercial quadrotor UAVs, ray-scan-based area division was adopted to segment the target area, and near-optimized paths in subareas were calculated by a simulated annealing algorithm to find near-optimized paths, which can achieve balanced task assignment for UAV formations and minimum energy consumption for each UAV. The proposed system was validated through a site experiment and achieved a reduction in path length of approximately 12.6% compared to the traditional zigzag path.

Multimodal feature fusion representation, e.g., hyperspectral image and light detection and ranging (HSI-LiDAR) fusion, is an essential topic for fusion perception. However, existing networks tend to employ mandatory feature stacking or local context fusion strategies between multiple modalities, ignoring the power of globally mutual-guided feature transmission. Therefore, this paper develops a mutually beneficial transformer method for multimodal data fusion (MBFormer), which contains the following steps. First, a spatial constraint-based self-attention (SCS) module . In this module, spectralwise attention and a spatialwise convolution are applied to HSI and LiDAR data individually, and then a spatial guide mask generated from LiDAR elevation information is used as an agent to bridge with HSI for spatial feature constraints. Second, a channel diversity-based transformer (CDT) module . On the basis of local spectral embedding explorations, an adaptive token-mixer mechanism is conducted on the groupwise classification token of HSI and individual LiDAR data for global information connectivity and transitivity. At last, the selected features are embedded into a classification layer for the final result calculation. Experimental results show that the proposed MBFormer can obtain 97.76% and 98.62% classification accuracies on Houston and Trento datasets, respectively, indicating the advantages and competitiveness of the MBFormer over the compared state-of-the-art methods.

The $k$ -nearest neighbor ( $k$ -NN) method relies on Euclidean distance as a classification measure to obtain the labels of the test samples. Recently, many studies show that joint region of test samples can make full use of the spatial information of hyperspectral image. However, traditional joint $k$ -NN algorithm holds that the weight of the each test sample in a local region is identical, which is not reasonable, since each test sample may have different importance and distribution. To solve this problem, a weighted joint nearest neighbor and sparse representation method is proposed in this paper, which consists of the following steps: first, a Gaussian weighted function has been introduced into the joint region of test pixels so as to obtain the weighted joint Euclidean distance. Next, the sparse representation-based method is adopted to obtain the representation residuals. Finally, a decision function is applied to achieve the balance between the weighted joint Euclidean distance and residual of the sparse representation. Experiments performed on the four real HSI datasets have demonstrated that the proposed methods can achieve better performance than several previous methods.

Acoustic holographic techniques are crucial in diverse applications, such as three-dimensional holographic display and particle manipulation. However, conventional methods for computer-generated acoustics holography rely heavily on iterative optimization algorithms, which are time-consuming and particularly hinder their capacity of generating a dynamic hologram in real time. Here, a deep learning approach based on U-Net is proposed to rapidly generate an acoustic hologram with optimal amplitude and phase maps. It is demonstrated that, after being trained with adequate data that are numerically synthesized by the pseudo-inverse method, the proposed deep learning approach can generate both amplitude and phase maps for new target images with an improved overall reconstruction quality. Remarkably, after the offline cost is compensated by a lower online cost for the proposed DL approach, the hologram generation speed is significantly accelerated by the proposed deep learning approach as compared with the pseudo-inverse method, especially for complicated or dynamic images. With the hierarchical feature learning capability and the fast online computational speed, the proposed deep learning approach can serve as a smart platform for rapidly generating complete maps of holograms for the sophisticated or dynamical target images, leading to the new possibility of real-time acoustic-hologram-based applications.