ABSTRACT This study presents DFSplat, a feed‐forward 3D Gaussian Splatting model utilizing depth feature fusion for the high‐quality reconstruction of a 3D scene from sparse multiview data and the production of novel view images. DFSplat enhances the robustness of depth prediction and the quality of geometric reconstruction by incorporating a pre‐trained monocular depth estimation module into the multiview feature matching branch, thereby addressing the limitations of current methods for multiview depth estimation in complex scenes. The method employs a content‐guided attention (CGA) module to adaptively integrate monocular depth features with multiview cost‐volume features, addressing the fusion difficulty arising from the disparity in encoding between low‐level and high‐level features. Experiments conducted on the extensive RealEstate10K and ACID data sets demonstrate that DFSplat surpasses current methodologies in PSNR, SSIM, and LPIPS measures, attaining state‐of‐the‐art performance. The innovation integrates the global consistency of monocular depth estimation with the local precision of multiview matching, optimizing 3D Gaussian parameters prediction through an efficient fusion strategy, thereby offering a novel approach for high‐quality scene reconstruction in sparse view scenarios.

Hyperspectral target detection faces challenges due to limited labeled data and high spectral similarity across material samples, making feature extraction difficult. Existing methods for spatial-spectral feature learning are often computationally constrained. To address these issues, we propose a framework combining unsupervised spectral contrastive learning and spatial-spectral knowledge distillation (UCLKD). Our approach uses an unsupervised contrastive learning module to enhance spectral feature representation while preventing model collapse. A spatial-spectral knowledge distillation strategy then leverages features from a hyperspectral Foundation Model to guide the spectral feature extraction network, effectively capturing discriminative spectral features. Extensive experiments demonstrate that UCLKD outperforms existing methods in target detection performance. The code is available at https://github.com/Li-ZK/UCLKD.

With the rapid growth in satellite imagery acquisition and decreasing revisit intervals, efficient on-orbit processing of hyperspectral data has become critical due to limited onboard computing resources. In this context, lifelong learning (LLL) offers a promising solution to enable continuous learning from new data without storing all previous data or retraining from scratch. However, the plasticity-stability dilemma remains a significant challenge, particularly in hyperspectral image (HSI) classification under class-incremental scenarios. To address this, we propose a novel network architecture that integrates contrastive learning and an angular penalty loss. The contrastive learning module facilitates adaptive knowledge fusion, enabling the model to effectively incorporate new information while preserving prior knowledge. The angular penalty loss allows the classifier to dynamically expand for new classes while maintaining discrimination between old and new categories. Together, these components ensure robust knowledge retention, transfer, and adaptability. Experimental results on three benchmark hyperspectral datasets demonstrate that our method significantly outperforms existing approaches, highlighting its efficacy in addressing LLL challenges in HSI classification. The code is available athttps://github.com/Li-ZK/LLL-AFCA.

The Kernel Correlation Filter Tracking Algorithm(KCF)is a lightweight tracking algorithm with the advantages of fast tracking speed and good effect. However, when the target is occluded and the scale changes, the algorithm will have tracking drift and tracking loss. Aiming at the kernel-related filter tracking algorithm that cannot solve the tracking failure caused by occlusion, target scale changes and other factors in the tracking process, an anti-occlusion and scale-adaptive kernel-related filtering algorithm is proposed. We build a scale pool and use the scale pool to train a one-dimensional fast scale filter to solve the problem of target scale changes. This paper uses the average occlusion distance metric and the size of the context occlusion factor to determine the occlusion state of the target, and dynamically select the learning update rate of the target model according to the target occlusion state. When it is judged that the target is severely occluded, the target position is predicted according to the previous motion state of the target, and the small-range re-detection positioning mechanism proposed in this paper is used to re-detect the target within a certain range of the predicted position. At the same time, the re-detected target is occluded again Judgment to determine whether the target is out of the occlusion. If it is determined that the target is still severely occluded, it means that the target is not out of the occlusion area, the re-detection of the target position is inaccurate, and the predicted position is output. Experimental results show that the accuracy and success rate of the algorithm in this paper are 0.819 and 0.669, which are 8.33% and 7.04% higher than the KCF algorithm. The tracking effect of this algorithm is better than that of KCF algorithm.

As an important information fusion method in video surveillance forensics, infrared and visible image fusion has received extensive attention and research. In order to reduce the sampling number of the original video surveillance image information in the fusion process, improve the processing efficiency and the background quality of the fused image, a frequency domain fusion algorithm of infrared and visible image based on compressed sensing for video surveillance forensics is proposed. Firstly, the compressed sensing is used for sampling, and then the subspace pursuit (SP) algorithm is applied to reconstruct the sparse coefficient. Secondly, the sparse coefficient is transformed in frequency domain, and the fusion is completed in frequency domain, and finally obtains the fused image. A large number of experiments show when the number of compressed sensing samples is 70%, the fusion quality is better than the other four comparison algorithms that take advantage of full sampling. Compared with similar algorithms, this algorithm has obvious advantages in the presentation of image background information and gradient structure information.

A lightweight intelligent vehicle system was developed to realize autonomous driving, face recognition, face anti-spoofing, remote control, infrared obstacle avoidance and other functions to improve the security of contactless delivery. In this system, BCM2711 was used as kernel control chip, and it was equipped with deep network learning models such as LaneNet, ResNet and LSTM. It had been proved that this system could realize the above functions and achieve real-time effects, thus gaining great economic value and market space in contactless delivery service.

An improved NS traffic flow model was built in this paper to simulate two safety factors of vehicle-pedestrian avoidance and vehicle-vehicle avoidance under different weather conditions. Then the regulations of changes on lanes and vehicle speed under two-lane conditions were optimized as well as the improved NS model based on cellular automata. Results showed that the improved NS model can predict road conditions effectively, thereby improving the safety of roads.

This paper presents a variable density (sinusoidal) antenna deployment scheme which is designed for mobile relay (MR) system of the high-speed train. By analyzing the large-scale fading under the high-speed railway (HSR) wireless channel environment, the instantaneous channel capacity and the total service amount of several antenna deployments are derived. Theoretical analysis and simulation results indicate that the proposed deployment can provide higher capacity for the HSR relay system. Comparing with several traditional deployments, the proposed deployment utilizes the feature of the HSR wireless environment, provides better coverage to the edge of the base stations (BS). In addition, an antenna selection scheme is proposed based on the sinusoidal deployment.

In recent years, high‐speed railways (HSRs) are being developed rapidly all over the world because of their convenience, safety, comfort, and other advantages. However, the reliability and security of HSR wireless communication systems have faced severe challenges, such as the Doppler effect, complicated wireless channel model, frequent handovers and so on. The International Union of Railways is pushing the evolution of the global system for mobile communications for railway (GSM‐R) internationally. In this study, a high‐speed based adaptive beamforming handover scheme is proposed to improve the handover performance for HSR wireless communication systems. When the high‐speed train enters the overlapping region, the serving evolved NodeB (eNodeB) and target eNodeB are using beamforming with different gain factors to improve the received signal quality. In addition, this scheme can dynamically adjust the handover hysteresis margins of the reference signal receiving power (RSRP) and the reference signal receiving quality (RSRQ) based on the speed and position. Simulation results have demonstrated that the proposed handover scheme can improve the handover performance by increasing the handover trigger probability and success probability effectively.