Search Results (965)

Gain and bias non-uniformities in infrared line-scanning detectors often result in horizontal streak noise, degrading image quality. This paper introduces a novel non-uniformity correction algorithm combining residual guidance and adaptive weighting, which achieves superior denoising and detail preservation compared to existing methods. The method combines residual and original images in a dual-guidance mechanism and significantly enhances denoising performance and detail preservation through iterative compensation strategies and locally weighted linear regression. Additionally, the algorithm employs local variance to adjust weights dynamically, achieving efficient correction in complex scenes while reducing computational complexity to meet real-time application requirements. Experimental results on both simulated and real infrared datasets demonstrate that the proposed method outperforms mainstream algorithms regarding peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) metrics, achieving an optimal balance between detail preservation and noise suppression. The algorithm demonstrates robust performance in complex scenes, making it suitable for real-time applications in high-resolution infrared imaging systems. Full article

Edge-assisted visual simultaneous localization and mapping (SLAM) is widely used in autonomous driving, robot navigation, and augmented reality for environmental perception, map construction, and real-time positioning. However, it poses significant privacy risks, as input images may contain sensitive information, and generated 3D point clouds can reconstruct original scenes. To address these concerns, this paper proposes a dual-component privacy-preserving approach for visual SLAM. First, a privacy protection method for images is proposed, which combines object detection and image inpainting to protect privacy-sensitive information in images. Second, an encryption algorithm is introduced to convert 3D point cloud data into a 3D line cloud through dimensionality enhancement. Integrated with ORB-SLAM3, the proposed method is evaluated on the Oxford Robotcar and KITTI datasets. Results demonstrate that it effectively safeguards privacy-sensitive information while ORB-SLAM3 maintains accurate pose estimation in dynamic outdoor scenes. Furthermore, the encrypted line cloud prevents unauthorized attacks on recovering the original point cloud. This approach enhances privacy protection in visual SLAM and is expected to expand its potential applications. Full article

Recently, with the advantages of easy deployment, flexibility, diverse functions, and low cost, flying ad hoc network (FANET) has captured great attention for its huge potential in military and civilian applications, whereas the high-speed movement and limited node energy of unmanned aerial vehicles (UAVs) leads to high dynamic topology and high packet loss rate in FANET. Thus, we introduce the software-defined networking (SDN) architecture into FANET and investigate routing scheme in an SDN-FANET to harvest the advantages of SDN centralized control. Firstly, a FANET segmented routing scheme based on the hybrid SDN architecture is proposed, where inter-segment conducts energy-balanced routing and intra-segment adopts three-dimensional (3D) greedy perimeter stateless routing (GPSR). Specifically, we design the specific process of message interaction between SDN controller and UAV nodes to ensure the execution of the inter-segment routing based on energy balance. Further, to reduce the packet loss rate in high-speed motion scenes, an adaptive extended Kalman prediction algorithm is also proposed to track and predict the 3D movement of UAVs. Simulations verify the effectiveness of the proposed routing scheme in terms of end-to-end delay and packet delivery ratio. Full article

Predicting the uncertain distribution of underwater acoustic fields, influenced by dynamic oceanic parameters, is critical for acoustic applications that rely on sound field characteristics to generate predictions. Traditional methods, such as the Monte Carlo method, are computationally intensive and thus unsuitable for applications requiring high real-time performance and flexibility. Current machine learning methods excel at improving computational efficiency but face limitations in predictive performance, especially in shadow areas. In response, a machine learning method is proposed in this paper that balances accuracy and efficiency for predicting uncertainties in deep ocean acoustics by decoupling the scene representation into two components: (a) a local radiance model related to environmental factors, and (b) a global representation of the overall scene context. Specifically, the internal relationships within the local radiance are first exploited, aiming to capture fine-grained details within the acoustic field. Subsequently, local clues are combined with receiver location information for joint learning. To verify the effectiveness of the proposed approach, a dataset of historical oceanographic data has been compiled. Extensive experiments validate the efficiency compared to traditional Monte Carlo techniques and the superior accuracy compared to existing learning method. Full article

Pneumonia is a respiratory infection that affects the lungs. The symptoms of viral and bacterial pneumonia are similar. In order to improve automatic detection efficiency regarding X-ray images of pneumonia, this paper, we propose a novel pneumonia detection method based on the Fast-YOLO network model. First, we re-annotated the open-source dataset of MIMIC Chest X-ray pneumonia, enhancing the model’s adaptability to complex scenes by incorporating Mixup, Mosaic, and Copy–Paste augmentation methods. Additionally, CutMix and Random Erasing were introduced to increase data diversity. Next, we developed a lightweight FASPA Fast Pyramid Attention Mechanism and designed the Fast-YOLO network based on this mechanism to effectively address the complex features in pneumonia X-ray images, such as low contrast and an uneven distribution of local lesions. The Fast-YOLO network improves upon the YOLOv11 architecture by replacing the C3k2 module with the FASPA attention mechanism, significantly reducing the network’s parameter count while maintaining detection performance. Furthermore, the Fast-YOLO network enhances feature extraction capabilities when handling scenes with geometric deformations, multi-scale features, and dynamic changes. It expands the receptive field, thereby balancing computational efficiency and accuracy. Finally, the experimental results demonstrate that the Fast-YOLO network, compared to traditional convolutional neural network methods, can effectively identify pneumonia regions and localize lesions in pneumonia X-ray image detection tasks, achieving significant improvements in FPS, precision, recall, mAP @0.5, and mAP @0.5:0.95. This confirms that Fast-YOLO strikes a balance between computational efficiency and accuracy. The network’s excellent generalization capability across different datasets has been validated, showing the potential to accelerate the pneumonia diagnostic process for clinicians and enhance diagnostic accuracy. Full article

Accurately detecting apple fruit can crucially assist in estimating the fruit yield in apple orchards in complex scenarios. In such environments, the factors of density, leaf occlusion, and fruit overlap can affect the detection and counting accuracy. This paper proposes an improved YOLOv7-Tiny-PDE network model based on the YOLOv7-Tiny model to detect and count apples from data collected by drones, considering various occlusion and lighting conditions. First, within the backbone network, we replaced the simplified efficient layer aggregation network (ELAN) with partial convolution (PConv), reducing the network parameters and computational redundancy while maintaining the detection accuracy. Second, in the neck network, we used a dynamic detection head to replace the original detection head, effectively suppressing the background interference and capturing the background information more comprehensively, thus enhancing the detection accuracy for occluded targets and improving the fruit feature extraction. To further optimize the model, we replaced the boundary box loss function from CIOU to EIOU. For fruit counting across video frames in complex occlusion scenes, we integrated the improved model with the DeepSort tracking algorithm based on Kalman filtering and motion trajectory prediction with a cascading matching algorithm. According to experimental results, compared with the baseline YOLOv7-Tiny, the improved model reduced the total parameters by 22.2% and computation complexity by 18.3%. Additionally, in data testing, the p-value improved by 0.5%; the R-value rose by 2.7%; the mAP and F1 scores rose by 4% and 1.7%, respectively; and the MOTA value improved by 2%. The improved model is more lightweight and can preserve a high detection accuracy well, and hence, it can be applied to detection and counting tasks in complex orchards and provides a new solution for fruit yield estimation using lightweight devices. Full article

Unsafe behaviors on construction sites are a major cause of accidents, highlighting the need for effective detection and prevention. Traditional methods like manual inspections and video surveillance often lack real-time performance and comprehensive coverage, making them insufficient for diverse and complex site environments. This paper introduces GeoIoU-SEA-YOLO, an enhanced object detection model integrating the Geometric Intersection over Union (GeoIoU) loss function and Structural-Enhanced Attention (SEA) mechanism to improve accuracy and real-time detection. GeoIoU enhances bounding box regression by considering geometric characteristics, excelling in the detection of small objects, occlusions, and multi-object interactions. SEA combines channel and multi-scale spatial attention, dynamically refining feature map weights to focus on critical features. Experiments show that GeoIoU-SEA-YOLO outperforms YOLOv3, YOLOv5s, YOLOv8s, and SSD, achieving high precision ([email protected] = 0.930), recall, and small object detection in complex scenes, particularly for unsafe behaviors like missing safety helmets, vests, or smoking. Ablation studies confirm the independent and combined contributions of GeoIoU and SEA to performance gains, providing a reliable solution for intelligent safety management on construction sites. Full article

Single-pixel imaging is a computational optical imaging technique that uses a single-pixel detector to obtain scene information and reconstruct the image. Compared with traditional imaging techniques, single-pixel imaging has the advantages of high sensitivity and a wide dynamic range, etc., which make it have broad application prospects in special frequency band imaging and scattering media imaging. This paper mainly introduces the history of development and the characteristics of the single-pixel detector, focuses on the typical applications of single-pixel imaging in coded aperture, transverse scanning, and longitudinal scanning systems, and gives an account of the application of deep learning technology in single-pixel imaging. At the end of this paper, the development of single-pixel imaging is summarized and future trends forecasted. Full article

The carbon stock in terrestrial ecosystems is closely linked to changes in land use. Understanding how land use alterations affect regional carbon stocks is essential for maintaining the carbon balance of ecosystems. This research leverages land use and driving factor data spanning from 2000 to 2020, utilizing the Patch-generating Land Use Simulation (PLUS) model alongside the InVEST ecosystem services model to examine the temporal and spatial changes in carbon storage across Beijing. Additionally, four future scenes for 2030—urban development, natural development, cropland protection, as well as eco-protection—are explored, with the PLUS and InVEST models employed to emulate dynamic land use changes and the corresponding carbon stock variations. The results show that the following: (1) Between 2000 and 2020, changes in land use resulted in a significant decline in carbon storage, with a total reduction of 1.04 × 10⁷ tons. (2) From 2000 to 2020, agricultural, forest, and grassland areas in Beijing all declined to varying extents, while built-up land expanded by 1292.04 km² (7.88%), with minimal changes observed in water bodies or barren lands. (3) Compared to the carbon storage distribution in 2020, carbon storage in the 2030 urban development scenario decreased by 6.99 × 10⁶ tons, highlighting the impact of rapid urbanization and the expansion of built-up areas on the decline in carbon storage. (4) In the ecological protection scenario, the optimization of land use structure resulted in an increase of 6.01 × 10⁵ tons in carbon storage, indicating that the land use allocation in this scenario contributes to the restoration of carbon storage and enhances the carbon sink capacity of the urban ecosystem. This study provides valuable insights for policymakers in optimizing ecosystem carbon storage from a land use perspective and offers essential guidance for the achievement of the “dual carbon” strategic objectives. Full article

Cooperative visual tracking involving unmanned aerial vehicles (UAVs) in low-altitude environments is a dynamic and rapidly evolving domain. Existing models encounter challenges with targets, such as scale variation, appearance similarity, and frequent occlusions, which hinder the effective use of target information for cross-view identity association. To address these challenges, this study introduces a model for multi-view, multi-target tracking in low-altitude scenes involving UAVs (MVTL-UAV), an effective multi-target tracking model specifically designed for low-altitude scenarios involving UAVs. The proposed method is built upon existing end-to-end detection and tracking frameworks, introducing three innovative modules: loss reinforcement, coupled constraints, and coefficient improvement. Collectively, these advancements enhance the accuracy of cross-view target identity matching. Our method is trained using the DIVOTrack dataset, which comprises data collected from a single UAV and two handheld cameras. Empirical results indicate that our approach achieves a 2.19% improvement in cross-view matching accuracy (CVMA) and a 1.95% improvement in the cross-view ID F1 metric (CVIDF1) when compared to current state-of-the-art methodologies. Importantly, the model’s performance is improved without compromising computational efficiency, thereby enhancing its practical value in resource-constrained environments. As a result, our model demonstrates satisfactory performance in various low-altitude target tracking scenarios involving UAVs, establishing a new benchmark in this research area. Full article

Multi-object tracking (MOT) is an important task in computer vision, particularly in complex, dynamic environments with crowded scenes and frequent occlusions. Traditional tracking methods often suffer from identity switches (IDSws) and fragmented tracks (FMs), which limits their ability to maintain consistent object trajectories. In this paper, we present a novel framework, called ReTrackVLM, that integrates multimodal embedding from a visual language model (VLM) with a zero-shot re-identification (ReID) module to enhance tracking accuracy and robustness. ReTrackVLM leverages the rich semantic information from VLMs to distinguish objects more effectively, even under challenging conditions, while the zero-shot ReID mechanism enables robust identity matching without additional training. The system also includes a motion prediction module, powered by Kalman filtering, to handle object occlusions and abrupt movements. We evaluated ReTrackVLM on several widely used MOT benchmarks, including MOT15, MOT16, MOT17, MOT20, and DanceTrack. Our approach achieves state-of-the-art results, with improvements of 1.5% MOTA and a reduction of 10. 3% in IDSws compared to existing methods. ReTrackVLM also excels in tracking precision, recording a 91.7% precision on MOT17. However, in extremely dense scenes, the framework faces challenges with slight increases in IDSws. Despite the computational overhead of using VLMs, ReTrackVLM demonstrates the ability to track objects effectively in diverse scenarios. Full article

Visual SLAM is widely applied in robotics and remote sensing. The fusion of Gaussian radiance fields and Visual SLAM has demonstrated astonishing efficacy in constructing high-quality dense maps. While existing methods perform well in static scenes, they are prone to the influence of dynamic objects in real-world dynamic environments, thus making robust tracking and mapping challenging. We introduce DyGS-SLAM, a Visual SLAM system that employs dual constraints to achieve high-fidelity static map reconstruction in dynamic environments. We extract ORB features within the scene, and use open-world semantic segmentation models and multi-view geometry to construct dual constraints, forming a zero-shot dynamic information elimination module while recovering backgrounds occluded by dynamic objects. Furthermore, we select high-quality keyframes and use them for loop closure detection and global optimization, constructing a foundational Gaussian map through a set of determined point clouds and poses and integrating repaired frames for rendering new viewpoints and optimizing 3D scenes. Experimental results on the TUM RGB-D, Bonn, and Replica datasets, as well as real scenes, demonstrate that our method has excellent localization accuracy and mapping quality in dynamic scenes. Full article

Array Geiger-mode avalanche photodiode (GM-APD) Light Detection and Ranging (LiDAR) has the advantages of high sensitivity and long imaging range. However, due to its operating principle, GM-APD LiDAR requires processing based on multiple-laser-pulse data to complete the target reconstruction. Therefore, the influence of the device’s movement or scanning motion during GM-APD LiDAR imaging cannot be ignored. To solve this problem, we designed a reconstruction method based on coordinate system transformation and the Position and Orientation System (POS). The position, attitude, and scanning angles provided by POS and angular encoders are used to reduce or eliminate the dynamic effects in multiple-laser-pulse detection. Then, an optimization equation is constructed based on the negative-binomial distribution detection model of GM-APD. The spatial distribution of photons in the scene is ultimately computed. This method avoids the need for field-of-view registration, improves data utilization, and reduces the complexity of the algorithm while eliminating the effect of LiDAR motion. Moreover, with sufficient data acquisition, this method can achieve super-resolution reconstruction. Finally, numerical simulations and imaging experiments verify the effectiveness of the proposed method. For a 1.95 km building scene with SBR ~0.137, the 2 × 2-fold super-resolution reconstruction results obtained by this method reduce the distance error by an order of magnitude compared to traditional methods. Full article

Event-based cameras are a new type of vision sensor in which pixels operate independently and respond asynchronously to changes in brightness with microsecond resolution, instead of providing standard intensity frames. Compared with traditional cameras, event-based cameras have low latency, no motion blur, and high dynamic range (HDR), which provide possibilities for robots to deal with some challenging scenes. We propose a visual–inertial odometry for stereo event-based cameras based on Error-State Kalman Filter (ESKF). The vision module updates the pose by relying on the edge alignment of a semi-dense 3D map to a 2D image, while the IMU module updates the pose using median integration. We evaluate our method on public datasets with general 6-DoF motion (three-axis translation and three-axis rotation) and compare the results against the ground truth. We compared our results with those from other methods, demonstrating the effectiveness of our approach. Full article

Oriented object detection has become a hot topic in SAR image interpretation. Due to the unique imaging mechanism, SAR objects are represented as clusters of scattering points surrounded by coherent speckle noise, leading to blurred outlines and increased false alarms in complex scenes. To address these challenges, we propose a novel noise-to-convex detection paradigm with a hierarchical framework based on the scattering-keypoint-guided diffusion detection transformer (SKG-DDT), which consists of three levels. At the bottom level, the strong-scattering-region generation (SSRG) module constructs the spatial distribution of strong scattering regions via a diffusion model, enabling the direct identification of approximate object regions. At the middle level, the scattering-keypoint feature fusion (SKFF) module dynamically locates scattering keypoints across multiple scales, capturing their spatial and structural relationships with the attention mechanism. Finally, the convex contour prediction (CCP) module at the top level refines the object outline by predicting fine-grained convex contours. Furthermore, we unify the three-level framework into an end-to-end pipeline via a detection transformer. The proposed method was comprehensively evaluated on three public SAR datasets, including HRSID, RSDD-SAR, and SAR-Aircraft-v1.0. The experimental results demonstrate that the proposed method attains an ${\mathrm{AP}}_{50}$ of 86.5%, 92.7%, and 89.2% on these three datasets, respectively, which is an increase of 0.7%, 0.6%, and 1.0% compared to the existing state-of-the-art method. These results indicate that our approach outperforms existing algorithms across multiple object categories and diverse scenes. Full article