Search Results (656)

This study developed a simple and scalable e-scooter riding evaluation platform that reduces on-site implementation effort by relying primarily on image analysis. The platform supports repeated riding trials under consistent conditions in an indoor environment. By combining the areas covered by two drones, the system recorded an entire long and narrow indoor riding lane. A pretrained YOLO model was fine-tuned to construct a rider detection model adapted to the experimental environment. ORB-based image registration and trajectory integration then transformed the riding trajectories obtained from the two cameras into a common coordinate system. Riding speed, riding duration, and the radius of curvature of the two curves were calculated. The results revealed differences among subjects in speed variation, stability across riding trials, and turning characteristics, including stable low-speed riding, sustained high-speed riding, and deceleration before turning. For most subjects, the inter-camera junction discrepancy was within 10 cm, indicating general internal consistency of trajectory integration under the experimental conditions. These results suggest that the proposed system can serve as a video-based platform for quantitatively evaluating observable riding behavior, including riding trajectory and speed characteristics, in an indoor riding lane. Full article

This work is devoted to the study of the influence of mechanical activation of aluminum, as well as its modification with graphite on the combustion processes of solid rocket propellant (SRP). Experiments were conducted to determine the ignition induction period, combustion rate and combustion products. Experimental work on ignition and combustion diagnostics was carried out using a diagnostic system at the Institute of Space Propulsion (ISP) of Nanjing University of Science and Technology. A multifunctional combustion diagnostic system was used to characterize the combustion processes of finished SRP samples. A high-speed camera was used to capture the combustion surface regression for various fuels, which can be used to derive their burning rate. The addition of graphite during the mechanochemical activation (MCA) of aluminum significantly affected particle shape and the state of the oxide film, improving the batch homogeneity and increasing the specific surface area. The optimal graphite concentration was approximately 10%, providing additional activation, while 20% graphite caused recoating of the particles and reduced processing efficiency. At constant pressure, it was found that MCA and the introduction of graphite accelerated composite combustion without disrupting stable combustion, with a relatively uniform front. The highest combustion velocity (3.0 mm s⁻¹) was achieved with 10% graphite due to effective heat transfer and catalysis, while increasing its content to 20% reduced the combustion velocity, indicating an optimum of approximately 10%. Full article

Augmented reality (AR) enhances the real world scene by overlaying virtual information onto it. Vision-based 3D tracking and registration is the key technology for ensuring the fusion of virtual and real content in monocular AR systems. Existing mainstream visual tracking and registration methods are susceptible to illumination variations, motion blur, target occlusion, and dynamic background interference in complex scenarios. They also suffer from low computational efficiency, cumulative pose errors, and insufficient stability, making them difficult to deploy on low power edge devices such as embedded systems and mobile terminals. To address these issues, this paper proposes a lightweight monocular AR 3D tracking and registration method that integrates ORB-FREAK features, mismatching outlier filtering, background weighted mean shift, and template-based relocalization. The method does not rely on depth sensors or neural network inference, enabling efficient and accurate lightweight pose estimation. Specifically, we first combine the ORB (Oriented FAST and Rotated BRIEF) descriptor with the FREAK (Fast Retina Keypoint) algorithm for feature detection and initial matching. Hamming distance is used for coarse filtering of mismatched point pairs, and an ascending sort combined with an iterative sequential sampling strategy is applied to solve the optimal homography matrix, significantly improving the accuracy and efficiency of matrix estimation. Then, distance constraints among feature points are imposed on the target registration region to optimize the selection, and camera pose is computed based on the matching between 2D feature points and their corresponding 3D spatial coordinates, eliminating the error accumulation problem of conventional algorithms. Real-time feature matching is further used to correct the optical flow tracking sequence and camera pose, ensuring the continuity of the AR tracking process. Finally, a background weighted mean shift algorithm is introduced to narrow the feature detection range and suppress background interference, complemented by a template-matching relocalization module and a dynamic model update strategy, which effectively enhance the robustness of continuous tracking and registration under complex conditions. Experimental results demonstrate that, in extreme scenarios such as low light conditions, high speed motion, and occlusion, the proposed method achieves AR 3D tracking and registration success rates of 86.7%, 82.3%, and 78.5%, respectively. It exhibits superior performance in pose estimation accuracy and anti-interference capability in complex environments, with significantly reduced computational overhead. Moreover, it can achieve robust and continuous AR 3D tracking and registration on low power edge devices, effectively adapting to demanding AR application scenarios and providing reliable technical support for lightweight AR applications. Full article

Automated wildlife monitoring systems are essential for studying bat populations in natural environments, where nocturnal behavior, high flight speeds, and limited illumination make conventional observation difficult. This paper presents a modular multi-sensor vision system that integrates RGB, near-infrared (NIR), and depth cameras for automated bat monitoring. The proposed architecture consists of one main module and two secondary modules that can be configured into multiple operating modes according to monitoring requirements. The main module operates independently to perform real-time habitat reconstruction using an integrated depth camera or bat detection using a YOLO-based model. When combined with one secondary module, it forms a stereo vision system for three-dimensional localization; when combined with both secondary modules, it generates panoramic images that substantially expand the field of view for monitoring large cave entrances and other complex environments. The proposed modular architecture enables flexible deployment while supporting multiple sensing configurations within a single platform. The modular design provides scalability, geometric consistency through multi-sensor calibration, and flexible deployment, enabling accurate bat detection, habitat reconstruction, and wide-area monitoring within a unified sensing framework. The proposed system provides a versatile and scalable solution for adapting wildlife monitoring to different environmental conditions and observation scenarios. Experimental results demonstrate a detection precision of 0.893, a panoramic field of view of 119°, and real-time processing at 60 fps, validating the effectiveness of the proposed modular architecture. Full article

The latest generation of 4D imaging radar demonstrates significant potential in autonomous driving environmental perception, leveraging its capability to provide target elevation data and dense point clouds. This paper introduces a complete method for constructing a multimodal 4D imaging radar dataset for three-dimensional traffic scenes. It illustrates the hardware and software configurations of the data-acquisition vehicle. Methods including multi-sensor coordination, parameter calibration, timestamp synchronization and spatial datum synchronization are proposed. And eight typical three-dimensional traffic scenarios are designed, such as rainy weather environments, dense heterogeneous targets, enclosed tunnels, high-speed cut-in of multiple vehicles, multi-layered stereoscopic structures and edge working condition reproduction. In addition, this paper puts forward a frame-by-frame processing method for high-resolution images and point cloud data collected by the high-definition camera-LiDAR-4D imaging radar collaborative system. A large model-based 3D annotation method for multiple types of targets is proposed, generating a spatio-temporal sequence-optimized four-dimensional annotation sequence, and finally constructs a complete and high-quality multimodal 4D imaging radar dataset for three-dimensional traffic scenes. The results show that the constructed dataset enables the synchronization of timestamps and spatial coordinate systems. The large model can achieve high-precision 3D annotation for the four predefined target types. The dataset contains 11,400 frames of data from high-definition cameras, LiDAR, and 4D imaging radar, with 131,642 labels. This study will provide reliable fundamental support for the training and verification of 4D imaging radar perception algorithms, vehicle decision-making and planning in complex scenarios, and multi-sensor fusion technologies. Full article

This paper presents a comprehensive wind tunnel investigation on the wind-induced vibration characteristics of a novel four-cable-supported photovoltaic (PV) structure. The proposed structure system integrates two adjacent dual-cable rows through rigid connecting rods to form a collaborative load-bearing framework, aiming to enhance overall stiffness and mitigate wind-induced vibrations. A 1:15-scale aeroelastic model was tested in a boundary-layer wind tunnel for both single-row and five-row configurations. Wind-induced displacements were measured using a non-contact high-definition camera system capable of real-time, multi-point monitoring across multiple rows, while cable tension forces were simultaneously recorded with load cells—a combined measurement approach rarely reported in existing studies. The effects of wind speed and wind direction angle on the vibration responses were systematically examined. Results reveal that vertical vibrations dominate, with mid-span displacements reaching maximum values. The shielding effect among multiple rows is pronounced: the windward first row consistently exhibits the largest displacements and cable forces under both wind pressure and suction. Wind directions of 0° and 180° are identified as the most unfavorable for pressure and suction, respectively. Cable forces under pressure exceed those under suction, and within each row, windward cables sustain greater forces than leeward cables. These findings provide essential experimental reference data for the wind-resistant design of multi-row cable-supported PV support structures. Full article

To investigate the rockburst failure characteristics and underlying mechanisms of deep straight-wall arch roadways containing structural planes, deep-mined limestone was selected as the rock material. True triaxial rockburst experiments were conducted on cubic limestone specimens containing a straight-wall arch roadway. A high-speed camera and an acoustic emission system were employed to monitor, in real-time, the initiation and evolution of the rockburst process. In addition, numerical simulations of straight-wall arch roadways containing structural planes with different spacings were carried out using the PFC software, and the failure patterns and rockburst evolution characteristics of surrounding rock with different structural-plane spacings were systematically analyzed. The results indicate that the presence of structural planes significantly alters the stress and energy transmission paths within the rock mass, leading to local stress concentration, enhanced rockburst impact intensity, and more complex microscopic morphologies of the ejected rock fragments. Compared with specimens without structural planes, specimens containing structural planes exhibited higher cumulative acoustic emission ring-down counts and cumulative absolute energy, accompanied by pronounced transient high-amplitude acoustic emission activity. Moreover, the proportion of shear failure in specimens containing structural planes was higher than that in intact specimens without structural planes. With increasing structural-plane spacing, the failure mode of the surrounding rock gradually changed, while the mutual constraint between the rock mass and the structural planes weakened. As the structural-plane spacing increased, the failure pattern of the surrounding rock changed, and the constraining effect of the rock mass on the structural planes gradually weakened. Consequently, crack propagation paths became increasingly oriented toward the free surface, resulting in a progressive decrease in the propagation angle of wing cracks. Based on the experimental data, a theoretical relationship was established between structural-plane spacing and the stress characteristic parameter of the straight-wall arch roadway, σ_i/σ_max. These findings can provide a useful reference for disaster prevention and mitigation, as well as rockburst prediction, in underground openings containing structural planes under impact disturbance. Full article

This study presents an optical reconstruction method for the quantitative evaluation of arc position in stud welding. Knowledge of the instantaneous arc position is essential for understanding arc dynamics and enabling process monitoring and control. The relative spatial position of the arc is determined from the geometric centroid of an intensity distribution recorded by eight directionally arranged photodiodes. Calculated using area-weighted polygon decomposition, the centroid provides a consistent two-dimensional representation of arc motion and enables the derivation of angular and amplitude-related quantities. The method is not intended to deliver absolute metric arc positions but to provide a reliable relative description of arc movement. Whereas laser-based stereoscopic systems require an additional illumination arrangement and high-speed camera systems rely on the acquisition and processing of image sequences, the proposed method evaluates photodiode signals acquired directly within the welding hardware. This enables a compact, process-integrated implementation suitable for real-time arc monitoring. The proposed reconstruction method is evaluated under conventional direct-current, magnetically influenced direct-current, and alternating-current welding conditions. Validation against stereoscopic high-speed camera measurements showed good agreement in trajectory shape and dynamic behaviour, with mean Pearson correlation coefficients above $0.97$ and a mean Euclidean deviation of 0.59 mm, corresponding to 10.4%, for rotational arc motion. The approach thus enables reliable analysis of arc movement and spatial energy distribution over the entire welding duration. Overall, the proposed methodology provides a robust and efficient tool for quantitative arc motion analysis in stud welding. Full article

Curling requires precise sweeping to control both the speed and trajectory of a stone, making quantitative analysis of sweeping behavior an important challenge for performance evaluation and tactical support. However, existing curling studies have primarily focused on stone motion analysis, while automatic estimation of sweep positions and brush orientations from monocular video remains largely unexplored. To address this problem, this study proposes a vision-based system for estimating the position and orientation of curling brushes during sweeping using a single RGB camera. The proposed framework integrates a two-stage YOLOv8-OBB detection pipeline for identifying sweeping players and brush regions, ByteTrack-based multi-object tracking for temporal association, and a geometric calibration model that transforms image coordinates into real-world curling-sheet coordinates through nonlinear optimization. The system enables frame-by-frame reconstruction of sweeping trajectories and brush orientations without requiring specialized sensing equipment. Experiments were conducted at a curling facility using synchronized monocular video and a Qualisys Miqus M3 motion-capture system as ground truth. The proposed method achieved position estimation errors of 0.03–0.06 m and orientation estimation errors of 4.80°–5.10°, while maintaining a high detection rate of 0.96–0.99 over a measurement range of 15 m. Additional analyses showed that estimation accuracy was influenced by camera placement, sweep style, and the apparent size of the detected brush region. These results demonstrate that accurate sweep-position and orientation estimation can be achieved using only a monocular camera under realistic curling conditions. The proposed system provides a practical framework for objective sweep analysis and has potential applications in technical training, performance assessment, and tactical support in curling. Full article

Surface quality is a critical indicator of product performance, creating a growing demand for real-time surface inspection in industrial manufacturing. However, conventional surface normal measurement techniques require sequential measurements with varying illumination or observation angles, making them unsuitable for high-speed online inspection. To overcome this limitation, this paper proposes a novel imaging technique, termed normalography, for visualizing pixel-wise surface normal distributions. Analogous to thermography, normalography visualizes the spatial distribution of surface normal directions over a material surface. The proposed method targets highly glossy and smooth surfaces and is based on reflectance measurements. A multi-angle collimator was developed to simultaneously illuminate the target surface from multiple incident directions, while multispectral illumination was employed to distinguish the reflected light corresponding to each direction. An imaging system incorporating red, green, and blue illumination sources enables single-shot acquisition of surface normal information over a 1024 × 1024-pixel field of view. The proposed normalography enables camera-like real-time visualization of surface normal distributions without sequential image acquisition, demonstrating its potential for online surface inspection and quality monitoring in industrial manufacturing. Full article

To address the challenges of low image quality and difficult feature extraction of weld beads caused by the complex dynamics of the molten pool, intense arc light, and spatter interference during wire arc additive manufacturing (WAAM) of 316L stainless steel, this paper develops a binocular vision-based dynamic molten pool tracking system and conducts image processing and feature analysis. Two high-speed CMOS cameras are employed to capture images of the molten pool and weld bead. Camera calibration is performed to convert pixel coordinates to world coordinates. The denoising performance of five filtering methods, namely mean, Gaussian, median, maximum, and minimum filters, is systematically compared, and the minimum filter is selected for noise reduction. Adaptive threshold binarization, adapthisteq image enhancement, and morphological threshold segmentation are integrated to effectively separate the weld bead from the background. Four edge detection algorithms—Sobel, Robert, Laplacian, and Canny—are compared, and the Canny algorithm combined with Hough transform line fitting is determined to achieve complete and continuous extraction of the weld bead contour. The Intersection over Union (IoU) metric is introduced for image quality screening. When IoU is set to 0.3, the detection accuracy exceeds 90%, effectively eliminating defective images caused by spatter, explosion, trailing, and other disturbances. The proposed method facilitates stable extraction of geometric parameters (e.g., pixel area of the weld bead and height/width of the molten pool), thereby offering a feasible image-processing solution for dynamic molten-pool monitoring and online quality assessment of 316L stainless steel components fabricated by wire arc additive manufacturing. Full article

Low-latency visual perception is essential for interactive machine vision on edge AI devices, but conventional frame-based image sensors impose frame period delays and generate dense image data that increase memory bandwidth and processing latency. Although Dynamic Vision Sensors (DVSs) are known to provide low latency, sparse output, and high dynamic range, these sensor-level properties do not automatically translate into practical application-level latency reduction on resource-constrained edge platforms. This paper presents a latency-driven sensing algorithm co-design approach for DVS-based low-latency machine vision. The main objective is to connect DVS sensor-level characteristics, event representations, task-dependent processing flows, and measured response times on mobile application processors. We first analyze latency requirements for three representative edge AI applications (i.e., person detection, gesture recognition, and Simultaneous Localization and Mapping (SLAM)), which correspond to different latency regimes and processing structures. We then describe the DVS operating principle, pixel-level event latency, and readout latency, showing how asynchronous event generation reduces sensing delay and suppresses redundant static background information before algorithmic processing. In contrast to prior event camera studies that mainly optimize a single task or a specific event representation, this work evaluates three task-specific event processing systems on mobile processors. Person detection achieves 92 ms processing latency on Exynos 7570, gesture recognition based on event-driven 4-DoF motion estimation achieves 20 ms latency on Exynos 5422, and SLAM achieves 15.9 ms latency on Snapdragon 845. These results satisfy the practical latency targets of the corresponding applications and demonstrate that DVS-based sensing can provide not only sensor-level speed advantages but also system-level latency benefits for AIoT, mobile, robotics, and AR/VR machine vision systems. Full article

Robotic belt grinding is an effective and widely adopted finishing method for superalloys, offering notable advantages such as high material removal capability, low heat input, and reduced workpiece damage. In addition, robots can readily integrate multiple sensors—such as infrared radiation cameras, force sensors, and high-speed cameras—which facilitate real-time monitoring of the grinding process and thereby enhance grinding quality control. With the establishment and continuous advancement of large-scale artificial intelligence (AI) data models, new breakthroughs have emerged in the optimization of robotic grinding processes. Owing to its dexterous workspace and advantages in high flexibility and cost-effectiveness, robotic belt grinding has become a critical process for the precision forming of complex curved components such as aero-engine blades and blisks. However, factors such as the limited absolute accuracy of industrial robots, time-varying grinding contact states, and significant transient boundary effects make it difficult for the current constant-parameter open-loop machining mode to simultaneously meet the demands for high material removal efficiency and high surface integrity on complex profiles. This paper systematically reviews the technologies for precision control and process optimization of robotic belt grinding aimed at pointwise precise material removal. First, the structural composition of the robotic belt grinding system and the material removal mechanism are analyzed. Then, centered on the compensation concept, a hierarchical progressive technical framework is outlined, covering geometric calibration compensation, force/position hybrid online compensation, transient entry boundary compensation, and system-level comprehensive compensation of multi-source errors, with a comparison of the applicable scenarios and the effects on shape and property control at each level. Furthermore, under the support of effective compensation, the collaborative optimization methods of material removal modeling, multi-objective optimization of process parameters, force-constrained trajectory planning, and intelligent adaptive processes are elaborated. Finally, current technical bottlenecks are summarized, and future trends in next-generation adaptive grinding technology driven by digital twins and embodied intelligence are envisioned. This review aims to provide a systematic theoretical reference for the high-precision and intelligent upgrading of robotic precision grinding systems. Full article

Laser-induced acoustic communication is a highly adaptable cross-medium technique that combines the advantages of optical transmission through air and acoustic transmission underwater. However, poor signal stability at high repetition frequencies currently hinders its widespread application. To address this, this paper proposes an innovative scanning focal-point method to enhance stability. Traditional methods such as beam scanning, focus control, and distributed interaction are primarily aimed at enhancing sound pressure in a specific direction, achieving near-field/far-field focusing, or improving the signal-to-noise ratio through coherent synthesis of ultrasonic intensity. In contrast, the method proposed in this paper is intended to avoid the interference of droplets and vapor generated by single-point breakdown under high repetition frequencies, which would otherwise degrade the laser-acoustic conversion efficiency. It is therefore an active defense strategy specifically targeting the stability of laser-induced acoustic communication. First, optical simulation software was used to analyze the effects of surface ripples and bubbles on focal spot displacement and size. Next, a single-pulse experimental system was developed to measure the range and duration of surface depressions caused by optical breakdown. Finally, a scanning focal-point system was constructed for comparative experiments, with results recorded via hydrophones and high-speed cameras. The maximum laser-induced acoustic signal generated by the scanning focal-point method is 7.4 times that produced by single-point breakdown. The experimental results demonstrate that the scanning focal-point method can effectively avoid the influence of water surface disturbance and steam on the optoacoustic conversion efficiency and significantly improve the amplitude and stability of the laser-induced acoustic signal. Full article

Background: Monitoring external load in team sports is essential for performance optimization, injury prevention, and individualized training prescription. Although Local Positioning Systems (LPS) are widely used for indoor athlete tracking, they require wearable devices and specialized infrastructure. Recent advances in artificial intelligence and computer vision allow markerless athlete tracking; however, their validity for basketball remains insufficiently explored. Objective: To evaluate the validity of a deep-learning multi-camera computer-vision system for quantifying external-load variables in basketball compared with a commercial LPS. Methods: The framework integrated fisheye video acquisition, player detection, and pose estimation using YOLOv11x-Pose and player re-identification through ResNet-50 and FAISS similarity search. Positional data were transformed into real-world court coordinates to derive distance, acceleration, deceleration, player load, and average speed metrics. Outputs were compared with measurements obtained from Kinexon LPS. Results: Strong correlations were observed for total distance (r = 0.92), acceleration counts (r = 0.90), deceleration counts (r = 0.92), and player load (r = 0.81), while average speed showed a moderate-to-strong correlation (r = 0.66). ICC and Bland–Altman analyses indicated agreement between systems. Conclusions: The proposed computer-vision system demonstrated high agreement with LPS, supporting its use as a valid, non-invasive, and scalable solution for external load monitoring in basketball. Full article