Search Results (4,750)

Reliable state identification is essential for intelligent flotation control because recovery, concentrate grade, entrainment, and mineral losses are only partially observable online. This systematic review examines field instrumentation, online analyzers, froth imaging, temporal synchronization, machine-vision methods, multimodal soft sensing, and the engineering requirements that determine whether a predictive model can operate as an industrial sensor. Scopus and Web of Science publications from 2021 to June 2026 were screened using a PRISMA-based protocol. The systematic evidence base includes 98 peer-reviewed technical studies published between 2021 and June 2026, and two PRISMA methodological publications are used to ensure the methodology for presenting the review. Additional methodological and contextual sources cited outside the systematic body of evidence are not included in the number of studies reflected in PRISMA. The evidence shows that machine vision is the most mature non-contact sensing approach, supporting bubble-size measurement, froth-velocity estimation, operating-state recognition, grade prediction, and visual monitoring. Current research is shifting from handcrafted descriptors toward convolutional, transformer, self-supervised, graph-based, temporal, and multimodal models. However, predictive accuracy alone does not demonstrate industrial readiness when camera geometry, illumination, contamination, delay compensation, temporal leakage, domain shift, uncertainty, inference latency, and SCADA/PLC integration are not evaluated. A five-dimensional Sensor Readiness Index is proposed to assess metrological validity, temporal integrity, validation rigor, operational robustness, and automation integration. The review defines the principal requirements for reliable industrial deployment of flotation sensing systems. Full article

This study presents a low-cost assistive reading system based on a distributed architecture, in which an ESP32-CAM performs image acquisition while an Android application executes the image-processing, optical character recognition, and text-to-speech stages. The processing pipeline includes fixed-threshold binarization, morphological dilation, median filtering, Canny edge detection, and projection-based text segmentation. A key contribution is the mechanical sliding rule-frame, designed to maintain horizontal alignment between the camera and the printed text and to reduce perspective-related errors. The system was evaluated through 36 trials conducted with six participants under Low, Medium, and High Lighting conditions. A repeated-measures analysis showed a statistically significant effect of lighting on the Word Recognition Rate (WRR), with significantly lower performance under Low Lighting than under Medium and High Lighting, while no significant difference was found between Medium and High Lighting. Across all trials, the prototype achieved an overall mean WRR of 76.53%, with a median of 78.75% and a maximum observed WRR of 97.50%. Exploratory participant-level correlations between mean WRR, age, and prior reading experience were not statistically significant and were interpreted cautiously because of the small sample size. These findings provide preliminary evidence of the technical feasibility of combining mechanical alignment with a low-cost portable reading architecture for educational accessibility, while highlighting the need for further validation with larger samples and more diverse operating conditions. 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

Markerless video-based motion capture is a fast-developing, low-burden and versatile alternative to marker-based stereophotogrammetry, yet independent validation evidence for several commercial solutions is still scarce. This study validates the markerless software CapturyStudio (The Captury GmbH, Saarbrücken, Germany) for the reconstruction of upper- and lower-limb joint kinematics, comparing it with a validated Xsens (Movella, Henderson, NV, USA) inertial measurement unit system. Ten healthy subjects (five females and five males) performed two standardized clinical tasks, a Reach-To-Grasp gesture for the upper limb and a Timed-Up and Go test for the lower limb, recorded simultaneously with eight BTS SMART EVO-DX 2 (BTS Bioengineering S.p.A., Garbagnate Milanese, Italy) cameras operating in markerless mode and a 17-sensor Xsens system. Elbow, shoulder, knee and hip flexion–extension angles were reconstructed from three-dimensional anatomical keypoints provided by CapturyStudio and compared with the Xsens angles through root mean square error in absolute and percentage terms, range-of-motion accuracy, intraclass correlation coefficient (ICC), Spearman’s coefficient (ρ), Bland–Altman analysis and non-parametric Wilcoxon Rank-Sum tests. CapturyStudio reproduced the temporal pattern of all four angles faithfully (ICC ≥ 0.87; ρ ≥ 0.89), with median discrepancies of about 12° for the elbow and below 8° for shoulder, knee and hip and with the best agreement for the upper limb; the main weakness was a systematic overestimation of hip range of motion. Since both systems are indirect measurement techniques, these values represent the discrepancy between two methods and an upper bound on the error of the markerless system rather than its absolute accuracy. Their magnitude is comparable to the changes regarded as clinically meaningful in goniometric assessment, so the system is presently suited to the analysis of movement patterns rather than to the measurement of absolute joint angles. The study is to be read as a technical comparison of two measurement systems, delimiting the conditions under which future clinical, rehabilitation and sports applications may be pursued. Full article

In planar compliant mechanisms, single-input dual-output (SIDO) displacement amplifiers driven by piezoelectric actuators are frequently used in precision positioning, micro/nano manipulation, and biomedical microdevices. Accurate experimental verification of these mechanisms remains challenging because traditional contact sensors can add excess stiffness and impact the structure’s normal behavior, and single-axis interferometers cannot measure multiple points simultaneously. In contrast, 2D vision-based measurement offers a non-contact alternative capable of capturing full planar motion and synchronized displacement tracking within a single image frame. This paper reviews the literature on camera calibration, homography-based planar reconstruction, sub-pixel edge extraction, vision-based characterization of compliant mechanisms, and benchmarking of vision systems against laser interferometers and coordinate measuring machines (CMMs). In the review, the SIDO-CDAM developed by Ozarkar et al. based on the Instantaneous Center Building Block (IC-BB) approach has been chosen as the target characterization system. The reviewed studies confirm that the major technical components needed for a high-precision 2D vision framework have been independently validated. However, there seems to be a lack of an integrated framework designed for synchronized dual-output SIDO-CDAM characterization. To overcome this gap, a seven-layer 2D vision-based characterization framework is proposed for scalable inspection of prototype-scale and MEMS-scale compliant mechanisms. Full article

Camera-based monitoring systems are increasingly adopted in healthcare settings for the continuous assessment of patient movement and activities. However, their technical performance under real-world indoor conditions remains insufficiently characterised, preventing appropriate selection when choosing cameras for clinical or home adoption and reproducibility. Existing validation studies typically assess either device metrological performance or algorithm accuracy in isolation, and often do not systematically account for practical deployment factors, such as lighting variability, occlusions, and camera positioning. To address this, we present two technical validation protocols that evaluate the same cameras at both the metrological and pose-estimation levels under systematically controlled deployment conditions rarely addressed together in prior work: the first evaluates the metrological performance of RGB and RGBD cameras, and the second assesses their use in supporting human pose estimation, validated using state-of-the-art pose estimators. The proposed protocols systematically assess five cameras (four RGBD and one RGB) under controlled variations in lighting, camera height, viewing angle, and occlusion level, within representative indoor scenarios. The experimental results show that metrological performance varies substantially across cameras, with depth bias at 5 m ranging from ∼10 mm to over 1400 mm depending on the device. For 2D pose estimation, all cameras achieve broadly comparable accuracy (mean mAP between ~78% and ~90%) across cameras and estimators, whereas 3D reconstruction error differs markedly across devices (MPJPE ranging from 104 mm to 365 mm), closely reflecting underlying depth sensing quality. Environmental factors have a camera- and estimator-dependent effect on 3D performance, while camera mounting height has minimal influence within the evaluated range. This work provides evidence-based guidance for the selection and deployment of cameras in healthcare monitoring applications, addressing an important gap in current technical validation practice. Full article

Human Activity Recognition (HAR) plays a significant role in various applications, from learning a discipline to physical rehabilitation. In older adults, activity patterns can indicate levels of frailty, which helps inform the design of physical training programs to prevent falls and maintain mobility. HAR sensing ranges from wearable sensors such as IMUs and RGB cameras to video, specialized gait laboratories, perturbation units, VR, and other modalities. This paper presents a comprehensive study of depth-based, skeleton-driven HAR and age group recognition (AGR) using data collected from real-world nursing home environments. Depth sensors offer a privacy-preserving and non-invasive alternative to wearable and RGB-based systems, enabling continuous 24-h monitoring without requiring user compliance. We systematically evaluate multiple modeling paradigms, including classical machine learning models (DT, RF, KNN, SVM, HMM, HMM+SVM), sequence-based models (LSTM, TCN, ARNN), and graph-based approaches, using skeletal joint data extracted from depth images. Experiments are conducted on two heterogeneous datasets: NTU RGB+D (younger adults) and ETAP-DID (older adults). We analyze the impact of different joint subset configurations (full-body, limb-only, leg-only, and torso-only) and compare raw joint representations with handcrafted time-series features (TSFEL) for frame-based HAR. Beyond activity recognition, we introduce an AGR pipeline to distinguish younger from older adults based on skeletal motion patterns. We investigate multiple feature representations, including absolute joint positions, root-relative coordinates, bone vectors, and joint velocities, and provide interpretability through feature importance and saliency analysis to identify age-discriminative joints and motion cues. Our study provides a comprehensive analysis of various HAR models applied to depth data, examining model performance and the contribution of joint-based features to HAR and AGR. Our study highlights the potential for personalized privacy-preserved monitoring and intervention in nursing homes. 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

With Industry 5.0, human–robot collaboration has become the center of attention. This has introduced new challenges, where workspaces become highly dynamic, leading to safety concerns for robots and especially for humans. This makes it important to have a realistic and accurate digital representation of a production cell and its components in environments for movement planning and remote supervision. This paper presents a digital twin of a smart production cell, synchronizing objects bidirectionally between a real and virtual workspace with minimal effort using only a single RGB-D camera. A fine-tuned YOLO-based detector identifies tools and items in the scene, estimates their spatial position, and spawns them in Unity relative to the robot via coordinate transformation. Experiments with different scanning velocities demonstrate a mean planar spawn deviation of $5.82\mathrm{mm}$ (standard deviation $2.24\mathrm{mm}$) at $0.1\mathrm{m}/\mathrm{s}$ at $0.7\mathrm{m}$ height, while maintaining a constant depth bias of $-0.54\mathrm{mm}$. Once spawned, objects can be manipulated freely via drag-and-drop within the simulation. Upon confirmation, a motion planning module calculates trajectories to execute these changes physically. Across 95 trials and 1805 object placements, the system achieves 100% success within working bounds, successfully executing complex tasks such as repositioning objects and stacking them into pyramid structures. The presented system provides a framework to see, spawn, and synchronize industrial workspaces, enabling rapid setup and safe remote supervision of smart production cells in highly dynamic industrial environments. Full article

With the expansion of broiler production and the transition toward intelligent and labor-saving management, conventional manual inspection is limited by high labor intensity, unintuitive spatial representation of abnormalities, and inefficient on-site verification. To address these limitations, this study developed an abnormality monitoring system for cage-reared broiler houses by integrating an inspection robot, a digital twin environment, and cloud-based data services. A parameterized three-dimensional model and semantic cage anchors were established according to the dimensions of the physical broiler house and the cage arrangement rules. Robot simultaneous localization and mapping (SLAM) poses, inspection aisles, camera identifiers, and cage arrangement parameters were combined to calculate the semantic locations of dead-bird events and map them within the digital twin environment. Open-mouth breathing, infrared abnormalities, and acoustic abnormalities were additionally visualized at the candidate-cage, local-region, or inspection-aisle level according to the completeness of the available localization information. The system also enabled virtual–physical synchronization of the robot’s position, orientation, and operating status, as well as remote interactive control through a WebGL-based interface. Field tests conducted over approximately 100 days showed that model optimization reduced the triangle count, vertex count, and file size by 48.34%, 30.47%, and 44.81%, respectively, while shortening the initial WebGL scene loading time from 3.84 to 3.05 s. When 500 abnormality markers were displayed simultaneously, the optimized scene maintained an average frame rate of 67.83 fps. Among 2035 dead-bird events, 1954 were correctly localized in terms of cage row, tier, and group, yielding a cage-level localization accuracy of 96.0%. A total of 6412 robot control-command records were evaluated, achieving an overall execution success rate of 99.50%, with mean feedback times ranging from 1.0 to 1.2 s. These results demonstrate that the proposed system provides an integrated workflow for abnormality event acquisition, cage-level localization, three-dimensional visualization, and inspection robot management. Full article

Accurate cloud segmentation from ground-based fisheye camera imagery is essential for solar irradiance forecasting and photovoltaic system optimization. Traditional computer vision approaches, such as HSV thresholding and K-means clustering, face significant limitations when applied globally to sky images due to the spectral similarity between cloud regions and sky areas under varying atmospheric conditions. This study presents a hybrid methodology that leverages YOLOv8 instance segmentation to provide contextual cloud regions followed by refined HSV thresholding within these detected areas. The approach incorporates solar trajectory modeling using pvlib for accurate sun disk detection and exclusion, preventing false cloud classification. The methodology was developed and validated at the CIESOL using Mobotix Q71 fisheye cameras, and later tested in Antofagasta (Chile) and Huelva (Spain). The YOLOv8l-seg model achieved a mask precision of 0.821 and box mAP@0.5 of 0.680 on validation data. The results show a promising correlation with radiometric measurements such as clearness index $k_t$ and diffuse fraction $k_d$ in preliminary validation cases. While YOLOv8 demonstrates good cross-site generalization, HSV thresholding requires camera-specific calibration for optimal performance. The method addresses the context-dependency limitations of traditional algorithms, though computational performance and broader validation remain areas for future work. Full article

Reliable unmanned surface vehicle (USV) recovery requires near-field maritime remote sensing outputs that remain stable during the final approach. Planar fiducial geometry provides metric pose estimates, but its depth channel is sensitive to corner-localization noise, apparent marker shrinkage, glare, reflection, and vessel vibration. We present VARE (Visual-Adaptive Ranging and Estimation), a geometry-anchored perception pipeline that combines ArUco-based Perspective-n-Point pose recovery, dual-path bearing fusion, MiDaS-assisted range stabilization, depth-consistency confidence weighting, and innovation-adaptive temporal filtering. VARE is a system-level integration rather than a new neural architecture, PnP solver, or end-to-end docking controller. The pipeline explicitly separates image-centroid bearing, translation-vector bearing, marker-normal heading, camera-frame horizontal approach range, and lateral offset. Independent RTK-synchronized external references, with measured lever-arm corrections between the RTK antenna, camera optical center, and marker reference point, are used for pool and near-shore evaluation. In controlled land tests, VARE reduced independent-reference angular RMSE by 34.0–49.3% relative to the pixel-only baseline and by 22.0–36.4% relative to a static-filter PnP variant. Across 15 pool-based approach trials, the full vision-only configuration achieved a horizontal bearing RMSE of 0.46 degrees, a range MAE of 0.82 m, and a range RMSE of 0.90 m. Relative to the matched IPPE-square geometry baseline with One-Euro filtering, the corresponding descriptive reductions were 14.8%, 4.7%, and 5.3%; the modest range differences are not presented as universally significant. Trial-level summaries, confidence intervals, and data-availability provisions are added to support reproducibility. The results support VARE as a candidate perception module for RTK-referenced USV recovery guidance, while full six-degree-of-freedom validation, session-level dropout survival, and closed-loop capture success remain future work. Full article

This single-site industrial case study compares a leased diesel shunting locomotive with a battery-electric shunting locomotive used for the same class of in-plant railway tasks at the Seza Cement Plant. The evidence base comprises plant leasing and fuel records, equipment specifications, site-reported electricity indicators, operator-reported operational observations, direct CO₂ calculations, and documented occupational safety and health (OSH) functions; it is not a controlled or statistically replicated time–motion experiment. The diesel system incurred a monthly lease cost of USD 10,000 and consumed approximately 1800 L/month, equivalent to 21,600 L/year. Cross-checking the direct CO₂ calculation with 2.692 and 2.683 kg CO₂/L factors gives 58.1 and 58.0 t CO₂/year, respectively. The approximately 24-month payback is treated as a plant-reported investment indicator and evaluated through a normalized sensitivity model because disaggregated costs for locomotive purchase, charging infrastructure, battery replacement, and historical maintenance are not available in the case-study dataset. Operational evidence is reported descriptively: the 20–40% reduction in task time is an operator-reported range rather than a statistical mean; the 7–9 min value refers to the complete 10-wagon weighing maneuver; and 25 loaded wagons (approximately 1450 t) represents the maximum documented field movement rather than a manufacturer-rated capacity. A force-balance check shows that this maximum movement is feasible only if total equivalent resistance remains below approximately 5.41 N/kN, using the 77 kN catalog tractive effort as an upper bound. The battery-electric locomotive produces no local exhaust emissions at the point of use and incorporates SIL 2 remote-control functions, a deadman function, emergency-stop controls, camera support, lighting, and warning systems; these features indicate risk-control capability but do not constitute a measured accident-rate reduction. The study therefore contributes facility-scale, evidence-bounded information for low-speed, repetitive industrial shunting within a defined operating area rather than a general proof of battery-electric superiority across railway applications. Full article

Conventional ultrasound phantoms typically lack anatomical surface geometry and procedural access constraints, limiting the transferability of acquired skills to clinical practice. The objective of the present work was to develop and describe such a platform for ultrasound-guided thyroid nodule puncture training, including its construction and initial ultrasound appearance. We present a modular, body-mimicking ultrasound phantom platform comprising three components: an anatomically shaped epoxy composite chassis cast from a healthy volunteer and covering the cervical and upper thoracic region, interchangeable gelatin-based inserts representing thyroid (including puncture target lesions) and surrounding tissue structures, and an integrated dual-camera optical feedback system for real-time and post-procedural needle trajectory visualization. Two chassis configurations reflecting different chin and shoulder positions allow deliberate modulation of procedural difficulty. Insert composition can be varied to simulate tissues of differing echogenicity and density, including liquid-filled targets. Under appropriate storage and disinfection conditions, inserts remained usable for up to four weeks in qualitative observation. The optical feedback system supports self-directed learning and structured debriefing. In combination with magnet-based ultrasound needle guidance technology, the platform is intended to support a longitudinal, competency-based training concept with quantifiable performance metrics, enabling systematic documentation of individual learning curves. The presented system is designed to more closely replicate the anatomical and procedural complexity of clinical ultrasound-guided interventions than conventional phantoms and represents a flexible simulation platform for interventional ultrasound education. Full article

Autonomous vehicles depend on heterogeneous sensing systems whose performance varies with range, illumination, weather, object material, traffic geometry, contamination, calibration quality, and cyber-physical interference. This PRISMA 2020 and PRISMA-S systematic review synthesized 65 peer-reviewed primary studies selected from 2143 records identified through four databases. After removal of 793 records before screening, 1350 titles and abstracts were screened; 273 full texts were assessed and 208 were excluded with documented reasons. The final evidence base covers cameras, LiDAR, radar, thermal and event cameras, GNSS/IMU localization, calibration, synchronization, multimodal fusion, adverse-weather perception, sensor-health monitoring, and fault-tolerant perception. No modality was universally superior: comparative performance depended on hardware generation, dataset, environmental severity, range, and metric. Direct evidence was strongest for component-level perception and controlled degradation, whereas health-conditioned fusion, ODD restriction, and minimum-risk behavior were supported mainly by partial experimental evidence and safety-oriented synthesis. The review therefore proposes, rather than claims to validate, a reliability-aware architecture that separates sensor health from task confidence, preserves uncertainty and provenance, adapts fusion, and constrains operation when residual evidence is insufficient. The review was retrospectively registered in PROSPERO on 30 July 2026 (CRD420261465869). Full article