Explore the vast range of titles available.

Characterization and testing of 3D‐printed robotic compliant systems for lifespan assessment is time‐consuming and costly. For this reason, this work introduces a computer vision approach for automated, non‐invasive monitoring of grippers and evaluation of failures. The vision system first detects colored fiducial markers placed on key points of the gripper. The detection model was trained using synthetic data to ensure robustness to background, illumination, and gripper color variations. Then, the marker positions across frames are used to train and detect anomalies in the gripper's displacement. This is performed by thresholding the reconstructed signal over temporal analysis windows, using the reconstruction error as an anomaly score. Validation was performed on real 3D‐printed grippers under controlled mechanical failures and uncontrolled lighting and background conditions, correctly classifying over 97% of actions corresponding to normal and anomalous gripper performance. The proposed framework offers a scalable and low‐cost alternative to embedded sensors for monitoring gripper performance and detecting early failures. The mechanical failures (low, medium, or severe) of compliant mechanisms include link disconnection and deformation of structural elements, leading to a loss of gripping performance and grasping accuracy compared to normal behavior. Through non‐invasive monitoring, autoencoders, and thresholding techniques, it is possible to identify the type of gripper failure.

A stable welding process is crucial to obtain high quality parts in wire arc additive manufacturing. The complexity of the process makes it inherently unstable, which can cause various defects, resulting in poor geometric accuracy and material properties. This demands for in-process monitoring and control mechanisms to industrialize the technology. In this work, process monitoring algorithms based on welding camera image analysis are presented. A neural network for semantic segmentation of the welding wire is used to monitor the working distance as well as the horizontal position of the wire during welding and classic image processing techniques are applied to capture spatter formation. Using these algorithms, the process stability is evaluated in real time and the analysis results enable the direction independent closed-loop-control of the manufacturing process. This significantly improves geometric fidelity as well as mechanical properties of the fabricated part and allows the automated production of parts with complex deposition paths including weld bead crossings, curvatures and overhang structures.

Contactless structural monitoring has in recent years seen a growing number of applications in civil engineering. Indeed, the elimination of physical installations of sensors is very attractive, especially for structures that might not be easily or safely accessible, yet requiring the experimental evaluation of their conditions, for example following extreme events such as strong earthquakes, explosions, and floods. Among contactless technologies, vision-based monitoring is possibly the solution that has attracted most of the interest of civil engineers, given that the advantages of contactless monitoring can be potentially obtained thorough simple and low-cost consumer-grade instrumentations. The objective of this review article is to provide an introductory discussion of the latest applications of vision-based vibration monitoring of structures and infrastructures through an overview of the results achieved in full-scale field tests, as documented in the published technical literature. In this way, engineers new to vision-based monitoring and stakeholders interested in the possibilities of contactless monitoring in civil engineering could have an outline of up-to-date achievements to support a first evaluation of the feasibility and convenience for future monitoring tasks.

This study examines how novice driving behaviors affect highway flow and collision potential. Driving behaviors of candidates receiving driving training are analyzed for the first time using drone images, in-car footage, and image processing methods. Driving parameters such as standstill distance [CC0], acceleration/deceleration, perception-reaction times, and speeds are extracted using image processing and field observation. These novice driver (ND) parameters are then incorporated into the VISSIM traffic micro-simulation model as a separate driving behavior dataset. The impact of NDs on traffic under different compositions and the resulting crash potential is then assessed. Safety analysis using the Collision Potential Index (CPI) reveal a 35% increase in CPI with only 10% novice drivers, while mobility analysis indicates a 14% average speed decrease with 50% ND traffic composition. Interestingly, a decrease in CPI values is observed when the ND ratio increases to 40% and 50%, which is explained by the more cautious behavior of experienced drivers and a decrease in traffic flow speed. The use of real-world data increases the authenticity and reliability of the study. The findings contribute to the understanding of the risks associated with novice drivers, highlighting the need for effective safety measures. This study provides valuable insights for policy makers and traffic safety experts to reduce the threats posed by inexperienced drivers and regulate the behavior of experienced drivers.

Displacements are among the most important engineering response parameters to be monitored during shake‐table testing, with experiments playing a key role in studying the seismic behavior of structures. However, their accurate measurement is not a trivial task when using contact sensors. Computer vision is an attractive alternative for monitoring absolute and relative displacements, and this study presents a new configuration to fully exploit its potential. The proposed solution combines internal and external video cameras. The former is installed on the roof and points downwards to simultaneously acquire the displacements of targets located throughout the height of the building. The latter was installed outside the shake‐table platen and tracked the roof displacements to provide redundant measures for control and noise compensation. In this way, the movements of the buildings can be reconstructed with high robustness and precision using a limited number of video cameras. The proposed configuration was applied for the first time during shake‐table testing of a full‐scale six‐story building on the outdoor shake table at the University of California, San Diego. The measurements obtained up to strong dynamic inputs showed the capacity of the proposed approach in real‐world environmental conditions and were used for a critical comparison with conventional contact sensors.

Video processing for structural monitoring has attracted much attention in recent years thanks to the possibility of measuring displacement time histories in the absence of stationary points close to the structure, using hardware that is simple to operate and with accessible costs. Experimental studies show a unanimous consensus on the potentialities of vision-based monitoring to provide accurate results that can be equivalent to those obtained from accelerometers and displacement transducers. However, past studies mostly involved steel bridges and footbridges while very few applications can be found for concrete bridges, characterised by a stiffer response with lower displacement magnitudes and different frequency contents of their dynamic behaviour. Accordingly, the attention of this experimental study is focused on the application of a vision-based structural monitoring system to a medium-span, post-tensioned, simply supported concrete bridge, a very common typology in many road networks. The objective is to provide evidence on the quality of the results that could be obtained using vision-based monitoring, understanding the role and influence on the accuracy of the measurements of various parameters relevant to the hardware settings and target geometry, highlighting possible difficulties, and providing practical recommendations to achieve optimal results.

Traditional post-earthquake structural health monitoring (SHM) methods based on dedicated sensors lack scalability due to installation and maintenance demands, leaving most buildings unmonitored. This study investigates the use of existing in-building surveillance cameras to infer structural demand by tracking earthquake-induced building motion. The proposed methodology repurposes ceiling-mounted surveillance cameras to estimate the inter-story drift (IDR) which is directly correlated with structural damage using FEMA guidelines. Shake-table experiments spanning a wide range of excitation intensities and dominant frequencies demonstrate that off-the-shelf surveillance cameras can estimate displacement with accuracy similar to dedicated vision-based SHM setups. To establish operating limits, we quantify how temporal sampling (frame rate) and spatial sampling (video resolution) affect drift estimation accuracy. We also evaluate peak drift/IDR estimation accuracy and peak timing sensitivity under reduced temporal sampling. The results highlight the potential of widely available camera networks as a low-cost, scalable, and rapidly deployable sensing network for post-earthquake assessment.

The composite material layering process has attracted considerable attention due to its production advantages, including high scalability and compatibility with a wide range of raw materials. However, changes in process conditions can lead to degradation in layer quality and non-uniformity, highlighting the need for real-time monitoring to improve overall quality and efficiency. In this study, an AI-based monitoring system was developed to evaluate layer width and assess quality in real time. Three deep learning models Faster Region-based Convolutional Neural Network (R-CNN), You Only Look Once version 8 (YOLOv8), and Single Shot MultiBox Detector (SSD) were compared, and YOLOv8 was ultimately selected for its superior speed, flexibility, and scalability. The selected model was integrated into a user-friendly interface. To verify the reliability of the system, bead width control experiments were conducted, which identified feed speed and extrusion speed as the key process parameters. Accordingly, a Central Composite Design (CCD) experimental plan with 13 conditions was applied to evaluate layer width and validate the system’s reliability. Finally, the proposed system was applied to the additive manufacturing of an aerospace component, where it successfully detected bead width deviations during printing and enabled stable fabrication with a maximum geometric deviation of approximately 6 mm. These findings demonstrate the critical role of real-time monitoring of layer width and quality in improving process stability and final product quality in composite material additive manufacturing.

Background The use of surveillance technologies is becoming increasingly common in inpatient mental health settings, commonly justified as efforts to improve safety and cost-effectiveness. However, their use has been questioned in light of limited research conducted and the sensitivities, ethical concerns and potential harms of surveillance. This systematic review aims to (1) map how surveillance technologies have been employed in inpatient mental health settings, (2) explore how they are experienced by patients, staff and carers and (3) examine evidence regarding their impact. Methods We searched five academic databases (Embase, MEDLINE, PsycInfo, PubMed and Scopus), one grey literature database (HMIC) and two pre-print servers (medRxiv and PsyArXiv) to identify relevant papers published up to 19/09/2024. We also conducted backwards and forwards citation tracking and contacted experts to identify relevant literature. The Mixed Methods Appraisal Tool assessed quality. Data were synthesised narratively. Results Thirty-two studies met the inclusion criteria. They reported on CCTV/video monitoring ( n  = 13), Vision-Based Patient Monitoring and Management ( n  = 9), body-worn cameras ( n  = 6), GPS electronic monitoring ( n  = 2) and wearable sensors ( n  = 2). Sixteen papers (50.0%) were low quality, five (15.6%) medium quality and eleven (34.4%) high quality. Nine studies (28.1%) declared a conflict of interest. Qualitative findings indicate patient, staff and carer views of surveillance technologies are mixed and complex. Quantitative findings regarding the impact of surveillance on outcomes such as self-harm, violence, aggression, care quality and cost-effectiveness were inconsistent or weak. Conclusions There is currently insufficient evidence to suggest that surveillance technologies in inpatient mental health settings are achieving their intended outcomes, such as improving safety and reducing costs. The studies were generally of low methodological quality, lacked lived experience involvement, and a substantial proportion (28.1%) declared conflicts of interest. Further independent coproduced research is needed to more comprehensively evaluate the impact of surveillance technologies in inpatient settings. If they are to be implemented, all key stakeholders should be engaged in the development of policies, procedures and best practice guidance to regulate their use, prioritising patients’ perspectives.

Accurately predicting the material removal rate (MRR) in belt grinding is challenging because of the randomly distributed multiple cutting edges, flexible contact, and continuous wear of the abrasive grains, undermining the ability to achieve the expected machining requirements for belt grinding using the planned parameters. With the development of sensing technology, big data, and intelligent algorithms, online identification methods for material removal through sensing signals have gained traction. A vision-based material removal monitoring method in the belt grinding process was investigated by adopting the gene expression programming (GEP) algorithm. First, the relationship between the grinding parameters and MRR was investigated through a series of experiments. Second, methods of image shooting distance calibration and automatic image segmentation were established. Furthermore, the definition and quantification method of 11 features related to the color, texture, and energy of spark images are described, based on which the features are extracted. Then, the optimal feature subset was determined by analyzing the fluctuation degree and correlation with MRR by computing the coefficient of variation of the features and Pearson’s coefficient of features and MRR, respectively. Finally, a continuous function model including the selected features was obtained using the GEP method. The predicted results and testing time were compared with those of other methods such as LightGBM, convolutional neural network (CNN), support vector regression (SVR), and BP neural network. The results show that the MRR prediction model based on the GEP algorithm can obtain explicit function expressions and is highly effective in predicting accuracy and test time, which is of utmost significance for accurate and efficient acquisition of MRR data online.