Real-time Safety Research Articles

Dynamic identification of crane load fall zone: A computer vision approach

Crane activities pose significant safety risks on construction sites. Thus, numerous studies have applied computer vision techniques to identify crane load fall zones or areas where the crane load could fall during an accident. However, past studies typically assumed that the fall zone directly below the load was static and had a fixed size. In reality, the fall zone is dynamic because its size changes with different crane load types and heights. Thus, this study proposes and validates a vision-based method that dynamically identifies the fall zone directly below the load. The proposed dynamic fall zone detection and tracking method first detects the key points of the load, and then the fall zone is estimated by projecting the area of the lifted load to the construction floor. Once the fall zone is detected and tracked, coupled with a previously established approach for worker detection and tracking, workers entering the fall zone can be automatically detected. Data were collected from a public housing project in Singapore to train and evaluate the proposed method. The proposed method achieved precision and recall of 0.86 and 0.83, respectively. The main contribution of this study is the development of a novel method to detect and track the crane load fall zone. Besides providing real-time safety alerts to crane operators and site supervisors, the proposed method can accumulate statistics on unsafe behaviour and situations to facilitate safety management.

Improved YOLOv4 algorithm for safety management of on-site power system work

To achieve real-time safety management of on-site power system work, the object detection algorithm of the cloud platform is transplanted into the edge computing device. However, due to the limited capacity and arithmetic power of the AI chip carried in the edge computing device, the algorithm needs to be studied in a lightweight manner. Therefore, this paper proposes an improved edge-side lightweight YOLOv4 (You Only Look Once version 4) algorithm, which is compressed and accelerated by improving the network structure and the confidence loss function of the algorithm without losing too much detection accuracy. Firstly, a lightweight convolutional neural network based on depthwise separable convolution and mobile inverted bottleneck convolution is constructed as the backbone network of YOLOv4, and the depthwise separable convolution is used instead of the conventional convolution, which can greatly reduce the parameters number and computation of the algorithm. Secondly, an improved bidirectional feature fusion network is constructed as the neck network of YOLOv4 algorithm to improve the information availability of multi-semantic features and enhance the extraction capability of small-scale features. Then, the confidence loss function of the algorithm is improved to alleviate the problems of uneven distribution of positive samples and negative samples and uneven distribution of easy-to-classify samples and hard-to-classify samples. Finally, the improved YOLOv4 algorithm is tested on an edge computing device equipped with an NVIDIA Jetson Xavier NX chip, and the dataset based on the compliance of on-site dress code in the electric power system field is tested. The experimental results show that, compared with the original YOLOv4 algorithm, the improved YOLOv4 algorithm reduces the parameters number by 93.11%, and the real-time video stream detection speed on the edge computing device is increased by 22.00%, reaching 15.05 FPS. The lightweight improved YOLOv4 algorithm is suitable for deployment at the edge end and can achieve real-time accurate identification of risks in the on-site work.

Time-temperature indicators based on Lipase@Cu3(PO4)2 hybrid nanoflowers

The application of enzyme-inorganic hybrid nanoflowers in time-temperature indicators (TTIs) was studied for the first time. Lipase@Cu3(PO4)2 hybrid nanoflowers were prepared by the interaction between lipase and copper ions. Compared with free lipase, the activity of lipase in the nanoflowers was increased by 137%, and the storage stability was also improved significantly. Based on the enzymatic reaction between lipase and glycerol triacetate, six TTI prototypes with different concentrations of lipase@Cu3(PO4)2 hybrid nanoflowers were investigated. The color of the six TTIs changed from purple to yellow, and finally to green, and the activation energies were 46.86 kJ/mol, 51.17 kJ/mol, 54.99 kJ/mol, 60.00 kJ/mol, 62.17 kJ/mol, and 65.63 kJ/mol, respectively. The shelf life of TTIs can be determined by changing the ratio of lipase@Cu3(PO4)2 hybrid nanoflowers to glycerol triacetate, therefore the TTIs can be easily applied to real-time safety evaluation of many perishable foods.

Deep Learning-Based Automatic Safety Helmet Detection System for Construction Safety

Worker safety at construction sites is a growing concern for many construction industries. Wearing safety helmets can reduce injuries to workers at construction sites, but due to various reasons, safety helmets are not always worn properly. Hence, a computer vision-based automatic safety helmet detection system is extremely important. Many researchers have developed machine and deep learning-based helmet detection systems, but few have focused on helmet detection at construction sites. This paper presents a You Only Look Once (YOLO)-based real-time computer vision-based automatic safety helmet detection system at a construction site. YOLO architecture is high-speed and can process 45 frames per second, making YOLO-based architectures feasible to use in real-time safety helmet detection. A benchmark dataset containing 5000 images of hard hats was used in this study, which was further divided in a ratio of 60:20:20 (%) for training, testing, and validation, respectively. The experimental results showed that the YOLOv5x architecture achieved the best mean average precision (mAP) of 92.44%, thereby showing excellent results in detecting safety helmets even in low-light conditions.

Cross Layer based Congestion Free Route Selection in Vehicular Ad Hoc Networks

For creating a mobile network, the moving cars consider as nodes in the Vehicular Ad-Hoc Networks (VANETs). Each participating car is turned into a wireless router in the VANETs that allows the connecting and creating a network. To improve the comfort and safety of driving of automotive users, the vehicular environment system develops in the vehicular environment systems using the wireless access. The channel congestion causes the degradation of quality of service in such cases with higher vehicle density. The real-time and reliable communication is required for various safety applications of VANETs. The dense traffic network has included one of the major challenges as avoiding the communication channels’ degradation. To provide the network with efficient operation, most of the studies are recommended to use the appropriate congestion control methods. It’s important to note that many congestion control mechanisms are not implemented for event-driven real-time safety messages. Based on the congestion probability approach estimation, CFRS-CP-Congestion free route selection is introduced for minimizing the total number of data flow packets that passing through the congested nodes. At each node, the congestion probability is estimated using the proposed technique of CFRS-CP based on link quality, MAC overhead, neighbour density & vehicle velocity. Then, the estimated congestion probability is used for route assessment. The estimated probability value is appended to the control packets for comparison. All the available routes are assessed based on the estimated congestion probability which results in congestion free routing path for every round of data communication. The simulation results prove that the proposed method decreases end to end delay by 32% and improves PDR up to 30% and throughput up to 45% compared to the existing protocols.

Design of Highway Icing Warning System

This design discloses a real-time safety device for warning road surface icing, including solar ground lamp (1) and solar LED display (2).This design adopts the facility device composed of solar ground lamp and solar LED display screen. The solar ground lamp is equipped with solar photovoltaic cell, charging and discharging battery, red and green LED lamp, temperature and humidity sensor and D data processing and transmission chip circuit;The solar LED display screen is equipped with solar photovoltaic cell, charging and discharging battery, LED graphic display screen, P data processing transmission chip circuit and red flashing warning light;According to the technical scheme of sending out red light and graphic warning according to the freezing critical data collected from the ground, and transmitting the data to the traffic control center through the Internet, it overcomes the problems and shortcomings of the existing technology that it is difficult to issue the pavement freezing warning in real time, so that the expressway can achieve the purpose of real-time and real-time pavement freezing warning.

Real-Time Safety Decision-Making Method for Multirotor Flight Strategies Based on TOPSIS Model

Multirotors play an important role in electric power inspection, border control, modern agriculture, forest fire fighting, flood control, disaster prevention, etc. Multirotor failures, such as a communication fault, a sensor failure, or a power system anomaly, may well lead to mission interruption, multirotor crashes, and even casualties. To ensure flight safety, a multirotor decision module should be established to prevent or reduce the adverse effects caused by failure. Therefore, this paper proposes a real-time safety decision-making method for multirotor flight strategies based on the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS). Firstly, the flight of the multirotor was simulated based on the Rflysim UAV flight simulation platform, and a fault-injection module was constructed to simulate different types of faults, so as to realize real-time monitoring of the flight status of the multirotor, and to collect flight data under various faults to establish condition assessment information sources. Then, based on the random forest algorithm, a failure level classification model of the multirotor was constructed, the model was trained and verified by inputting flight data of three types of safety level failures, and the model effectively classified the failure levels of the multirotor. Under this framework, a real-time safety decision-making model for the multirotor based on the TOPSIS model was constructed to realize the flight safety decision-making of the multirotor under different faults. This method can effectively realize the real-time decision-making for the flight strategy of a multirotor. By comparison with other models, the classification accuracy of the failure level classification model is higher, and the consideration of flight decision-making is more comprehensive and accurate, thus effectively ensuring the flight safety of the multirotor.

Real Time Safety Assessment Based on Hira for Portable and Overhanging Lifting Machine- Construction Site

Cranes are an essential machine for many industries. Due to the fact that it is so large, and it moves such heavy gadget, it can also be very dangerous. Most cranes manage to move at the base in order to put the load where it needs to go. When this occurs, anyone who is in the area surrounding the crane could get trapped in a pinch point and get crushed. The crane driver typically does not have visibility to the area directly on every side the machine, which produces this safety hazard even more serious. Unfortunately, when large numbers of cranes have been dispatched to a Industrial site, the hazard exposure also increases for workers who work with, around or under these cranes. As per OSHA, 85% of all cranes unsettle and structural failures can be allocate to exceeding the crane’s operational capacity. When a crane is overloaded, it is subject to structural stresses that may cause irreparable damage. Swinging or unexpected dropping of the load, using defective components, hoisting a load far away capacity, pull a load, and side-loading a boom can all cause overloading. The only way to eliminate the accidents is Identify the Hazards to assess the associated controls with the Material Lifting machinery with their operations to bring the hazard to tolerable level. As the part of this dissertation hazard recognition will carried out with the help of checklist and different mathematical methodology with respect to various lifting machines and their control measures will also be given in this work, improved safe work conditions in industry for using this engineering tool this is the objective of this project work and to reduce the number of accidents. The main objective of this study was to examine the various hazards associated with using cranes will help of checklist to identify the risks in the workplace.

Engineering Poster E8: Dynamic Risk Advisor – real time bowtie risk assessment monitoring system

Poster E8 This paper describes the development of a Dynamic Risk Advisor to improve safety of operations in fleet marine construction and production oil and gas vessels, well, and subsea equipment in process safety management, providing rig management onshore and offshore tools to assess health, safety, and environment (HSE) performance in real time. This tool uses the bowtie methodology, and provides a useful framework to present a risk ‘picture’ and to satisfy management of health and safety at work regulations. The focus of the methodology is the relationship between safeguards (control systems/equipment, process controls, and human factors) and an overview of the safety process monitoring in a simple and visual form. As facilities and equipment age, those safeguards may become less effective or degraded (different from the condition at the moment of risk assessment creation/revision), risks may increase and that is why they must be monitored using an effective, continuous, and temporal-aware method. Ocyan and Viasat/Intelie created a solution to identify and analyse the hazards dynamically (in real time) using the LIVE streaming data platform, providing an interface that allows the client to manage the current risk profile of their assets using a bowtie diagram. Additionally, this also provides a way for the user to assess the effect of degraded safeguards on complex operations in real time, by monitoring the actual residual risk. The study started in 2019 and was developed together with the Ocyan HSE team and was extended to all fleet due to the positive results. The tool can be accessed from mobile phones and tablets. To access the poster click the link on the right. To read the full paper click here

Construction site safety monitoring and excavator activity analysis system

With the recent advancements in deep learning and computer vision, the AI-powered construction machine such as autonomous excavator has made significant progress. Safety is the most important section in modern construction, where construction machines are more and more automated. In this paper, we propose a vision-based excavator perception, activity analysis, and safety monitoring system. Our perception system could detect multi-class construction machines and humans in real-time while estimating the poses and actions of the excavator. Then, we present a novel safety monitoring and excavator activity analysis system based on the perception result. To evaluate the performance of our method, we collect a dataset using the Autonomous Excavator System (AES) (Zhang et al., Sci Robot 6(55):eabc3164) including multi-class of objects in different lighting conditions with human annotations. We also evaluate our method on a benchmark construction dataset. The results showed our YOLO v5 multi-class objects detection model improved inference speed by 8 times (YOLO v5 x-large) to 34 times (YOLO v5 small) compared with Faster R-CNN/YOLO v3 model (Zhang et al., In Proceedings of the 38th International Symposium on Automation and Robotics in Construction 461 (ISARC), pp. 49–56. InternationalAssociation for Automation and Robotics in Construction (IAARC), Dubai, UAE (2021). https://doi.org/10.22260/ISARC2021/0009). Furthermore, the accuracy of YOLO v5 models is improved by 2.7% (YOLO v5 x-large) while model size is reduced by 63.9% (YOLO v5 x-large) to 93.9% (YOLO v5 small). The experimental results show that the proposed action recognition approach outperforms the state-of-the-art approaches on top-1 accuracy by about 5.18%. The proposed real-time safety monitoring system is not only designed for our Autonomous Excavator System (AES) in solid waste scenes, it can also be applied to general construction scenarios.

Dynamic Risk Advisor – real time bowtie risk assessment monitoring system

This paper describes the development of a Dynamic Risk Advisor to improve safety of operations in fleet marine construction and production oil and gas vessels, well, and subsea equipment in process safety management, providing rig management onshore and offshore tools to assess health, safety, and environment (HSE) performance in real time. This tool uses the bowtie methodology, and provides a useful framework to present a risk ‘picture’ and to satisfy management of health and safety at work regulations. The focus of the methodology is the relationship between safeguards (control systems/equipment, process controls, and human factors) and an overview of the safety process monitoring in a simple and visual form. As facilities and equipment age, those safeguards may become less effective or degraded (different from the condition at the moment of risk assessment creation/revision), risks may increase and that is why they must be monitored using an effective, continuous, and temporal-aware method. Ocyan and Viasat/Intelie created a solution to identify and analyse the hazards dynamically (in real time) using the LIVE streaming data platform, providing an interface that allows the client to manage the current risk profile of their assets using a bowtie diagram. Additionally, this also provides a way for the user to assess the effect of degraded safeguards on complex operations in real time, by monitoring the actual residual risk. The study started in 2019 and was developed together with the Ocyan HSE team and was extended to all fleet due to the positive results. The tool can be accessed from mobile phones and tablets.

VLCKD: a real time safety study in obesity

Searchable abstracts of presentations at key conferences in endocrinology ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Real-Time Passenger Safety System Using Facial Emotion Recognition

Abstract: In this paper, a “REAL-TIME PASSENGER SAFETY SYSTEM USING FACIAL EMOTION RECOGNITION” is presented using the help of Neural Network Models. Our facial expressions keep on changing, and we have the intelligence of understanding the meaning of several of them. Broadly, we can classify the facial expression as - Happy, Sad, Angry, Scared, Surprised, Disgusted and Neutral. So, the major aim in this project is to show the mood of the person in front of the camera. And this requires that we provide our computer with the intelligence required to do so like our brains. Our brains have neural networks which are responsible for all kinds of thinking (decision-making, understanding) that we do, and we try to develop these neuron capabilities artificially called artificial neural network. Moreover, we will be using the concept of these neural networks to build the application which can recognize and inform about the expression on the face of the passenger. I will also work on adding a feature which notifies close contacts, in case of negative emotions like fear, sad etc. Keywords: Emotion Recognition, Facial Expression, CNN, Deep Learning, Computer Vision

Application of explainable machine learning for real-time safety analysis toward a connected vehicle environment

Due to the difficulty of obtaining traffic flow data and conflicts simultaneously, conflict-based analysis using macroscopic traffic features is much less studied. This research aims to analyze real-time safety by a disaggregate study and explore the benefit of the connected vehicle (CV) for real-time safety evaluation. To avoid the endogeneity problem regarding conflicts and traffic features in regression models, machine learning is employed to obtain a reliable and practical real-time safety model. The results show that the Random Forest outperforms eXtreme Gradient Boosting, Support Vector Machine and Adaptive Boosting models, achieving the best performance with the highest AUC of 0.827. For a deep understanding of conflict mechanisms, the explainable machine learning method SHAP (SHapley Additive exPlanation) is introduced to improve the model interpretability providing insights into the impacts of traffic flow features. Lane difference regarding average speed is found to have the most significant impacts on real-time safety. Speed variation, the proportion of trucks and traffic volume are associated with conflict occurrence. Further analysis highlights that the impacts of traffic features are heterogeneous and there may exist specific patterns of paired features affecting real-time safety. Encouragingly, SHAP appears to be able to complement the traditional model with random components in terms of revealing heterogeneity. The explainable machine learning can also provide a solid basis for discretizing continuous variables while previous studies perform discretization mainly based on prior knowledge and experience. The experimental result regarding CV Market Penetration Rate (CV-MPR) demonstrates that the model performance is gradually elevated with the increase of penetration rate. The initial stage of the CV market (20%, 40% CV-MPR) yields the most significant gains in real-time safety evaluation. These findings can be used beneficially in active traffic management.

Dashboard development for near real-time visualization of COVID-19 vaccine safety surveillance data in the Vaccine Safety Datalink

The Vaccine Safety Datalink (VSD) conducts active surveillance and vaccine safety research studies. Since the start of the U.S. COVID-19 vaccination program, the VSD has conducted near real-time safety surveillance of COVID-19 vaccines using Rapid Cycle Analysis. VSD investigators developed an internal dashboard to facilitate visualization and rapid reviews of large weekly automated vaccine safety surveillance data. Dashboard development and maintenance was informed by vaccine surveillance data users and vaccine safety partners. Key metrics include population demographics, vaccine uptake, pre-specified safety outcomes, sequential analyses results, and descriptive data on potential vaccine safety signals. Dashboard visualizations are used to provide situational awareness on dynamic vaccination coverage and the status of multiple safety analyses conducted among the VSD population. This report describes the development and implementation of the internal VSD COVID-19 Vaccine Dashboard, including metrics used to develop the dashboard, which may have application across various other public health settings.

Adaptive guided wave-based damage identification under unknown load conditions

Damage identification methods based on guided waves (GWs) have been widely researched in the field of aircraft structural health monitoring. Notably, the existing research has not extensively considered the realization of accurate damage localization in an unknown load environment, although this aspect is of significance to the real-time safety assessment of aircraft structures. To address this issue, we propose an adaptive damage imaging method based on GW signals for typical aircraft structures subjected to unknown workloads. First, through the GW signals and load data sampled in the working load range, a mathematical model between the characteristics of the GW signal and load is established. Second, the active GW signal is adaptively compensated by the real-time identified load through the passive strain signal during the service process of the structure. Finally, the alternating time-reversal phase synthesis (ATRPS) method is used to accurately locate the damage in the monitoring area. The feasibility, applicability, and accuracy of the proposed methodology are validated by three levels of experimental cases. All results consistently indicate that reliable damage localization can be achieved under an unknown load environment.

Multi-gas leakage synchronous monitoring system based on the distributed optical fiber sensing technology

In the process of long-distance transportation of different gas, the remote pipeline plays an irreplaceable role in energy transmission. When the pipeline is laid in remote areas for a long distance, it is easy to be influenced by geological disasters and complex working conditions, which may lead to corrosion and leakage. therefore, it is necessary to conduct pipeline gas real-time safety monitoring. An optical fiber gas leakage synchronous monitoring system was proposed and demonstrated based on distributed optical sensing technology for simultaneous multi-gas measurements. In this study, we discuss that the principle of multi-gas leakage synchronous monitoring system is investigated and then validated by the theoretical simulation experiments. Furthermore, gas concentration and leakage location discrimination tests are also conducted in laboratory. The experimental results show that the output intensity values increased obviously along with the gas concentration changes, and the response time of the sensor system is about 40 seconds, and it’s concluded that the multi-gas leakage synchronous monitoring system based on distributed optical fiber sensing technology exhibited good sensing and location discrimination performance.

An Online safety monitoring system of hydropower station based on expert system

The hydropower station real-time safety monitoring is not only related to the safety, but also to the generation benefit of the power station. However, as the amount of monitoring data is very huge, and the dam structure is very complex, real-time monitoring cannot be carried out in most of the dam. It is useful to develop a high-performance safety monitoring system to assist the dam safety management work. As such, the paper introduces an expert system for dam safety management with a mode which can carry out the entire process of real-time dam safety monitoring, covering abnormal importing data, abnormal data changes, monitoring work, dam safety. The system has abundant functions such as dam safety analysis, visual query and remote consultation, etc. The system has been successfully applied to more than ten dam projects in China. The application shows that reliable data evaluation can identify data anomalies such as single point jump, multi point outlier and step change. Trend tracking can find possible structural changes. And clustering comprehensive evaluation can give a reasonable evaluation of the safety state of different parts of the dam. Complete online system can provide a supporting decision-making platform for the safety monitoring of the dams and improve the efficiency of power station management.

Deep-Seated Deformation Mechanisms of Excavated Slope and Control Methods: A Case Study

High slope stability control is a crucial technical issue in water conservancy and hydropower project construction. This issue is relevant throughout the entire engineering-related survey, design, construction, and operation process. This paper reports a recent case in which significant deep-seated deformation occurred during slope excavation. Several unfavorable structural surfaces were revealed throughout the slope’s undercutting process, particularly fault fz39, which ran diagonally through the entire slope from top to bottom. On the slope surface, damage such as cracking, bulging, and local slippage appeared. The measured value of multipoint displacement meters increased in a stepwise pattern with each blast, and the maximum measured value exceeded 100 mm. The studied slope’s whole excavation and reinforcement process is described in detail. We analyzed the deformation mechanism of the excavation slope by combining the geological conditions, real-time safety monitoring findings, and site deformation characteristics. The slope reinforcement scheme was optimized based on the preceding analyses. We used a three-dimensional numerical calculation program to simulate the slope excavation and unloading process to reproduce the current mechanical state of the slope, evaluate the reinforcement measures, and provide a useful reference for decision making. The analysis shows that the deformation behavior of the studied slope is closely related to fault fz39 and fracture L920, and presents a typical wedge failure mode. The deep-seated deformation of the excavation slope occurs at the junction of the wedge and bedrock, which is mainly affected by unfavorable geological structure surfaces, excavation unloading, and blasting vibration. The reinforcement scheme of unloading at the top of the slope and large-tonnage cables are established. The numerical analysis results showed that the design of the slope reinforcement measures was appropriate and ensured slope stability. The work described in this paper deepens our understanding of the occurrence of large deep-seated deformation on the structural-controlled slope under the action of excavation and improves the design philosophy.

Bernstein Polynomial-Based Method for Solving Optimal Trajectory Generation Problems.

This paper presents a method for the generation of trajectories for autonomous system operations. The proposed method is based on the use of Bernstein polynomial approximations to transcribe infinite dimensional optimization problems into nonlinear programming problems. These, in turn, can be solved using off-the-shelf optimization solvers. The main motivation for this approach is that Bernstein polynomials possess favorable geometric properties and yield computationally efficient algorithms that enable a trajectory planner to efficiently evaluate and enforce constraints along the vehicles’ trajectories, including maximum speed and angular rates as well as minimum distance between trajectories and between the vehicles and obstacles. By virtue of these properties and algorithms, feasibility and safety constraints typically imposed on autonomous vehicle operations can be enforced and guaranteed independently of the order of the polynomials. To support the use of the proposed method we introduce BeBOT (Bernstein/Bézier Optimal Trajectories), an open-source toolbox that implements the operations and algorithms for Bernstein polynomials. We show that BeBOT can be used to efficiently generate feasible and collision-free trajectories for single and multiple vehicles, and can be deployed for real-time safety critical applications in complex environments.