35 results about "Safety behaviour" patented technology

The invention relates to a method, a system and equipment for intelligently identifying unsafe behaviors of coal mine operating personnel and a medium. The method comprises the following steps: extracting individual trajectory data from an underground video stream through a multi-target tracking algorithm, decomposing continuous actions into an atomic behavior sequence with a space-time mark by using attitude estimation and a space-time diagram convolutional network, and capturing space-time relevance of a behavior chain through an attention mechanism enhanced long and short term memory network to generate a feature vector; risk reasoning is carried out in combination with the coal mine safety knowledge graph to predict the risk event type and probability, and finally early warning information is generated based on a multi-level early warning strategy. According to the scheme, the crossing from isolated action recognition to behavior chain risk prediction is realized, and the early warning capability of potential safety risks in coal mine operation is remarkably improved through fusion of space-time correlation analysis and domain knowledge of the behavior sequence.

The invention discloses a method and system for automatically detecting unsafe behaviors of workers based on machine vision, and relates to the field of industrial safety, and the method comprises the steps: collecting and preprocessing the behavior data of the workers, the behavior data of the workers comprising position information, equipment state, certificate state and motion trail; performing unsafe behavior detection on the preprocessed operator behavior data to generate a behavior target detection frame; inputting the behavior target detection frame into a target tracking algorithm to perform multi-target tracking, and generating a personnel movement track data set; performing multivariate statistical analysis on the behavior feature vector by using a quantitative theory to generate an unsafe behavior reaction matrix; and evaluating a risk level in the unsafe behavior reaction matrix through a fuzzy analytic hierarchy process, and performing safety early warning notification according to the risk level. According to the method, multivariate statistical analysis is carried out on the behavior feature vectors through a quantification theory, and deep quantification and comprehensive evaluation of the unsafe behavior features of the workers are realized.

The invention discloses an unsafe behavior recognition method and system based on 5G and artificial intelligence, and relates to the technical field of unsafe behavior recognition, and the method comprises the steps: building a reaction matrix through multi-dimensional feature correlation analysis based on an unsafe behavior motion feature sequence with confidence score, and obtaining a feature sequence of the unsafe behavior; predicting the contribution degree of the unsafe behavior action characteristics with confidence scores to the unsafe behaviors, distributing action weights through FAHP, and dynamically dividing the risk levels of the unsafe behaviors; and a multi-agent cooperation framework is constructed based on a dynamic cooperation optimization strategy and a task decomposition and distribution mechanism, detection is performed according to the risk level, and an unsafe behavior identification report is generated. According to the method, the reaction matrix is constructed through multi-dimensional feature correlation analysis, the weight is dynamically adjusted in combination with the FAHP, and the risk levels are divided, so that adaptive risk assessment and efficient decision are realized, and the recognition precision and the response efficiency are improved.

The invention relates to the technical field of chemical industry park safety risk monitoring and management, in particular to a chemical industry park safety risk monitoring and management system based on the Internet of Things, and the system comprises a multi-dimensional risk data collection unit which is used for collecting dynamic operation data; the individual comprehensive risk resolving unit is used for resolving to obtain an individual comprehensive risk index; the regional risk situation assessment unit is used for generating a safety state, a first-level early warning signal or a second-level early warning signal; the safety behavior deviation monitoring unit is used for constructing an individual safety behavior data set; calculating a safety performance index; the risk closed-loop correction control unit is used for calculating a risk correction factor; the corrected individual comprehensive risk is obtained; triggering message notification management, and sending a warning message to the positioning terminal; a safe evacuation instruction is started, and the large-screen monitoring center is linked to start real-time tracking and video monitoring; according to the method, differentiated risk upgrading management of individuals with bad safety habits is realized, and the problem of disjunction between instantaneous risks and long-term behavior trends is effectively solved.

The invention relates to the field of monitoring systems, and particularly discloses a work site safety behavior monitoring system based on AI video monitoring, which comprises a data acquisition module used for acquiring a high-definition video stream of a work site in real time; the edge calculation module is used for preprocessing the video stream; the behavior recognition module is used for detecting and classifying safety behaviors in the video stream through a deep learning model and outputting behavior types and coordinates; and the dynamic weight library comprises a weight configuration unit, an environment factor acquisition unit, a dynamic adjustment unit and a weight output unit. According to the operation site safety behavior monitoring system based on AI video monitoring, environmental factors are fused in real time through the dynamic weight library, the safety behavior risk weight is automatically adjusted, the risk assessment precision in a complex scene is remarkably improved, and false alarm and missing alarm caused by environmental interference are reduced.

The invention discloses a latent variable guided unsafe behavior intervention system and a latent variable guided unsafe behavior intervention method for building workers. A decision laboratory-interpretation structure model is used for analyzing the mutual relation between unsafe behavior factors such as unsafe psychology, unsafe physiology, safety ability and unsafe motivation of the building workers and safety culture, safety atmosphere and safety education. And determining an action path of the intervention measures on the influence factors. And constructing an unsafe behavior occurrence mechanism and intervention mechanism model in combination with the structural equation model. And then simulation research and analysis of the action effect of the intervention strategy are carried out by using system dynamics, and the internal occurrence mechanism and intervention effect of the unsafe behaviors of the building workers are deeply analyzed. The method has the advantages that the relation between unsafe behavior occurrence and intervention paths of workers is studied statically, the unsafe behavior occurrence and intervention process is studied dynamically, the safety management level in the building construction process is improved, and safety accidents are reduced.

The invention discloses a community monitoring system. According to the invention, the behavior identification module can identify various abnormal behaviors in real time, such as fighting, suspicious wandering, crowd gathering and the like, and immediately triggers an alarm to notify a manager to deal with the abnormal behaviors, so that the occurrence of safety accidents is effectively prevented. And the behavior modeling module can identify potential criminal behaviors by establishing normal and abnormal behavior models, so that the crime rate is further reduced. And meanwhile, the behavior trajectory analysis module can track personnel trajectories in real time and quickly position personnel positions in an emergency, so that the emergency response capability is improved, and the safety of residents is ensured. And the behavior analysis module can effectively improve the management efficiency of the cell. Through automatic behavior analysis, the behavior analysis module can reduce the manual workload, and managers can concentrate on key tasks, thereby improving the management efficiency. A behavior analysis report is provided for managers, the managers are helped to know the safety condition of the cell, and a more effective safety management strategy is formulated.

The invention discloses a high plateau airport operator unsafe behavior monitoring and early warning method and device, and relates to the technical field of plateau operation monitoring. According to the method, algorithm such as YOLOv8 and feature fusion are combined to collect physical and behavior data from multiple sources and extract features, then a bimodal model is constructed by SVM and logistic regression, and a comprehensive fatigue level is generated by matching with an evidence theory, so that the limitation of single-modal data is avoided; meanwhile, a fuzzy Bayesian dynamic fusion model is introduced, and a fatigue score is corrected in combination with specific environmental data (low pressure, low temperature and the like) of the high plateau, so that risk judgment is more adaptive to physiological-environmental collaborative risk characteristics of the high plateau scene; and finally, the graded early warning is triggered based on the comprehensive risk grade of fatigue, environment and space dimensions, so that the accuracy of risk identification is realized, the timeliness of high-plateau operation safety management is improved through differentiated intervention, and the potential safety hazard of high-plateau operation is effectively reduced.

The present invention discloses a comprehensive identification method for unsafe behaviors of construction site personnel, which specifically relates to the field of construction site image analysis and identification, including: constructing a boundary segment connection diagram through edge detection and Hough transform, and generating a risk mask area based on plane projection trajectory and buffer zone; performing instance segmentation and posture estimation on personnel, extracting joint points and calculating center of gravity points, screening key points according to operation type and constructing a relative vector sequence; extracting the angle between the main posture axis and the boundary normal of continuous frames, generating a direction consistency matrix and an outward-inclined behavior mark; finally, identifying the resident section within the risk mask area and outputting an oppressive illegal approach behavior label; realizing a "spatial oppression" judgment mechanism based on image feature recognition, ensuring early identification and risk warning of unsafe edge approach behaviors that exist when construction personnel perform different types of operations.

The application discloses a safety education content pushing system and method for a construction site, and belongs to the technical field of safety education of the construction site. The safety education content pushing system comprises the following steps: listing education materials in a tree structure to generate a safety education library, and creating at least one dimension of identification tags for each education material to obtain an identification tag set; performing real-name authentication and numbering on workers to create work information files about the workers; creating a priority pushing model, obtaining pushing keywords required by the workers based on the work information files; mapping the pushing keywords to the identification tag set, matching identification tags having a mapping relationship with the pushing keywords in the identification tag set to obtain a temporary tag set; calling education materials corresponding to the temporary tag set in the safety education library to obtain to-be-pushed education materials, and pushing the to-be-pushed education materials to corresponding workers. The safety education content pushing system achieves efficient and high-precision education content pushing effect based on safety awareness and safety behaviors of the workers, greatly reduces learning time of the workers, and can also make up for part of defects of the workers.

The invention relates to the field of production safety management, in particular to a tipping paper production safety monitoring method and system, and the method comprises the steps: calculating the abnormal score of a person behavior in each frame of real-time image; and generating an abnormal behavior judgment result of the personnel according to the abnormal score, and for the personnel judged to be abnormal, obtaining identity information of the personnel so as to supervise safety behaviors of all personnel in the production place. The method and the device have the effect of improving the smoking behavior recognition accuracy in the production process of personnel.

The utility model provides limited space networking equipment based on unsafe behaviors of a power plant. The limited space networking equipment comprises a mounting platform, a shell, a first sealing part and a second sealing part, the mounting platform is a platform with a certain thickness; the shell is a hemispherical transparent shell, and the shell and the end face of the installation platform are arranged in a separable mode. The first sealing part is arranged on the end surface of the mounting platform and is arranged along the side line of the end surface; the second sealing part is arranged at the edge of the shell; wherein the inner wall of the shell and the end surface of the mounting platform form an accommodating space, and the accommodating space is used for accommodating monitoring equipment; and the first sealing part is matched with the second sealing part, so that the accommodating space is in a sealed state isolated from the outside. According to the application, the shell is arranged on the mounting platform, the sealed accommodating space is formed between the shell and the mounting platform through the sealed matching of the shell and the mounting platform, particularly through the tight fitting of the first sealing part and the second sealing part, and the monitoring equipment is mounted in the accommodating space for moisture-proof and waterproof sealed protection.

This invention discloses a network security abnormal behavior early warning management method in the field of network security technology, including the following steps: S1: constructing a security behavior architecture diagram of the target network; S2: inputting security monitoring behaviors into the security behavior architecture diagram and analyzing the target security behaviors; S3: setting an early warning evaluation interval; S4: judging abnormal monitoring behaviors and analyzing the behavior early warning methods of the abnormal monitoring behaviors; S5: analyzing the architecture layer anomaly coefficient of the behavior architecture layer and analyzing the abnormal behavior architecture layer and its architecture layer early warning method; S6: analyzing the network anomaly coefficient of the target network and analyzing the network early warning method of the target network; and performing abnormal behavior early warning of the target network based on the behavior early warning method, architecture layer early warning method, and network early warning method. This invention improves the comprehensiveness and accuracy of target network abnormal behavior early warning management.

The application discloses a workshop safety production visual early warning method and system based on digital twinning, comprising: dividing workshop system entities into several workshop subsystems, determining production factors in each workshop subsystem, and constructing a workshop digital twinning model from the geometric dimension, behavior dimension and rule dimension; selecting a corresponding data acquisition form to obtain workshop safety production data, classifying and storing the workshop safety production data; and based on the real-time mapping relationship of the data, completing the binding between the workshop safety production data and the workshop digital twinning model; constructing a target detection model to monitor whether unsafe behavior exists in the operating personnel in real time, and issuing a warning information if unsafe behavior exists. Through the target detection technology, the safety early warning is carried out, the digital twinning technology is combined to accurately and timely monitor and manage the safety problems in the complex production process, a safer and more reliable production environment is constructed, the efficiency of safety production management is effectively improved, and the risk of accidents is reduced.

The invention relates to a power construction safety behavior intelligent monitoring and early warning method based on artificial intelligence. The method comprises the following steps: acquiring a scene data snapshot of a construction site; the scene data snapshot is mapped to a preset construction site three-dimensional model, and a dynamic digital twinning scene is obtained; traversing each independent risk source in the dynamic digital twinborn scene, and calculating an influence area of the independent risk source to obtain an independent risk field; performing risk superposition mapping based on the independent risk field to obtain a risk level thermodynamic diagram; performing risk detection based on the risk level thermodynamic diagram and the dynamic digital twinborn scene to obtain a risk behavior; and generating an early warning instruction set based on the risk behavior. By adopting the method, the multi-risk coupling effect can be identified, and when the risks of different working planes are superposed, accurate risk level assessment and early warning decision can be made.

The embodiment of the invention provides an unsafe behavior propagation early warning method and system, and relates to the field of safety management and early warning. The method comprises the following steps: acquiring a risk propagation map; in response to the detection that the first object in the plurality of objects has the unsafe behavior, determining the probability that the second object has the unsafe behavior based on the risk propagation map; and sending early warning information based on the probability. According to the invention, before the unsafe behavior is not widely spread, the probability of the unsafe behavior of the second object is determined based on the constructed risk spreading map, and the early warning information is sent based on the probability result, so that early discovery, early warning and timely intervention of the unsafe behavior are realized, and the occurrence of mass safety accidents is effectively prevented and controlled.

The application discloses a kind of unsafe behavior intelligent identification method and system of construction personnel based on YOLO algorithm, comprising: according to the personnel behavior feature set obtained, the coarse-grained position information is handled using hierarchical identification framework, and the basic action state is determined;If display violation signs, then obtain equipment deviation index by the wearing integrity of safety equipment through medium-grained analysis;Dangerous action details are obtained by using fine-grained recognition to fuse different levels of results;Determine the construction scene by extracting the environmental type characteristics through the scene adaptive module, and call the corresponding parameter configuration to obtain the generalization enhanced behavior recognition output;If exceed the preset threshold, then calculate the angle deviation and spatial deviation with standard action template to obtain risk level score;According to risk level score, track the action of safety officer and correction, judge closed-loop management integrity and record disposal archives.The application realizes the precise identification and closed-loop management and control of unsafe behavior of construction personnel.

The utility model provides a single-well oil pulling flow velocity monitoring and controlling device which comprises a well site oil storage tank, a flow velocity control valve group and a flow velocity control visual control cabinet, and the well site oil storage tank is connected with the flow velocity control valve group through a flange. The flow velocity control valve bank is connected with the oil tank truck through a flange, and the flow velocity control visualization control cabinet is connected to the flow velocity control valve bank. The single-well oil pulling flow rate monitoring and controlling device is convenient to operate, safe and rapid, can achieve the effect of rapid installation through standard flange connection, can be widely applied to a single-well oil pulling site, eliminates unsafe behaviors controlled by experience in the oil loading process, and improves the oil loading efficiency. Potential safety hazards are eliminated from intrinsic safety, the personal safety of workers is ensured, and the labor intensity of the workers is reduced.

The invention relates to the field of monitoring and early warning, in particular to an unsafe behavior monitoring and early warning method and device based on deep learning. The system comprises a wearable monitoring device, a fluid impedance tomography monitoring module and a fluid-structure interaction wave equation inverse solution model. The wearable monitoring device integrates a low-pressure pneumatic flexible network, a magneto-rheological fluid layer and a negative Poisson's ratio auxetic skeleton, collects a pressure pulsation frequency spectrum and electromagnetic parameters in real time, and inverts a normalized stiffness index of a human body contact part by using deep learning; the method is characterized in that magnetic field PWM and air pressure frequency are regulated and controlled in a linkage mode according to a rigidity judgment result, and air pressure pretightening force is used for inducing phase change of magnetorheological fluid and generating high-frequency vibration early warning; according to the invention, the geometric antagonism mechanism of the auxetic skeleton is utilized to ensure the tight fitting of the device and the high fidelity of signals, so that accurate tactile feedback monitoring is realized.

This invention provides an intelligent identification and early warning system for production line personnel safety, comprising: an environmental and personnel feature extraction module for constructing a digital environment model of the work site and extracting key personnel features; a multimodal fusion perception and diagnosis module for receiving the key personnel features and fusing multi-source sensor data to identify and diagnose unsafe behaviors or states of personnel; a situation prediction and assessment module for constructing a personnel safety status deterioration trend model and a comprehensive evaluation model; a data fusion and processing hub module for performing unified data standardization fusion and optimization processing; and an intelligent linkage and decision optimization module for generating and executing optimal safety early warning and linkage control strategies based on the fused and optimized information. This invention can form a complete closed loop from perception, diagnosis, prediction to linkage, significantly improving the intelligence level and response efficiency of production line personnel safety management.

The present application relates to a kind of unsafe behavior intelligent identification method based on YOLO, belong to image recognition field.The present application is for the input scene picture, using YOLOv7 to carry out target detection to picture, output the position information of personnel contained in scene, whether wear safety helmet, whether wear work clothes;Using high-low lens algorithm judges input scene picture belongs to high lens or low lens;According to the input scene picture, select the dangerous area discrimination decision tree model under the corresponding scene;Decision tree model discriminates according to the input 5 parameters whether the human body corresponding to this data is in dangerous area;The personnel position, whether wear work clothes, whether wear safety helmet and whether in dangerous area in the result of step S1, S4 are integrated, and the unsafe behavior grade of personnel in scene is output according to the requirement of label.The algorithm scheme proposed in the present application can quickly and accurately identify the unsafe behavior grade of personnel in video monitoring scene.

The application discloses a research method for the influence of a safe atmosphere on an unsafe behavior supervision decision-making process in a military-civil airport construction, and comprises the following steps: S1, model assumption: determining a game group, a strategy set and a safe and stable strategy; S2, establishing a behavior game value perception matrix; S3, using a replication dynamic equation to perform numerical solution on the model and performing evolution equilibrium point analysis to obtain conditions required to be met by the model stability; and S4, parameter simulation and analysis.The application uses the prospect theory and the psychological account theory to establish the value perception matrix, determines the non-rational and risk preference factors of the safety behavior decision-making of the supervision and construction personnel, simulates the safety behavior decision-making evolution process of the construction related persons through the establishment of an evolution game model conforming to the safety management characteristics of the airport non-stop construction, and according to the simulation experiment result, the unsafe behavior in the construction process can be effectively controlled through the single factor control of the safety supervision.

The invention relates to the technical field of safety behavior assessment, and provides a metallurgical industry employee safety behavior risk index quantification method and device. The method comprises the following steps: analyzing individual evaluation scores to obtain the grade distribution proportion of each evaluation index; constructing an evaluation matrix between the safety factor set and the corresponding evaluation indexes by using the grade distribution proportion; weighting the corresponding evaluation matrix by using the sub weight vectors to obtain sub comprehensive evaluation values; determining a total weight vector of all the safety factor sets; constructing a sub-target evaluation matrix by using the sub comprehensive evaluation values of all the safety factor sets; weighting the sub-target evaluation matrix by using the total weight vector to obtain a total comprehensive evaluation value; and according to the total weight vector and the total comprehensive evaluation value, generating a total target evaluation vector so as to obtain an evaluation result of the group safety behavior, and solving the problem that targeted improvement measures cannot be effectively put forward due to the fact that the group safety behavior condition cannot be quantified.

The present invention relates to the field of trend prediction technology, and in particular, to a method and system for predicting the trend of unsafe behavior of workers based on dual-indicator clustering. The method comprises: stratified sampling of all coal miners; constructing basic risk indicators and unsafe behavior risk indicators; performing kmeans clustering on individuals, teams or district teams among all coal miners based on the basic risk indicators and unsafe behavior risk indicators; constructing a Bayesian discriminant model, and clustering the individuals, teams or district teams among all coal miners based on the Bayesian discriminant model; predicting the unsafe behavior of individuals, teams or district teams among all coal miners based on the results of the clustering, and formulating corresponding preventive safety education plans and preventive control measures. This solves the problem that existing prediction methods cannot be applied to both basic risk indicators and unsafe behavior risk indicators.

The invention provides an unsafe behavior identification method and device based on cross-modal alignment, and a medium. The method comprises the following steps: respectively obtaining inertial measurement unit data and human skeleton key point data of a scaffolding worker; performing cross-modal alignment training on an inertial measurement unit encoder and a human skeleton key point encoder by using the inertial measurement unit data and the human skeleton key point data, so that the inertial measurement unit encoder and the human skeleton key point encoder can align semantic information of heterogeneous modals in the same feature space; only either an inertial measurement unit encoder or a human skeleton key point encoder is used to analyze the input single modal data to identify the unsafe behavior of the worker. Through the innovative design of the cross-modal alignment training framework and the high-performance encoder, the robustness, flexibility and economy of the system in actual deployment are remarkably enhanced while the recognition precision of the unsafe behaviors of the scaffolding workers is improved.

The invention discloses an escalator unsafe behavior data set synthesis method and system, and belongs to the technical field of computer vision and artificial intelligence. The method mainly solves the problem that in an existing escalator monitoring system, the model training effect is poor due to the fact that dangerous behavior data are difficult to collect and samples are insufficient. By innovatively combining a collaborative optimization mechanism of a diffusion model and a large language model (LLM), automatic generation of high-authenticity escalator unsafe behavior video and image data is realized, including but not limited to typical dangerous scenes such as illegal riding of a baby carriage and falling of pedestrians. Compared with a traditional method, the method has the remarkable advantages that on one hand, large-scale and diversified training data can be efficiently generated, and the anomaly detection capability of a security and protection monitoring system is remarkably improved; and on the other hand, the security risk and ethical problems existing in the real scene data acquisition process are completely avoided.

To provide a technology for accurately performing feedback necessary for securing safety of a worker in delivery work.SOLUTION: The behavior extracting unit 13 monitors a worker who performs delivery work using a vehicle, and determines whether or not a safety behavior, which is a behavior for ensuring the safety of the work, is appropriately performed, thereby extracting an inappropriate safety behavior as an unsafe behavior. The psychological analysis unit 14 estimates the psychological state of the worker. The information generating unit 15 generates feedback information including the content of the extracted unsafe behavior, the timing at which the unsafe behavior occurred, and the psychological state of the worker estimated by the psychological analysis unit 14. The notification unit 17 notifies at least one of the worker and a manager who manages the worker of the generated feedback information.SELECTED DRAWING: Figure 1

The present application relates to the technical field of high-altitude operation safety monitoring, and discloses an intelligent safety rope control system and a control method, wherein the working safety sensing data of safety equipment is acquired by a sensing module, then the equipment life state of the safety equipment is evaluated by an intelligent analysis module, and the working safety sensing data is used for working behavior safety evaluation, finally, identity verification is performed by a rope coiling box, an identity verification result is obtained, and a safety determination is made by combining the working behavior evaluation result and the equipment life state result; when the safety determination result meets the preset opening box requirement, the rope coiling box is opened to obtain a safety rope. The present application can avoid the situation that the working personnel does not use the safety equipment in a standard manner due to fear of trouble or lucky psychology, and the working personnel cannot open the rope coiling box to normally work before wearing the safety equipment in a reasonable and compliant manner, thereby standardizing the safety behavior of the working personnel and improving the safety.

The present invention discloses a method for generating unsafe behavior risk portraits based on multimodal data fusion and dynamic visualization, which relates to the technical field of safety management and risk assessment, and includes the following steps: continuously and in real time collecting original data from different modal data sources through network communication to ensure that the information relied on by the overall fusion process is fresh and complete. The present invention effectively identifies high-frequency, low-value redundant data through real-time perception and dynamic regulation of the value of data source information, avoids redundant modal interference in fusion results, and enhances the dominant position of key modal information in risk portraits. The dynamic adjustment of fusion contribution and the adaptive recovery mechanism enhance the system's ability to respond to data changes, making the generated unsafe behavior risk portraits more accurate, dynamic, and targeted, thereby improving the accuracy and reliability of overall safety monitoring and accident warning.