Park safety non-inductive monitoring and emergency response management system

Abstract

The invention discloses a park safety non-inductive monitoring and emergency response management system, relates to the technical field of non-inductive monitoring, and constructs a multi-level and three-dimensional laboratory safety risk assessment system, improves objectivity, precision and dimensionality of dangerous source identification, forms a more real and accurate risk portrait, and improves the risk assessment efficiency. The problems of subjective deviation and monitoring blind areas in traditional manual patrol are avoided, the accuracy of risk identification and the timeliness of emergency response can be improved, meanwhile, a risk source personalized risk assessment strategy is achieved, abstract multi-dimensional environment parameters and personnel behavior characteristics are converted into visual and quantifiable risk situation portraits, and the risk assessment strategy is optimized. According to the invention, scientific and precise regional risk management and control are realized, comprehensive assessment is carried out in combination with a teacher supervision intensity coefficient, an environmental parameter deviation coefficient and a behavior risk coefficient, environmental state monitoring and personnel behavior analysis are considered, and the pertinence of emergency response and the safety management efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of non-intrusive monitoring technology, specifically to a non-intrusive monitoring and emergency response management system for park safety. Background Technology

[0002] University laboratories, as important venues for scientific research and teaching, have complex internal environments, numerous hazards, and frequent personnel flows. The accuracy of safety risk identification and emergency response efficiency during experiments directly affect the level of personal safety protection for teachers and students and the ability to handle emergencies. Therefore, it is essential to construct intelligent sensing and dynamic risk assessment technologies based on multimodal data fusion for comprehensive safety monitoring and scientific emergency decision-making.

[0003] In existing technologies, laboratory safety monitoring systems rely on traditional equipment such as smoke detectors and temperature and humidity sensors located in fixed positions. These systems cannot cover the dynamic behavioral characteristics of personnel in the laboratory and identify multi-level risk factors. They are also difficult to objectively assess personnel's operational compliance, behavior when approaching hazards, the effectiveness of teacher supervision, and students' attention. When faced with various types of hazards such as chemical reagent storage areas, high-temperature equipment areas, high-pressure equipment areas, and radiation source areas, problems such as delayed risk identification, untimely warnings, and inaccurate emergency response are likely to occur.

[0004] Existing technologies lack the ability to jointly collect, deeply analyze, and link risk warnings to multimodal behavioral data such as students' gaze direction, eyelid opening and closing, pupil diameter changes, and deviation from the operation path during experimental operations. They also lack a mechanism to identify the differentiated risk impact of different types of hazards under different environmental conditions and personnel states. They cannot effectively achieve site-specific environmental baseline modeling and personalized risk assessment. Furthermore, existing systems suffer from defects such as low accuracy of environmental parameter collection, difficulty in quantifying teacher supervision intensity, inability to assess students' risk awareness, and static emergency plans, resulting in a significant gap between risk assessment results and the actual safety status of the laboratory. Summary of the Invention

[0005] The purpose of this invention is to provide a park safety non-intrusive monitoring and emergency response management system, which solves the problems existing in the background technology.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a campus security non-intrusive monitoring and emergency response management system, including: a personnel behavior recognition module, used to collect video images of university laboratories, identify each unidentified person in the university laboratory, and verify their identity, thereby distinguishing each student and each teacher in the university laboratory, and obtaining the location of each student and each teacher in the university laboratory.

[0007] The experimental environment monitoring module is used to extract experimental environment features of each student and their location in the university laboratory, thereby determining the characteristic areas of each hazard source in the university laboratory, and mapping environmental parameters to obtain the environmental baseline model of the university laboratory and the initial values ​​of environmental parameters of each hazard source characteristic area.

[0008] The risk assessment and analysis module is used to periodically and dynamically monitor the location of each teacher and the characteristic areas of each hazard source in the university laboratory. It combines the environmental baseline model of the university laboratory with the initial values ​​of environmental parameters of each characteristic area of ​​each hazard source to evaluate the comprehensive risk deviation index of each characteristic area of ​​each hazard source in the university laboratory.

[0009] The emergency response decision module is used to analyze the emergency response level of each hazardous source characteristic area in the university laboratory based on the comprehensive risk deviation index of each hazardous source characteristic area, and to carry out laboratory safety response procedures.

[0010] The beneficial effects of this invention are as follows: This invention constructs a multi-level, three-dimensional laboratory safety risk assessment system, improving the objectivity, accuracy, and dimensionality of hazard identification. It can comprehensively capture students' experimental operation behavior characteristics, teacher supervision status, and dynamic changes in environmental parameters, forming a more realistic and accurate risk profile. This avoids the subjective bias and monitoring blind spots in traditional manual inspections. This invention can deeply explore students' multimodal behavioral characteristics and cognitive states during experiments, such as gaze direction, eyelid opening and closing, and pupil diameter changes, improving the accuracy of risk identification and the timeliness of emergency response. At the same time, this invention realizes a personalized risk assessment strategy for hazard sources, and based on the spatial grid modeling method of the environmental baseline model and the geometric modeling method of the spatial influence sphere, it transforms abstract multi-dimensional environmental parameters and personnel behavior characteristics into a visualized and quantifiable risk situation profile, realizing scientific and precise regional risk management. It also combines teacher supervision intensity coefficient, environmental parameter deviation coefficient, and behavioral risk coefficient for comprehensive evaluation, taking into account environmental status monitoring and personnel behavior analysis, improving the pertinence of emergency response and the efficiency of safety management. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Reference Figure 1 As shown, the present invention provides a park safety non-intrusive monitoring and emergency response management system, including: a personnel behavior recognition module, an experimental environment monitoring module, a risk assessment and analysis module, and an emergency response decision-making module.

[0015] It should be noted that the personnel behavior recognition module is connected to the experimental environment monitoring module, the experimental environment monitoring module is connected to the risk assessment and analysis module, and the risk assessment and analysis module is connected to the emergency response decision-making module.

[0016] The personnel behavior recognition module is used to capture video images of university laboratories, identify each unidentified person in the university laboratory, and verify their identity, thereby distinguishing between each student and each teacher in the university laboratory, and obtaining the location of each student and each teacher in the university laboratory.

[0017] In a specific embodiment of the present invention, each student and each teacher in the university laboratory is distinguished, and the location of each student and each teacher in the university laboratory is obtained. The specific method is as follows: video images of the university laboratory are acquired and human target detection processing is performed on them to obtain human target bounding box data of each person to be identified in the university laboratory at the current monitoring time point, and facial region images of each person to be identified in the university laboratory are extracted based on this data, and feature extraction is performed on them to obtain facial feature vectors of each person to be identified in the university laboratory.

[0018] In one specific embodiment, the method for acquiring video images of a university laboratory is as follows: acquiring video images of the university laboratory through an ultra-high-definition camera in the laboratory.

[0019] In one specific embodiment, the bounding box data of the human body of each person to be identified in the university laboratory at the current monitoring time is obtained, and the facial region images of each person to be identified in the university laboratory are extracted based on this data. Feature extraction is then performed on these images to obtain the facial feature vectors of each person to be identified in the university laboratory. The specific method is as follows: the video images of the university laboratory are processed using an existing human body recognition model to obtain the bounding box data of the human body of each person to be identified in the university laboratory at the current monitoring time. The facial region images of each person to be identified in the university laboratory are then extracted using a facial recognition model. The facial region images of each person to be identified in the university laboratory are then cropped, and feature extraction is performed using a pre-trained facial feature recognition model to obtain the facial feature vectors.

[0020] The system obtains the feature vector sets of the student information database and the feature vector sets of the teacher information database from the local database, and performs similarity calculations by combining them with the facial feature vectors of each unidentified person in the university laboratory, thereby distinguishing each student and each teacher in the university laboratory.

[0021] It should be noted that the local database is used to store the feature vector sets of the student information database and the feature vector sets of the teacher information database, the face similarity threshold, the preset student surrounding range, the hazard source identification model trained by the data, the basic hazard level weight coefficients corresponding to each hazard source type, the safe distance thresholds corresponding to each hazard source type, the influence radius parameters corresponding to each hazard level scoring interval, the personnel exposure risk coefficient threshold, the preset baseline learning time window, the environmental baseline modeling model, the location of each teacher at each historical monitoring time point, the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold corresponding to each personnel number interval, and the emergency response plan templates corresponding to each hazard source type at each emergency response level.

[0022] In one specific embodiment, similarity calculation is performed to distinguish between students and teachers in the university laboratory. The specific method is as follows: obtain the facial feature vectors of each recorded student from the feature vector set of the student information database; calculate the facial similarity between each unidentified person in the university laboratory and each recorded student based on the existing facial similarity calculation model and the facial feature vectors of each unidentified person in the university laboratory; obtain the facial similarity threshold from the local database; if the facial similarity between an unidentified person in the university laboratory and a recorded student is greater than the facial similarity threshold, then the unidentified person is marked as a student, thereby obtaining each student in the university laboratory. Similarly, each teacher in the university laboratory is obtained.

[0023] Based on the bounding box data of the human body of each person to be identified in the university laboratory, the bounding box data of the human body of each student in the university laboratory is extracted, and the pixel coordinates of each student in the university laboratory are extracted, so as to obtain the location of each student.

[0024] Similarly, the location of each teacher in the university laboratory can be obtained.

[0025] The experimental environment monitoring module is used to extract experimental environment features of each student and their location in the university laboratory, thereby determining the characteristic areas of each hazard source in the university laboratory, and mapping environmental parameters to obtain the environmental baseline model of the university laboratory and the initial values ​​of environmental parameters of each hazard source characteristic area.

[0026] In a specific embodiment of the present invention, the characteristic areas of each hazard source in the university laboratory are determined by: analyzing the activity range of each student in the university laboratory based on the location of each student, and merging them spatially to obtain the total activity area of ​​the students in the university laboratory.

[0027] In one specific embodiment, the activity range of each student in the university laboratory is analyzed by: obtaining a preset student perimeter from a local database, and drawing the activity range of each student in the university laboratory based on their location.

[0028] The system acquires environmental monitoring images of the total student activity area in the university laboratory, obtains data from a local database to train a hazard identification model, and inputs the environmental monitoring images of the total student activity area in the university laboratory into the hazard identification model to identify each hazard source in the total student activity area of ​​the university laboratory.

[0029] In one specific embodiment, the environmental monitoring image of the total student activity area in the university laboratory is obtained by means of a camera.

[0030] It should be noted that the hazard identification model is a ResNet (residual network) hazard identification model.

[0031] It should also be noted that the types of hazardous sources include: chemical reagent storage areas, high-temperature equipment areas, high-pressure equipment areas, and radiation source areas.

[0032] Based on the various hazards in the overall student activity area of ​​the university laboratory, analyze the spatial coordinates and hazard level scores of each hazard.

[0033] In one specific embodiment, the spatial location coordinates and hazard level scores of each hazard source in the overall student activity area of ​​a university laboratory are analyzed. The specific method is as follows: Based on each hazard source in the overall student activity area of ​​the university laboratory, the bounding box coordinate data of each hazard source in the environmental monitoring image are extracted from the output of the hazard source identification model. The pixel coordinates of the center point of each hazard source bounding box are calculated, and the pixel coordinates are converted into three-dimensional spatial location coordinates in the laboratory spatial coordinate system through camera calibration parameters and coordinate transformation algorithms, thereby obtaining the spatial location coordinates of each hazard source. The area occupied by each hazard source is calculated based on the bounding box size of each hazard source, and this is used as the scale parameter of each hazard source. The identification label information of each hazard source area is identified through OCR text recognition technology. The storage capacity of chemical reagents, the rated power of high-temperature equipment, the rated pressure of high-pressure equipment, and the activity information of radiation sources are extracted and used as the storage capacity parameter of each hazard source. Based on the hazard source type of each hazard source, the basic hazard level weight coefficient corresponding to each hazard source type is obtained from the local database. Combined with the scale parameter and storage capacity parameter of each hazard source, the hazard level score of each hazard source is obtained through weighted calculation.

[0034] Based on the location of each student in the university laboratory and the spatial coordinates of each hazard in the overall student activity area of ​​the university laboratory, the spatial distance between each student and each hazard is calculated. Combined with the hazard level score of each hazard in the overall student activity area of ​​the university laboratory, the characteristic areas of each hazard in the university laboratory are evaluated.

[0035] In one specific embodiment, the method for evaluating the characteristic areas of each hazard source in a university laboratory is as follows: The safe distance threshold corresponding to each hazard source type is obtained from a local database. Based on the spatial distance values ​​between each student in the university laboratory and each hazard source, when the spatial distance value between a student in the university laboratory and a hazard source is less than the safe distance threshold corresponding to that hazard source type, the hazard source is marked as an active hazard source, thus obtaining each active hazard source in the university laboratory. The influence radius parameter corresponding to each hazard level score interval is obtained from the local database. Based on the hazard level score of each active hazard source in the university laboratory, the influence radius parameter of each active hazard source in the university laboratory is obtained. The higher the hazard level score, the larger the influence radius parameter.

[0036] Using the spatial coordinates of each active hazard source in the university laboratory as the center and the corresponding influence radius parameter as the radius, a spatial influence sphere is established for each active hazard source in the university laboratory. The number of students 'a' within each active hazard source's spatial influence sphere is counted. x Where x represents the number of each active hazard, x = 1, 2, ..., y, and y is a positive integer greater than 2, combined with the hazard level score b of each active hazard in the university laboratory.x Calculate the personnel exposure risk coefficients for each active hazard source in the university laboratory. Where e represents the natural constant, the threshold of personnel exposure risk coefficient is obtained from the local database. If the personnel exposure risk coefficient of an active hazard source in the university laboratory is greater than the threshold, the active hazard source is marked as a vigilant hazard source, thereby obtaining each vigilant hazard source in the university laboratory. The spatial influence sphere of each vigilant hazard source in the university laboratory is extracted, and adjacent vigilant hazard sources are merged through a spatial clustering algorithm, thereby evaluating the characteristic regions of each hazard source in the university laboratory.

[0037] In a specific embodiment of the present invention, the environmental baseline model of the university laboratory and the initial values ​​of environmental parameters of each hazardous source characteristic area are obtained. The specific method is to collect environmental parameter data of the university laboratory.

[0038] It should be noted that the environmental parameter data includes temperature data, humidity data, gas concentration data, and radiation dose data.

[0039] In one specific embodiment, environmental parameter data of the university laboratory is collected by means of various sensors in the university laboratory.

[0040] Based on environmental parameter data from university laboratories, time series analysis is performed to extract the baseline mean and baseline fluctuation range of environmental parameters, thereby constructing an environmental baseline model for university laboratories.

[0041] The environmental baseline model of the university laboratory is spatially gridded, and the initial values ​​of environmental parameters for each hazard source characteristic area of ​​the university laboratory are extracted based on the characteristic areas of each hazard source.

[0042] In a specific embodiment of the present invention, an environmental baseline model for a university laboratory is constructed. The specific method is as follows: time series analysis is performed based on the environmental parameter data of the university laboratory to calculate the environmental baseline data of the university laboratory. The environmental baseline data includes: the mean temperature baseline, the temperature baseline fluctuation range, the mean humidity baseline, the humidity baseline fluctuation range, the mean gas concentration baseline, the gas concentration baseline fluctuation range, the mean radiation dose baseline, and the radiation dose baseline fluctuation range.

[0043] In one specific embodiment, the environmental baseline data of a university laboratory is calculated as follows: A preset baseline learning time window is obtained from a local database. Based on the environmental parameter data of the university laboratory, temperature, humidity, gas concentration, and radiation dose data are collected at various historical monitoring points within this time window to form time-series datasets for each environmental parameter. Outlier detection and removal are performed on the time-series datasets for each environmental parameter. Statistical analysis is then performed on the processed temperature time-series data. Using existing baseline statistical methods, the arithmetic mean of the temperature is calculated as the temperature baseline mean. The standard deviation of the temperature is calculated and combined with the confidence interval to determine the temperature baseline fluctuation range. Similarly, statistical analysis is performed on the humidity time-series data, gas concentration time-series data, and radiation dose time-series data to calculate the humidity baseline mean, humidity baseline fluctuation range, gas concentration baseline mean, gas concentration baseline fluctuation range, radiation dose baseline mean, and radiation dose baseline fluctuation range, thereby obtaining the environmental baseline data of the university laboratory.

[0044] The environmental baseline model is obtained from the local database, and the environmental baseline data of the university laboratory is input into the environmental baseline modeling algorithm to construct the environmental baseline model of the university laboratory.

[0045] It should be noted that the environmental baseline modeling model is a neural network model or statistical regression model trained based on historical environmental data. Its inputs are the baseline mean and baseline fluctuation range of each environmental parameter, and its outputs are the parameter matrix and spatial grid distribution of the environmental baseline model.

[0046] In a specific embodiment of the present invention, the initial values ​​of environmental parameters of each hazardous source characteristic area of ​​the university laboratory are extracted. The specific method is as follows: the environmental baseline model of the university laboratory is spatially gridded to obtain each environmental monitoring grid unit of the university laboratory.

[0047] Based on the characteristic areas of each hazard source in the university laboratory, the spatial coordinates of each characteristic area of ​​each hazard source in the university laboratory are extracted, and the environmental monitoring grid units covered by each characteristic area of ​​each hazard source in the university laboratory are identified.

[0048] Environmental parameter data for each environmental monitoring grid unit were extracted from the environmental baseline model of the university laboratory.

[0049] It should be noted that the environmental parameter data includes: the baseline mean values ​​of temperature, humidity, gas concentration, and radiation dose for the corresponding environmental monitoring grid unit.

[0050] Based on the environmental monitoring grid units covering the characteristic areas of each hazard source in the university laboratory, and combined with the environmental parameter data of each environmental monitoring grid unit in the university laboratory, the initial values ​​of environmental parameters for each characteristic area of ​​each hazard source in the university laboratory are evaluated.

[0051] In one specific embodiment, the initial values ​​of environmental parameters for each hazardous source characteristic area of ​​the university laboratory are evaluated. The specific method is as follows: based on the environmental monitoring grid units covered by each hazardous source characteristic area of ​​the university laboratory, the spatial overlap area between each hazardous source characteristic area of ​​the university laboratory and each environmental monitoring grid unit is calculated, and the proportion of the overlap area of ​​each environmental monitoring grid unit of the university laboratory to the total area of ​​each hazardous source characteristic area is calculated. This proportion is used as the area weighting coefficient of each environmental monitoring grid unit of each hazardous source characteristic area of ​​the university laboratory. The average temperature baseline of each covered grid unit is extracted from the environmental baseline model of the university laboratory.

[0052] Based on the area weighting coefficients of each environmental monitoring grid unit in the characteristic areas of each hazard source in the university laboratory, the average temperature baseline of each covered environmental monitoring grid unit is weighted and averaged to obtain the initial temperature value of each characteristic area of ​​the hazard source. Similarly, the average humidity baseline, average gas concentration baseline, and average radiation dose baseline of each covered grid unit are weighted and averaged to obtain the initial humidity value, initial gas concentration value, and initial radiation dose value of each characteristic area of ​​the hazard source. The initial temperature value, initial humidity value, initial gas concentration value, and initial radiation dose value are combined to evaluate the initial environmental parameter values ​​of each characteristic area of ​​the hazard source in the university laboratory.

[0053] The risk assessment and analysis module is used to periodically and dynamically monitor the location of each teacher and the characteristic areas of each hazard source in the university laboratory. It combines the environmental baseline model of the university laboratory with the initial values ​​of environmental parameters of each characteristic area of ​​each hazard source to evaluate the comprehensive risk deviation index of each characteristic area of ​​each hazard source in the university laboratory.

[0054] In a specific embodiment of the present invention, the comprehensive risk deviation index of each hazardous source characteristic area of ​​a university laboratory is evaluated by means of: obtaining the location of each teacher at each historical monitoring time point from a local database, simulating the movement trajectory and patrol coverage of each teacher in the university laboratory based on the location of each teacher in the university laboratory, and evaluating the teacher supervision intensity coefficient of each hazardous source characteristic area of ​​the university laboratory by combining the spatial coordinates of each hazardous source characteristic area of ​​the university laboratory.

[0055] In one specific embodiment, the movement trajectories and patrol coverage areas of each teacher in a university laboratory are simulated. Combined with the spatial coordinates of the characteristic areas of each hazard source in the university laboratory, the teacher supervision intensity coefficient for each hazard source characteristic area is evaluated. The specific method is as follows: Based on the location of each teacher at each historical monitoring time point, the position coordinates of each teacher at each historical monitoring time point are connected in chronological order to simulate the movement trajectory of each teacher in the university laboratory. A circular coverage area is established with the coordinates of each trajectory point as the center and a preset teacher's field of vision radius as the radius, thus obtaining the patrol coverage area of ​​each teacher. Based on the spatial coordinates of the characteristic areas of each hazard source in the university laboratory, it is determined whether there is spatial overlap between each hazard source characteristic area and the patrol coverage area of ​​each teacher. When spatial overlap exists, it is recorded as one patrol. The total number of patrols and patrol duration for each hazard source characteristic area within a preset time window are counted. Based on the total number of patrols g for each hazard source characteristic area in the university laboratory... n Distance h between each patrol and the hazard source ni Patrol duration k ni Where n represents the number of each hazardous source characteristic area, n = 1, 2, ..., m, where m is a positive integer greater than 2, and i represents the number of each inspection, i = 1, 2, ..., j, where j is a positive integer greater than 2, thereby evaluating the teacher supervision intensity coefficient of each hazardous source characteristic area in the university laboratory.

[0056] The system acquires real-time environmental parameter data of each hazardous source characteristic area in the university laboratory at the current monitoring time point, and calculates the environmental parameter deviation coefficient of each hazardous source characteristic area in the university laboratory based on the initial values ​​of the environmental parameters of each hazardous source characteristic area.

[0057] In one specific embodiment, the deviation coefficient of environmental parameters for each hazardous source characteristic area of ​​the university laboratory is calculated. The specific method is as follows: based on the real-time environmental parameter data d of each hazardous source characteristic area of ​​the university laboratory at the current monitoring time point... n And combined with the initial values ​​of environmental parameters f of the characteristic areas of each hazardous source in the university laboratory n Calculate the deviation coefficient of environmental parameters for each hazardous source characteristic area in the university laboratory.

[0058] Obtain student field-of-view data for the characteristic areas of various hazards in university laboratories at the current monitoring time.

[0059] It should be noted that student behavior data includes: gaze direction and eye physiological characteristics data.

[0060] It should also be noted that the data on eye physiological characteristics include: interocular distance data, pupil diameter data, and eyelid opening and closing data.

[0061] Based on the students' field of vision data of each hazardous source characteristic area in the university laboratory, the behavioral risk coefficient of each hazardous source characteristic area in the university laboratory is calculated.

[0062] Based on the teacher supervision intensity coefficient, environmental parameter deviation coefficient, and behavioral risk coefficient of each hazardous source characteristic area in the university laboratory, the comprehensive risk deviation index of each hazardous source characteristic area in the university laboratory is evaluated.

[0063] In one specific embodiment, the comprehensive risk deviation index of each hazard source characteristic area of ​​the university laboratory is evaluated. The specific method is as follows: based on the teacher supervision intensity coefficient ε of each hazard source characteristic area of ​​the university laboratory... n Environmental parameter deviation coefficient δ n and behavioral risk coefficient φ n Calculate the comprehensive risk deviation index of each hazard source characteristic area in the university laboratory.

[0064] In a specific embodiment of the present invention, the behavioral risk coefficient of each hazardous source characteristic area in a university laboratory is calculated by: extracting the gaze direction and eye physiological characteristics data of each student in each hazardous source characteristic area of ​​the university laboratory based on the student's visual field data at the current monitoring time point.

[0065] Based on the eye physiological characteristic data of each student in each hazard source characteristic area of ​​the university laboratory, an eye trajectory fitting analysis was performed. Combined with the gaze direction of each student in each hazard source characteristic area of ​​the university laboratory, the coordinates of the eye fixation point of each student in each hazard source characteristic area of ​​the university laboratory were obtained, thereby determining the real-time field of vision of each student in each hazard source characteristic area of ​​the university laboratory.

[0066] In one specific embodiment, the eye fixation point coordinates of each student within the characteristic areas of each hazard source in a university laboratory are obtained, thereby determining the real-time visual field range of each student within these areas. The specific method is as follows: Based on the physiological eye characteristic data of each student within the characteristic areas of each hazard source in the university laboratory, the interocular distance data, pupil diameter data, and eyelid opening / closing degree data of each student are extracted. Combined with existing eye angle estimation techniques, the eye fixation point coordinates of each student within the characteristic areas of each hazard source in the university laboratory can be obtained. Centered on the eye fixation point coordinates of each student, a visual field cone with a 120-degree visual field angle in the horizontal direction and a 90-degree visual field angle in the vertical direction is established based on human eye visual field angle parameters. The visual field range is adaptively magnified or reduced by combining the eyelid opening / closing degree data of each student, thereby determining the real-time visual field range of each student within the characteristic areas of each hazard source in the university laboratory.

[0067] Based on the real-time field of vision of each student within the characteristic area of ​​each hazard source in the university laboratory, and combined with the spatial coordinates of each characteristic area of ​​each hazard source in the university laboratory, the behavioral risk coefficient of each characteristic area of ​​each hazard source in the university laboratory is calculated.

[0068] In one specific embodiment, the behavioral risk coefficient of each hazard-source characteristic area in a university laboratory is calculated as follows: Based on the real-time field of vision of each student within each hazard-source characteristic area of ​​the university laboratory, and combined with the spatial coordinates of each hazard-source characteristic area, a three-dimensional spatial intersection algorithm is used to determine whether the real-time field of vision of each student overlaps with the hazard-source characteristic area. This yields the visual overlap volume rnp between each student and the area within each hazard-source characteristic area of ​​the university laboratory, where p represents the student's number, p = 1, 2, ..., q, and q is a positive integer greater than 2. Based on the pupil diameter data of each student, the rate of change t of the pupil diameter of each student within each hazard-source characteristic area of ​​the university laboratory when approaching the hazard is calculated. np This allows us to obtain the behavioral risk coefficients of the characteristic areas of each hazard source in the university laboratory.

[0069] The emergency response decision module is used to analyze the emergency response level of each hazardous source characteristic area in the university laboratory based on the comprehensive risk deviation index of each hazardous source characteristic area, and to carry out laboratory safety response procedures.

[0070] In a specific embodiment of the present invention, the emergency response level of each hazardous source characteristic area of ​​the university laboratory is analyzed. The specific method is as follows: obtain the number of personnel in each hazardous source characteristic area of ​​the university laboratory, obtain the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold corresponding to each personnel number interval from the local database, and thus map the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold of each hazardous source characteristic area of ​​the university laboratory.

[0071] Based on the comprehensive risk deviation index of each hazardous source characteristic area in the university laboratory, and by comparing it with the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold of each hazardous source characteristic area in the university laboratory, the emergency response level of each hazardous source characteristic area in the university laboratory is obtained.

[0072] It should be noted that the emergency response levels include: Level I, Level II, and Level III. Level I is more dangerous than Level II, and Level II is more dangerous than Level III.

[0073] In one specific embodiment, the emergency response level of each hazardous source characteristic area in the university laboratory is obtained by the following method: when the comprehensive risk deviation index of a hazardous source characteristic area is less than the second-level comprehensive risk deviation index threshold, the emergency response level of the hazardous source characteristic area is the third-level response level; when the comprehensive risk deviation index of a hazardous source characteristic area is between the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold, the emergency response level of the hazardous source characteristic area is the second-level response level; and when the comprehensive risk deviation index of a hazardous source characteristic area is greater than the first-level comprehensive risk deviation index threshold of the area, the emergency response level of the hazardous source characteristic area is the first-level response level.

[0074] In a specific embodiment of the present invention, laboratory safety response processing is carried out by determining the hazard source type of each hazard source characteristic area of ​​the university laboratory based on the hazard sources in the total student activity area of ​​the university laboratory.

[0075] The emergency response plan templates for each type of hazard source at each emergency response level are obtained from the local database. Based on the emergency response level and hazard source type of each hazard source characteristic area in the university laboratory, the emergency response plan templates for each hazard source characteristic area in the university laboratory are determined.

[0076] Based on the locations of students and teachers in the university laboratory, and in conjunction with the emergency response plan templates for each hazardous area in the university laboratory, the emergency response plan parameters were adjusted to obtain the revised emergency response plan templates for each hazardous area in the university laboratory, and then safety response procedures were implemented.

[0077] In one specific embodiment, emergency response plan parameter adjustment is performed to obtain an adjusted emergency response plan template for each hazardous source characteristic area of ​​the university laboratory. The specific method is as follows: based on the location of each student and each teacher in the university laboratory, the number of students and teachers in each hazardous source characteristic area is counted, and the average distance between each hazardous source characteristic area and the nearest safety exit is calculated. Based on the emergency response plan template for each hazardous source characteristic area, adjustable parameters are extracted from the plan template, including parameters for the number of evacuees, evacuation route selection, emergency resource allocation, teacher guidance personnel allocation, and estimated evacuation time. The adjustable parameters in the plan template are adjusted according to the number of students, the number of teachers, and the average distance to the nearest safety exit in each hazardous source characteristic area, thereby obtaining the adjusted emergency response plan template for each hazardous source characteristic area of ​​the university laboratory.

[0078] This invention constructs a multi-layered, three-dimensional laboratory safety risk assessment system, enhancing the objectivity, accuracy, and dimensionality of hazard identification. It comprehensively captures students' experimental operational behaviors, teacher supervision status, and dynamic changes in environmental parameters, forming a more realistic and accurate risk profile. This avoids the subjective bias and monitoring blind spots inherent in traditional manual inspections. The invention deeply mines students' multimodal behavioral characteristics and cognitive states during experiments, such as gaze direction, eyelid opening and closing, and pupil diameter changes, improving the accuracy of risk identification and the timeliness of emergency response. Furthermore, this invention implements personalized hazard assessment strategies and, based on spatial grid modeling of environmental baseline models and geometric modeling of spatial influence spheres, transforms abstract multidimensional environmental parameters and personnel behavioral characteristics into visualized and quantifiable risk profiles. This enables scientific and precise regional risk management. Simultaneously, it combines teacher supervision intensity coefficients, environmental parameter deviation coefficients, and behavioral risk coefficients for comprehensive evaluation, taking into account both environmental status monitoring and personnel behavior analysis, thereby improving the targeting of emergency response and the efficiency of safety management.

[0079] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0080] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A park safety non-sense monitoring and emergency response management system, characterized in that, The method comprises the following steps: The personnel behavior recognition module is used for collecting video images of the university laboratory, recognizing each person to be determined in the university laboratory, and verifying the identity of each person to be determined, so as to distinguish each student and each teacher in the university laboratory, and obtain the positions of each student and each teacher in the university laboratory; The experimental environment monitoring module is used for extracting experimental environment features of each student and the position of each student, so as to determine each dangerous source feature area of the university laboratory, and perform environment parameter mapping to obtain an environment baseline model of the university laboratory and initial values of environment parameters of each dangerous source feature area; The risk assessment analysis module is used for performing periodic dynamic behavior detection on the positions of each teacher and each dangerous source feature area of the university laboratory, and combining the environment baseline model of the university laboratory and the initial values of environment parameters of each dangerous source feature area, to evaluate a comprehensive risk deviation index of each dangerous source feature area of the university laboratory; The emergency response decision module is used for analyzing an emergency response level of each dangerous source feature area of the university laboratory according to the comprehensive risk deviation index of each dangerous source feature area of the university laboratory, and performing laboratory safety response processing.

2. The park safety in-sight monitoring and emergency response management system of claim 1, wherein, The method for distinguishing each student and each teacher in the university laboratory, and obtaining the positions of each student and each teacher in the university laboratory, comprises the following steps: Video images of the university laboratory are obtained, and human target detection processing is performed on the video images, so as to obtain human target frame data of each person to be determined in the university laboratory at a current monitoring time point, and human face region images of each person to be determined in the university laboratory are extracted according to the human target frame data, and feature extraction is performed on the human face region images, so as to obtain human face feature vectors of each person to be determined in the university laboratory; Student information database feature vector sets and teacher information database feature vector sets are obtained from a local database, and similarity calculation is performed in combination with the human face feature vectors of each person to be determined in the university laboratory, so as to distinguish each student and each teacher in the university laboratory; According to the human target frame data of each person to be determined in the university laboratory, human target frame data of each student in the university laboratory are extracted, and pixel coordinates of each student in the university laboratory are extracted, so as to obtain the positions of each student; The positions of each teacher in the university laboratory are obtained in the same way.

3. The park safety in-sight monitoring and emergency response management system of claim 2, wherein, The method for determining each dangerous source feature area of the university laboratory comprises the following steps: According to the positions of each student in the university laboratory, the activity range of each student in the university laboratory is analyzed, and spatial merging is performed, so as to obtain a total activity area of students in the university laboratory; An environment monitoring image of the total activity area of students in the university laboratory is obtained, a dangerous source recognition model trained by data is obtained from a local database, the environment monitoring image of the total activity area of students in the university laboratory is input into the dangerous source recognition model, so as to recognize each dangerous source of the total activity area of students in the university laboratory; According to each dangerous source of the total activity area of students in the university laboratory, the spatial position coordinates and the danger level score of each dangerous source of the total activity area of students in the university laboratory are analyzed. According to the positions of the students in the university laboratory, and in combination with the spatial position coordinates of the danger sources in the total student activity area of the university laboratory, the spatial distance values between the students and the danger sources in the university laboratory are calculated, and in combination with the danger level scores of the danger sources in the total student activity area of the university laboratory, the characteristics of the danger sources in the university laboratory are evaluated.

4. The park safety in-sight monitoring and emergency response management system of claim 3, wherein, The environmental parameter initial values of the environment baseline model and the characteristics of the danger sources in the university laboratory are obtained, and the specific method is as follows: The environmental parameter data of the university laboratory is collected; According to the environmental parameter data of the university laboratory, time series analysis is performed to extract the baseline mean and baseline fluctuation range of the environmental parameters, thereby constructing the environmental baseline model of the university laboratory; The environmental baseline model of the university laboratory is spatially gridded, and the environmental parameter initial values of the characteristics of the danger sources in the university laboratory are extracted according to the characteristics of the danger sources in the university laboratory.

5. The park safety in-sight monitoring and emergency response management system of claim 4, wherein, The specific method for constructing the environmental baseline model of the university laboratory is as follows: According to the time series analysis of the environmental parameter data of the university laboratory, the environmental baseline data of the university laboratory is calculated, including the temperature baseline mean, the temperature baseline fluctuation range, the humidity baseline mean, the humidity baseline fluctuation range, the gas concentration baseline mean, the gas concentration baseline fluctuation range, the radiation dose baseline mean, and the radiation dose baseline fluctuation range; The environmental baseline modeling model is obtained from the local database, and the environmental baseline data of the university laboratory is input into the environmental baseline modeling algorithm, thereby constructing the environmental baseline model of the university laboratory.

6. The park safety in-sight monitoring and emergency response management system of claim 4, wherein, The specific method for extracting the environmental parameter initial values of the characteristics of the danger sources in the university laboratory is as follows: The environmental baseline model of the university laboratory is spatially gridded, thereby obtaining each environmental monitoring grid cell in the university laboratory; According to the characteristics of the danger sources in the university laboratory, the spatial position coordinates of the characteristics of the danger sources in the university laboratory are extracted, and the environmental monitoring grid cells covered by the characteristics of the danger sources in the university laboratory are identified; The environmental parameter data of each environmental monitoring grid cell is extracted from the environmental baseline model of the university laboratory; According to the environmental monitoring grid cells covered by the characteristics of the danger sources in the university laboratory, and in combination with the environmental parameter data of each environmental monitoring grid cell in the university laboratory, the environmental parameter initial values of the characteristics of the danger sources in the university laboratory are obtained.

7. The park safety in-sight monitoring and emergency response management system of claim 6, wherein, The specific method for evaluating the comprehensive risk deviation index of the characteristics of the danger sources in the university laboratory is as follows: The positions of the teachers at each historical monitoring time point are obtained from the local database, and according to the positions of the teachers in the university laboratory, the moving track and the patrol coverage range of the teachers in the university laboratory are simulated, and in combination with the spatial position coordinates of the characteristics of the danger sources in the university laboratory, the teacher supervision intensity coefficient of the characteristics of the danger sources in the university laboratory is evaluated. The real-time environmental parameter data of each hazard source characteristic region of the university laboratory at the current monitoring time point is acquired, and the environmental parameter deviation coefficient of each hazard source characteristic region of the university laboratory is calculated according to the environmental parameter initial value of each hazard source characteristic region of the university laboratory; The student visual field data of each hazard source characteristic region of the university laboratory at the current monitoring time point is acquired; The behavior risk coefficient of each hazard source characteristic region of the university laboratory is calculated according to the student visual field data of each hazard source characteristic region of the university laboratory. The comprehensive risk deviation index of each hazard source characteristic region of the university laboratory is evaluated according to the teacher supervision intensity coefficient, the environmental parameter deviation coefficient and the behavior risk coefficient of each hazard source characteristic region of the university laboratory.

8. The park safety in-sight monitoring and emergency response management system of claim 7, wherein, The behavior risk coefficient of each hazard source characteristic region of the university laboratory is calculated according to the student visual field data of each hazard source characteristic region of the university laboratory. The line-of-sight direction and the eye physiological feature data of each student in each hazard source characteristic region of the university laboratory are extracted according to the student visual field data of each hazard source characteristic region of the university laboratory at the current monitoring time point. The eye fixation point coordinates of each student in each hazard source characteristic region of the university laboratory are obtained by performing eye trajectory fitting analysis according to the eye physiological feature data of each student in each hazard source characteristic region of the university laboratory, and combining the line-of-sight direction of each student in each hazard source characteristic region of the university laboratory, so as to determine the real-time visual field range of each student in each hazard source characteristic region of the university laboratory. The behavior risk coefficient of each hazard source characteristic region of the university laboratory is calculated according to the real-time visual field range of each student in each hazard source characteristic region of the university laboratory, and combining the spatial position coordinates of each hazard source characteristic region of the university laboratory.

9. The park safety in-sight monitoring and emergency response management system of claim 1, wherein, The emergency response level of each hazard source characteristic region of the university laboratory is analyzed according to the personnel quantity of each hazard source characteristic region of the university laboratory, the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold corresponding to each personnel quantity interval are acquired from the local database, so as to map the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold of each hazard source characteristic region of the university laboratory. The emergency response level of each hazard source characteristic region of the university laboratory is obtained by comparing the comprehensive risk deviation index of each hazard source characteristic region of the university laboratory with the first-level comprehensive risk deviation index threshold and the second-level comprehensive risk deviation index threshold of each hazard source characteristic region of the university laboratory. The laboratory safety response processing is performed according to the following steps:

10. The park safety in-sight monitoring and emergency response management system of claim 3, wherein, The hazard source type of each hazard source characteristic region of the university laboratory is determined according to the total activity region of the students of the university laboratory; The emergency response plan template corresponding to each emergency response level is acquired from the local database according to each hazard source type, and the emergency response plan template of each hazard source characteristic region of the university laboratory is determined according to the emergency response level and the hazard source type of each hazard source characteristic region of the university laboratory. ​ According to the positions of the students and the positions of the teachers in the university laboratory, and in combination with the emergency response plan templates of the characteristic regions of the dangerous sources of the university laboratory, the emergency response plan templates are adjusted, so that the adjusted emergency response plan templates of the characteristic regions of the dangerous sources of the university laboratory are obtained, and the safety response is processed.

Images