Intelligent protective helmet rapid response method and system based on emergency rescue scenario

Through the intelligent sensing module, the emergency rescue scenarios are obtained and mapped, the risk indicators are generated and the early warning response is sent, which solves the problems of inaccurate detection and unstable communication in complex environments, and improves the response speed and safety of emergency rescue .

CN118823992BActive Publication Date: 2024-11-29JIANGSU KETIANSHI INTELLIGENT TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411311641.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-20
Publication Date
2024-11-29
Estimated Expiration
2044-09-20

AI Technical Summary

Technical Problem

Existing intelligent protective helmets do not detect danger signals accurately or respond quickly enough in complex environments, resulting in poor safety protection effects of emergency rescue personnel and may be affected by the environment, resulting in unstable communication signals.

Method used

Use intelligent sensing modules to obtain environment, vital signs, image and location data flows, conduct emergency rescue scene mapping, traverse preset scene features for similarity comparison, generate risk indicators and send early warning response information, and use feature extraction and noise reduction Convergence model improves response accuracy.

Benefits of technology

It realizes the faster and more accurate response of intelligent protective helmets in emergency rescue scenarios, improving the safety and communication stability of rescue personnel.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118823992B_ABST
    Figure CN118823992B_ABST
Patent Text Reader

Abstract

The present invention discloses a quick response method and system for an intelligent protective helmet based on an emergency rescue scenario, and relates to the field of data processing technology. The method includes: obtaining an intelligent sensing module of a first intelligent protective helmet; putting the first intelligent protective helmet on a first rescuer, collecting environmental data streams, vital signs data streams, image data streams, and position data streams; performing emergency rescue scenario mapping, and obtaining dynamic rescue scenarios; traversing the preset scenario features in the emergency rescue scenario library for similarity comparison, and generating a first risk indicator; performing an early warning response, generating a first early warning response message, and sending it to a rescue command center. The method solves the technical problem that it is difficult for existing emergency rescue protective helmets to accurately perceive complex and changeable environmental parameters and vital signs of personnel in real time, resulting in inaccurate and rapid identification of danger signals, and making the protective helmets respond poorly, and achieves the technical effect of faster response of intelligent protective helmets during emergency rescue.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to fields related to data processing technology, and specifically to a rapid response method and system for an intelligent protective helmet based on emergency rescue scenarios. Background Art

[0002] In modern emergency rescue scenarios, facing the complex and changeable natural environment, potential safety hazards and high-intensity work pressure, the safety protection of rescuers is extremely important. Smart protective helmets have become an indispensable smart equipment in emergency rescue. With the rapid development of science and technology, intelligence and informatization have become important trends in modern emergency rescue. Smart protective helmets, as the "smart bodyguards" of emergency rescue personnel, not only have the protective functions of traditional protective helmets, but also integrate advanced sensor technology, Internet of Things technology, artificial intelligence algorithms, etc., to achieve monitoring and response to the rescue environment and data analysis. However, traditional smart protective helmets may have certain limitations in sensor technology and data processing capabilities, resulting in the helmet's inaccurate detection of dangerous signals in complex environments or insufficient response, thereby affecting the rescuers' judgment and actions. At the same time, traditional smart protective helmets may be affected by the on-site environment (such as building obstruction, electromagnetic interference, etc.), resulting in unstable or interrupted communication signals, affecting the real-time communication and collaboration between rescuers and the command center or other rescue team members.

[0003] Therefore, in the current event denoising convergence related technologies, there is a technical problem that the load data feature acquisition is inaccurate, resulting in an inaccurate denoising model, which makes the accuracy and reliability of the event denoising convergence results low. Summary of the invention

[0004] This application provides a quick response method and system for intelligent protective helmets based on emergency rescue scenarios, and adopts technical means such as feature extraction and matching, and construction of noise reduction convergence models, to solve the technical problem that existing emergency rescue protective helmets are difficult to accurately perceive complex and changeable environmental parameters and personnel vital signs in real time, resulting in inaccurate and rapid identification of danger signals, resulting in poor response effect of protective helmets. This application achieves the technical effect of faster response of intelligent protective helmets during emergency rescue.

[0005] The present application provides a rapid response method for an intelligent protective helmet based on an emergency rescue scenario, the method comprising: obtaining an intelligent sensor module of a first intelligent protective helmet, wherein the intelligent sensor module comprises an environmental monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module; putting the first intelligent protective helmet on a first rescuer, and collecting an environmental data stream, a vital sign data stream, an image data stream and a position data stream through the intelligent sensor module; performing emergency rescue scene mapping based on the environmental data stream, the vital sign data stream, the image data stream and the position data stream to obtain a dynamic rescue scene; based on the dynamic rescue scene, traversing the preset scene features in the emergency rescue scene library for similarity comparison to generate a first risk indicator; performing an early warning response based on the first risk indicator, generating a first early warning response message and sending it to a rescue command center.

[0006] In a possible implementation, emergency rescue scene mapping is performed based on the environmental data stream, the vital signs data stream, the image data stream and the position data stream to obtain a dynamic rescue scene, and the following processing is performed: static indicator identification is performed on the environmental data stream and the vital signs data stream to obtain a static factor state; dynamic indicator identification is performed on the environmental data stream and the vital signs data stream to obtain a dynamic factor state; moving target identification is performed based on the image data stream to generate a moving target state, wherein the moving target state includes the scale and falling frequency of the moving target; scene factor state is constructed with the static factor state, the dynamic factor state and the moving target state; scene destruction simulation is performed based on the position data stream to obtain a destruction simulation result; and the dynamic rescue scene is constructed with the destruction simulation result and the scene factor state.

[0007] In a possible implementation, moving target recognition is performed based on the image data stream to generate a moving target state, wherein the moving target state includes the scale and dropping frequency of the moving target, and the following processing is performed: according to the first step length, images in the image data stream are extracted and down-sampled to obtain a first extracted image set; according to the second step length, images are selected from the image data stream to obtain a second extracted image set, wherein the second step length is greater than the first step length, and the number of images in the second extracted image set is less than the number of images in the first extracted image set; based on the slowfast network, motion feature recognition is performed on the first extracted image set and the second extracted image set, and the moving target state is output.

[0008] In a possible implementation, a scene destruction simulation is performed based on the position data stream to obtain a destruction simulation result, and the following processing is also performed: a pre-disaster three-dimensional model of the building within a preset range is obtained with the position recognition result corresponding to the position data stream as the center; an external post-disaster image of the position recognition result is obtained through the rescue command center; and a disaster building damage simulation is performed in combination with the external post-disaster image and the pre-disaster three-dimensional model of the building to obtain the destruction simulation result.

[0009] In a possible implementation, based on the dynamic rescue scenario, the preset scenario features in the emergency rescue scenario library are traversed for similarity comparison to generate a first risk index, and the following processing is performed: based on the dynamic rescue scenario, the preset scenario features and the preset risk index are obtained; the dynamic rescue scenario and the preset scenario features are compared for similarity to generate a similarity index; if the similarity index is greater than the preset similarity, the preset risk index of the preset scenario features is tolerantly processed based on the similarity index to generate the first risk index.

[0010] In a possible implementation, the preset scene feature is obtained based on the dynamic rescue scene, and the following processing is also performed: the emergency rescue scene library contains multiple scene samples, wherein any scene sample includes a scene destruction simulation sample, a scene factor state sample and a preset risk indicator, wherein the preset risk indicator is calibrated by an expert system; the scene destruction simulation sample and the scene factor state sample of any scene sample are extracted from the multiple scene samples as the preset scene feature.

[0011] In a possible implementation, a warning response is performed based on the first risk indicator, a first warning response information is generated and sent to a rescue command center, and the following processing is also performed: the first smart protective helmet communicates data with the rescue command center through a first communication channel, and when the first communication channel sends a communication interruption reminder signal, the risk indicator sequence and interruption location information sent by the first smart protective helmet to the rescue command center are obtained; risk prediction is performed in a preset time zone based on the risk indicator sequence, and a risk prediction time sequence is generated; and an emergency response decision is generated based on the interruption location information and the risk prediction time sequence.

[0012] This application also provides a smart protective helmet rapid response system based on emergency rescue scenarios, including:

[0013] An intelligent sensor module acquisition module, the intelligent sensor module acquisition module is used to acquire the intelligent sensor module of the first intelligent protective helmet, wherein the intelligent sensor module includes an environment monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module;

[0014] A data stream collection module, wherein the data stream collection module is used to collect environmental data streams, vital sign data streams, image data streams and position data streams through the intelligent sensor module when the first rescuer wears the first intelligent protective helmet;

[0015] An emergency rescue scene mapping module, which performs emergency rescue scene mapping based on the environmental data stream, the vital sign data stream, the image data stream and the position data stream to obtain a dynamic rescue scene;

[0016] A first risk indicator generation module, the first risk indicator generation module is used to traverse the preset scene features in the emergency rescue scene library to perform similarity comparison based on the dynamic rescue scene, and generate a first risk indicator;

[0017] The first warning response information generating module performs a warning response based on the first risk indicator, generates first warning response information and sends it to the rescue command center.

[0018] Through the intelligent protective helmet rapid response method and system based on emergency rescue scenarios proposed in this application, the intelligent sensing module of the first intelligent protective helmet is obtained; the first rescuer is equipped with the first intelligent protective helmet to collect environmental data streams, vital signs data streams, image data streams and location data streams; emergency rescue scene mapping is performed to obtain dynamic rescue scenes; the preset scene features in the emergency rescue scene library are traversed for similarity comparison to generate the first risk index; an early warning response is performed to generate the first early warning response information and send it to the rescue command center. This solves the technical problem that the existing emergency rescue protective helmets are difficult to accurately perceive the complex and changing environmental parameters and vital signs of personnel in real time, resulting in the inaccurate and rapid identification of danger signals, which makes the protective helmets respond poorly, and achieves the technical effect of faster response of intelligent protective helmets during emergency rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0020] Figure 1 A schematic diagram of a quick response method for an intelligent protective helmet based on an emergency rescue scenario provided in an embodiment of the present application;

[0021] Figure 2A schematic diagram of the structure of a rapid response system for an intelligent protective helmet based on an emergency rescue scenario provided in an embodiment of the present application.

[0022] Explanation of the accompanying drawings: intelligent sensor module acquisition module 10, data stream acquisition module 20, emergency rescue scene mapping module 30, first risk indicator generation module 40, first warning response information generation module 50. DETAILED DESCRIPTION

[0023] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0024] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0025] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0026] The present application embodiment provides a quick response method for a smart protective helmet based on an emergency rescue scenario, such as Figure 1 As shown, the method includes:

[0027] Step S100, obtaining the intelligent sensor module of the first intelligent protective helmet, wherein the intelligent sensor module includes an environment monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module. The intelligent sensor module of the first intelligent protective helmet is a module that integrates multiple sensors and technologies, and is used to collect and process various data related to the helmet wearer and its surrounding environment. The intelligent sensor module includes an environment monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module. Specifically, the environment monitoring module includes a temperature sensor, a humidity sensor, an air pressure sensor (measuring altitude or air pressure changes), a gas sensor (detecting harmful gas concentrations), etc., which are used to monitor and evaluate the environmental conditions around the helmet wearer; the vital sign monitoring module includes a heart rate sensor, a body temperature sensor, a blood oxygen saturation sensor, etc., which are used to continuously monitor and record the wearer's vital sign data; the image monitoring module includes a high-definition camera and an infrared camera, etc., and captures images and videos within the wearer's field of view through image sensors such as cameras; the positioning module includes GPS (Global Positioning System), Beidou (Beidou Satellite Navigation System), etc., to accurately locate the position of the protective helmet wearer. These modules are integrated and interacted through components such as circuit boards, microprocessors, memories and communication interfaces inside the helmet to achieve data collection, processing, transmission and storage.

[0028] Step S200: Put the first intelligent protective helmet on the first rescuer, and collect environmental data stream, vital sign data stream, image data stream and position data stream through the intelligent sensor module. When the first rescuer wears the first intelligent protective helmet, through its built-in intelligent sensor module, multiple data streams can be collected and transmitted in real time, including environmental data streams, vital signs data streams, image data streams and position data streams. Specifically, the environmental data stream includes temperature, humidity, air pressure, concentration of harmful gases, etc., which reflects the environmental conditions of the rescue site in real time; the vital signs data stream includes heart rate, blood oxygen saturation, body temperature, respiratory rate, etc., which can reflect the physical condition and health status of the rescuer, and help to timely discover potential health problems or physical fatigue, thereby ensuring the safety of the rescuer; the image data stream is transmitted in the form of video stream or image sequence, which can reflect the actual situation of the rescue site in real time, including the disaster site conditions, the location of trapped persons, the progress of rescue, etc., to provide support for remote command and decision-making; the data of the position data stream is presented in the form of latitude and longitude, altitude, speed, etc., which can reflect the movement trajectory and location information of the rescuer in real time, which helps the command center to grasp the location and distribution of the rescuers and conduct resource scheduling and coordination.

[0029] Step S300, perform emergency rescue scene mapping based on the environmental data stream, the vital signs data stream, the image data stream and the position data stream to obtain a dynamic rescue scene. By performing emergency rescue scene mapping through the environmental data stream, the vital signs data stream, the image data stream and the position data stream, a digital model or visualization interface that dynamically reflects the current emergency rescue scene is constructed and updated, so as to more intuitively and comprehensively understand the actual situation at the rescue site. Specifically, the environmental data stream (such as temperature, humidity, air pressure, concentration of harmful gases, etc.) will be converted into digital or graphic form and displayed on a digital map of the rescue scene or on a special monitoring interface; the vital signs data stream (such as heart rate, blood oxygen saturation, body temperature, respiratory rate, etc.) can be directly displayed to the rescue personnel through the built-in display screen of the helmet, or transmitted through wireless communication. Go to the command center to let the command personnel know the health status of the rescuers; the image data stream is captured by the camera of the helmet and transmitted to the command center or the rescuer's display screen in real time, which can intuitively show the actual situation at the rescue site, including the location of the trapped people, the degree of damage caused by the disaster, the progress of the rescue, etc.; the location data stream (such as GPS coordinates, movement trajectory, etc.) will be used to mark the location of the rescuers and equipment on the digital map in real time. The command center can clearly understand the distribution of rescue forces and conduct resource scheduling and coordination. At the same time, rescuers can also understand each other's location through the location data stream and collaborate more effectively. Combining the above four aspects of data streams, a dynamic and comprehensive emergency rescue scene mapping can be constructed to better understand the on-site situation and improve rescue efficiency and safety.

[0030] In a possible implementation, step S300 further includes step S310, performing static indicator identification on the environmental data stream and the vital signs data stream to obtain a static factor state. Among them, static indicators usually refer to statistical indicators calculated according to a unified caliber at a specific time point or a fixed period, reflecting a certain condition at a certain time point, usually called a time point indicator. When static indicator identification is performed on the environmental data stream and the vital signs data stream, the system will analyze and process these real-time collected data to extract key features or parameters, representing the static state of a specific time point or time period, that is, the static factor state. For example, for the environmental data stream, a pattern recognition algorithm is used to detect whether parameters such as temperature and humidity exceed a preset safety threshold; for the vital signs data stream, an anomaly detection algorithm is used to identify whether the rescuer has abnormal heart rate, decreased blood oxygen saturation, and other unstable vital signs. It also includes step S320, performing dynamic indicator identification on the environmental data stream and the vital signs data stream to obtain a dynamic factor state. Dynamic indicators usually refer to certain changes that occur within a certain period of time, and can also be called period indicators. When dynamic indicators are identified, the environmental data stream and the vital signs data stream are analyzed in real time, the values ​​of these parameters at different time points are extracted, and their changes over time are observed. For example, the dynamic characteristics such as the change trend, fluctuation cycle, and abnormal changes of these data are identified through time series analysis and trend prediction models, so as to obtain the dynamic factor state, which reflects the dynamic changes of environmental conditions and vital signs over a period of time. For example, if it is identified that the concentration of harmful gases in the environment is rising rapidly, the rescue command center will immediately take measures to prevent the spread of harmful gases and protect the safety of rescuers. If it is identified that the heart rate of a rescuer continues to rise, the command center will require him to suspend work and receive medical examination. It also includes step S330, performing moving target identification based on the image data stream to generate a moving target state, wherein the moving target state includes the scale and drop frequency of the moving target. Use background subtraction, inter-frame difference, optical flow method, deep learning, etc. to identify moving targets, identify and extract moving targets in images, and generate moving target states, including the scale and drop frequency of moving targets. Specifically, the position, shape, size and other scale information of the moving target are detected by calculating the number of target pixels, the size of the bounding box, etc. The drop frequency refers to the frequency of the moving target falling from a certain position or height and appearing in the image, that is, detecting whether there is a new falling target in each frame of the image, and counting the number of falling targets within a period of time, which is used to evaluate the falling speed and density of the target, so as to help judge the situation and possible risks on the scene. It also includes step S340, forming a scene factor state with the static factor state, the dynamic factor state and the moving target state.The static factor state, the dynamic factor state and the moving target state are integrated together to form a comprehensive state that fully describes a specific scene or environment, namely, the scene factor state. It also includes step S350, performing scene destruction simulation based on the position data stream to obtain a destruction simulation result. The collected position data stream (usually including the location information of rescue personnel, equipment or other key elements) is used to simulate and predict the destruction or effect that may occur in the emergency rescue scene. For example, if a fire or explosion occurs in a certain area, the impact of the spread of the fire or the explosion shock wave on the surrounding environment and personnel is simulated according to factors such as the fire or explosion range, wind direction, etc., to obtain a destruction simulation result, which may include an image or video of the destroyed area, a list of damaged equipment, the number of affected personnel, etc. It also includes step S360, constructing the dynamic rescue scene with the destruction simulation result and the scene factor state. Combine the simulated destruction and the scene factor information collected in real time to dynamically and in real time construct and update a virtual scene that reflects the actual situation of the current rescue scene to obtain a dynamic rescue scene.

[0031] In a possible implementation, step S330 further includes step S331, extracting images from the image data stream and performing downsampling processing according to the first step length to obtain a first extracted image set. Extracting images from the image data stream according to the preset first step length (i.e., time interval or frame number interval), for example, if the first step length is set to extract 1 frame per second, the system will extract an image from the data stream every 1 second. Downsampling (also known as downsampling) is an image processing technology used to reduce the resolution or size of an image. Downsampling processing is performed on the extracted image through nearest neighbor interpolation, bilinear interpolation, etc., to reduce the number of pixels of the image or reduce the resolution of the image. The purpose is to meet the size of a specific display area, generate thumbnails of corresponding images, and reduce the amount of calculation of large images, so as to obtain a set of downsampled images, i.e., the first extracted image set. It also includes step S332, selecting images from the image data stream according to the second step length to obtain a second extracted image set, wherein the second step length is greater than the first step length, and the number of images in the second extracted image set is less than the number of images in the first extracted image set. The method further includes step S333, based on the slowfast network, performing motion feature recognition on the first extracted image set and the second extracted image set, and outputting the state of the moving target. The first extracted image set and the second extracted image set are input into the slowfast network for motion feature recognition, including motion information such as the motion trajectory, speed, acceleration, and the size and shape of the target, so as to achieve a comprehensive understanding and accurate recognition of the moving target, and finally the slowfast network outputs the state of the moving target.

[0032] In a possible implementation, step S333 also includes that the core idea of ​​the SlowFast network is to adopt a dual-path design, namely a slow path and a fast path. The slow path runs at a low frame rate and is mainly responsible for capturing spatial semantic information in the image, such as the shape, color, texture, etc. of the object; the fast path runs at a high frame rate and captures motion information in the image with fine temporal resolution, such as the moving trajectory and speed of the target. Specifically, in the SlowFast network, the slow path and the fast path extract the spatial and temporal features of the image respectively. The slow path adopts a standard convolutional neural network structure and is responsible for extracting the spatial features of the image frame; the fast path focuses on extracting temporal features by reducing the spatial resolution and increasing the frame rate. At the end of the network, the SlowFast network fuses the features extracted by the slow path and the fast path, so that the two paths can complement each other and make full use of their respective advantages, that is, the network can process the spatial and temporal dimensions of the image at the same time to achieve a comprehensive understanding of the moving target.

[0033] In a possible implementation, step S350 further includes step S351, taking the position recognition result corresponding to the position data stream as the center, obtaining the pre-disaster three-dimensional model of the building within a preset range. Specific position information is extracted from the position data stream as the position recognition result, such as the location where an event occurred or the location of a rescuer, and it is used as the center point to obtain a spatial area of ​​a preset range, such as a circular area with the center point as the center and a certain distance as the radius, or a rectangular area with the center point as the vertex, etc., and the pre-disaster three-dimensional model of the building in the range is retrieved and obtained through a geographic information system (GIS) database, a city three-dimensional model library or other relevant data sources, which usually includes detailed information such as the geometric shape, size, height, material, and the spatial relationship between the buildings. It also includes step S352, obtaining the external post-disaster image of the position recognition result through the rescue command center. The rescue command center retrieves and obtains the post-disaster image of the corresponding location from relevant data sources (such as videos taken by drones, satellite images, ground surveillance cameras, etc.) based on the position recognition result. The method further includes step S353, combining the external post-disaster image and the pre-disaster building three-dimensional model to simulate the damage to the building after the disaster, and obtain the damage simulation result. Advanced image processing, three-dimensional modeling and simulation technology are used to simulate and predict the damage to the building after the disaster, and obtain the damage simulation result, which is usually presented in the form of three-dimensional models, images, charts, etc., showing the degree and scope of damage to the building in the disaster. The damage simulation result can also include detailed information such as the collapse direction of the building, the distribution of debris, and the rescue channel.

[0034] Step S400, based on the dynamic rescue scenario, traverse the preset scenario features in the emergency rescue scenario library for similarity comparison, and generate a first risk index. The dynamic rescue scenario refers to a virtual scene that reflects the actual situation of the current rescue scene, which is comprehensively constructed by collecting environmental data streams, vital signs data streams, image data streams, and position data streams in real time through devices such as smart protective helmets. In the emergency rescue scenario library, a large number of typical or possible rescue scenario features are pre-stored, such as environmental parameters (such as temperature, humidity, concentration of harmful gases, etc.), rescuer status, location of trapped persons, disaster type and its impact range, etc. After the dynamic rescue scenario is constructed, the system will perform a similarity comparison between it and the preset scenario features in the emergency rescue scenario library, for example, using a sequence alignment algorithm to find the maximum match or similarity between the two, and based on the result of the similarity comparison, generate a quantitative risk index, i.e., the first risk index, which reflects the similarity between the current dynamic rescue scenario and one or some scene features in the preset scenario library, thereby indirectly reflecting the risk level of the current rescue scenario.

[0035] In a possible implementation, step S400 further includes step S410, obtaining the preset scene features and preset risk indicators based on the dynamic rescue scene. The preset scene features are used to describe the key attributes and states of the scene and predict future trends, and may include scene types (such as cities, mountains, waters, etc.), disaster types (such as earthquakes, floods, fires, etc.), building structures, infrastructure conditions, weather conditions, road capacity, and other aspects; the preset risk indicators are standards for quantitatively evaluating potential risks in rescue operations, and may include casualty risks, property loss risks, rescue operation failure risks, etc. It also includes step S420, performing a similarity comparison between the dynamic rescue scene and the preset scene features to generate a similarity index. Compare the current rescue scene with the preset ideal scene. Specifically, extract key feature information from the dynamic rescue scene, match the extracted dynamic rescue scene features with the preset scene features through histogram calculation, hash method, cosine similarity calculation, etc., calculate the similarity between the dynamic rescue scene and the preset scene features, which is called the similarity index. The size of the similarity index reflects the closeness of the two scenes in terms of features. The larger the value, the more similar, and the smaller the value, the greater the difference. It also includes step S430. If the similarity index is greater than the preset similarity, the preset risk index of the preset scene feature is tolerantly processed based on the similarity index to generate the first risk index. If the similarity index is greater than the preset similarity, it means that the current dynamic rescue scene and the preset scene feature have a high similarity in key features. The preset risk index of the preset scene feature is tolerantly processed based on the similarity index, that is, the preset risk index is appropriately adjusted to make it more consistent with the actual situation of the current dynamic rescue scene, and a first risk index based on the current dynamic rescue scene is obtained.

[0036] In a possible implementation, step S410 further includes step S411, wherein the emergency rescue scenario library includes a plurality of scenario samples, wherein any scenario sample includes a scenario destruction simulation sample, a scenario factor state sample, and a preset risk index, wherein the preset risk index is calibrated by an expert system. In the described emergency rescue scenario library, a plurality of scenario samples are included, each of which describes in detail the emergency rescue situation under a specific situation, and the key components include a scenario destruction simulation sample, a scenario factor state sample, and a preset risk index. Specifically, the scenario destruction simulation sample is usually based on a physical model, a mathematical model, or a computer simulation technology, and is the result of simulating the possible destruction under a specific emergency rescue scenario; the scenario factor state sample describes the state information of the scenario sample at a specific time point, including weather conditions, traffic conditions, personnel distribution, resource availability, and other factors; the preset risk index is calibrated by an expert system, and specifically, by applying artificial intelligence technology and reasoning technology, the expert system can perform qualitative and quantitative evaluations of the risks in the scenario sample based on preset algorithms and rules, combined with a large amount of professional knowledge and experience, thereby generating accurate preset risk indicators. The method further includes step S412, extracting a scene destruction simulation sample and a scene factor state sample of any scene sample from the multiple scene samples as the preset scene feature. When constructing an emergency rescue scene, a specific scene sample is selected from the multiple pre-stored scene samples, and two key components of the scene sample are extracted from the scene sample: the scene destruction simulation sample and the scene factor state sample, as the preset scene feature.

[0037] Step S500, based on the first risk indicator, a warning response is performed, and the first warning response information is generated and sent to the rescue command center. According to the evaluation result of the first risk indicator, the warning response is triggered. If the risk indicator exceeds the preset threshold value, or reaches the level that requires specific measures to be taken, the warning response is initiated and the first warning response information is generated, which usually includes the description of the current rescue scene, the specific value of the risk indicator, the recommended response measures and other key contents. After the first warning response information is generated, the system will send this information to the rescue command center through a wireless communication network or other communication methods. After receiving this information, the rescue command center can quickly understand the current rescue scene situation, and adjust the rescue strategy and dispatch rescue resources as needed, so as to realize the informatization and intelligence of emergency rescue, significantly improve the rescue efficiency, reduce the rescue risk, and provide timely and accurate information support for rescue personnel.

[0038] In a possible implementation, step S500 further includes step S510, wherein the first smart protective helmet communicates data with the rescue command center through the first communication channel, and when the first communication channel sends a communication interruption reminder signal, the risk indicator sequence and interruption location information sent by the first smart protective helmet to the rescue command center are obtained. The first smart protective helmet communicates data with the rescue command center through the first communication channel, and when the first communication channel sends a communication interruption reminder signal due to network failure, signal interference, equipment failure, etc., the risk indicator sequence and interruption location information sent by the first smart protective helmet to the rescue command center are obtained. Specifically, the risk indicator sequence reflects the environmental conditions, potential risks and other information at the location of the helmet, and the interruption location information generally includes the specific location where the communication interruption occurs (such as network node, equipment location, etc.) and possible causes of the failure. It also includes step S520, which performs risk prediction in a preset time zone based on the risk indicator sequence to generate a risk prediction time series. Use historical risk indicator data (i.e., risk indicator sequence) to predict the risk level in a preset time period (i.e., preset time zone) in the future, and arrange these prediction results in chronological order to form a risk prediction time series, for example, use an autoregressive model (AR model), a moving average model (MA model), an autoregressive moving average model (ARIMA model), etc., to predict the risks and challenges that may be faced in the future period of time based on the patterns and trends in historical data. It also includes step S530, generating an emergency response decision based on the interruption location information and the risk prediction time series. Emergency rescue strategies and action plans are formulated and adjusted based on this information. Specifically, the interruption location information can help the rescue command center quickly determine the specific location and possible causes of the communication interruption. Based on the prediction results provided by the risk prediction time series, the rescue command center can evaluate the priority, resource allocation and action route of the current rescue operation. For example, if it is predicted that the risk level in a certain area will increase significantly, the rescue command center may increase the rescue force in that area, adjust the allocation of rescue resources, and formulate corresponding emergency evacuation or dispersal plans. Finally, combined with the interruption location information and the risk prediction time series, the rescue command center formulates a comprehensive emergency response decision, which may include specific plans to repair communication failures, detailed steps to adjust the rescue action plan, specific arrangements for deploying rescue resources, etc.

[0039] In the above, refer to Figure 1 The quick response method of the smart protective helmet based on the emergency rescue scenario according to the embodiment of the present invention is described in detail. Figure 2 The invention describes an intelligent protective helmet rapid response system based on emergency rescue scenarios according to an embodiment of the invention.

[0040] According to the embodiment of the present invention, the intelligent protective helmet rapid response system based on the emergency rescue scenario is used to solve the technical problem that the existing emergency rescue protective helmets are difficult to accurately perceive the complex and changeable environmental parameters and personnel vital signs in real time, resulting in the inaccurate and rapid judgment of danger signals, which makes the protective helmets respond poorly, and achieves the technical effect of faster response of the intelligent protective helmets during emergency rescue. The intelligent protective helmet rapid response system based on the emergency rescue scenario includes: an intelligent sensor module acquisition module 10, a data stream acquisition module 20, an emergency rescue scenario mapping module 30, a first risk indicator generation module 40, and a first warning response information generation module 50.

[0041] An intelligent sensor module acquisition module 10, wherein the intelligent sensor module acquisition module 10 is used to acquire an intelligent sensor module of a first intelligent protective helmet, wherein the intelligent sensor module includes an environment monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module;

[0042] A data stream collection module 20, wherein the data stream collection module 20 is used to collect environmental data streams, vital sign data streams, image data streams and position data streams through the intelligent sensor module when the first rescuer wears the first intelligent protective helmet;

[0043] An emergency rescue scene mapping module 30, wherein the emergency rescue scene mapping module 30 performs emergency rescue scene mapping based on the environmental data stream, the vital sign data stream, the image data stream and the position data stream to obtain a dynamic rescue scene;

[0044] A first risk indicator generating module 40, which is used to traverse the preset scene features in the emergency rescue scene library to perform similarity comparison based on the dynamic rescue scene, and generate a first risk indicator;

[0045] The first warning response information generating module 50 performs a warning response based on the first risk indicator, generates first warning response information and sends it to the rescue command center.

[0046] The specific configuration of the emergency rescue scene mapping module 30 will be described in detail below. The emergency rescue scene mapping module 30 further includes: performing static indicator identification on the environmental data stream and the vital signs data stream to obtain a static factor state; performing dynamic indicator identification on the environmental data stream and the vital signs data stream to obtain a dynamic factor state; performing moving target identification based on the image data stream to generate a moving target state, wherein the moving target state includes the scale and drop frequency of the moving target; forming a scene factor state with the static factor state, the dynamic factor state and the moving target state; performing scene destruction simulation based on the position data stream to obtain a destruction simulation result; and constructing the dynamic rescue scene with the destruction simulation result and the scene factor state.

[0047] The specific configuration of the emergency rescue scene mapping module 30 will be described in detail below. The emergency rescue scene mapping module 30 may further include: extracting images from the image data stream and performing downsampling processing according to the first step length to obtain a first extracted image set; selecting images from the image data stream according to the second step length to obtain a second extracted image set, wherein the second step length is greater than the first step length, and the number of images in the second extracted image set is less than the number of images in the first extracted image set; based on the slowfast network, performing motion feature recognition on the first extracted image set and the second extracted image set, and outputting the motion target state.

[0048] The specific configuration of the emergency rescue scene mapping module 30 will be described in detail below. The emergency rescue scene mapping module 30 may further include: taking the position recognition result corresponding to the position data stream as the center, obtaining a pre-disaster building three-dimensional model within a preset range; obtaining an external post-disaster image of the position recognition result through the rescue command center; combining the external post-disaster image and the pre-disaster building three-dimensional model to simulate the damage of the disaster building and obtain the damage simulation result.

[0049] The specific configuration of the first risk indicator generating module 40 will be described in detail below. The first risk indicator generating module 40 further includes: obtaining the preset scenario feature and the preset risk indicator based on the dynamic rescue scenario; performing a similarity comparison between the dynamic rescue scenario and the preset scenario feature to generate a similarity indicator; if the similarity indicator is greater than the preset similarity, performing a tolerant processing on the preset risk indicator of the preset scenario feature based on the similarity indicator to generate the first risk indicator.

[0050] The specific configuration of the first risk indicator generation module 40 will be described in detail below. The first risk indicator generation module 40 further includes: the emergency rescue scenario library includes multiple scenario samples, wherein any scenario sample includes a scenario destruction simulation sample, a scenario factor state sample and a preset risk indicator, wherein the preset risk indicator is calibrated by an expert system; and the scenario destruction simulation sample and the scenario factor state sample of any scenario sample are extracted from the multiple scenario samples as the preset scenario features.

[0051] The specific configuration of the first warning response information generation module 50 will be described in detail below. The first warning response information generation module 50 may further include: the first smart protective helmet communicates data with the rescue command center through the first communication channel, when the first communication channel sends a communication interruption reminder signal, obtains the risk indicator sequence and interruption location information sent by the first smart protective helmet to the rescue command center; performs risk prediction in a preset time zone based on the risk indicator sequence, and generates a risk prediction sequence; generates an emergency response decision based on the interruption location information and the risk prediction sequence.

[0052] The intelligent protective helmet rapid response system based on emergency rescue scenarios provided in an embodiment of the present invention can execute the intelligent protective helmet rapid response method based on emergency rescue scenarios provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0053] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0054] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. Rapid response method of intelligent protective helmet based on emergency rescue scenario, It is characterized in that The method comprises: Acquire an intelligent sensor module of a first intelligent protective helmet, wherein the intelligent sensor module includes an environment monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module; Putting the first intelligent protective helmet on the first rescuer, and collecting environmental data stream, vital sign data stream, image data stream and position data stream through the intelligent sensor module; Perform emergency rescue scene mapping based on the environmental data stream, the vital sign data stream, the image data stream, and the position data stream to obtain a dynamic rescue scene; Based on the dynamic rescue scenario, traverse the preset scenario features in the emergency rescue scenario library to perform similarity comparison and generate a first risk indicator; Perform an early warning response based on the first risk indicator, generate first early warning response information and send it to the rescue command center; The step of performing emergency rescue scene mapping based on the environment data stream, the vital sign data stream, the image data stream, and the position data stream to obtain a dynamic rescue scene includes: Performing static indicator identification on the environmental data stream and the vital sign data stream to obtain a static factor state; Performing dynamic index identification on the environmental data stream and the vital sign data stream to obtain a dynamic factor state; Performing moving target recognition based on the image data stream to generate a moving target state, wherein the moving target state includes a scale and a drop frequency of the moving target; The static factor state, the dynamic factor state and the moving target state are used to form a scene factor state; Performing scene destruction simulation based on the position data stream to obtain a destruction simulation result; Constructing the dynamic rescue scenario based on the destruction simulation result and the scenario factor state; Wherein, the moving target is identified based on the image data stream to generate the moving target state, wherein the moving target state includes the scale and drop frequency of the moving target, including: According to the first step, images in the image data stream are extracted and down-sampled to obtain a first extracted image set; selecting images from the image data stream according to a second step length to obtain a second extracted image set, wherein the second step length is greater than the first step length, and the number of images in the second extracted image set is less than the number of images in the first extracted image set; Based on the slowfast network, performing motion feature recognition on the first extracted image set and the second extracted image set, and outputting the motion target state; Wherein, performing scene destruction simulation based on the position data stream to obtain a destruction simulation result includes: Taking the position recognition result corresponding to the position data stream as the center, obtaining a three-dimensional model of the building before the disaster within a preset range; Acquiring an external post-disaster image of the location recognition result through the rescue command center; The disaster building damage simulation is performed by combining the external post-disaster image and the pre-disaster building three-dimensional model to obtain the damage simulation result.

2. The intelligent protective helmet rapid response method based on emergency rescue scenarios as claimed in claim 1, It is characterized in that Based on the dynamic rescue scenario, traversing the preset scenario features in the emergency rescue scenario library for similarity comparison to generate a first risk indicator, including: Acquire the preset scenario characteristics and preset risk indicators based on the dynamic rescue scenario; Performing a similarity comparison between the dynamic rescue scene and the preset scene features to generate a similarity index; If the similarity index is greater than a preset similarity, a tolerance processing is performed on the preset risk index of the preset scene feature based on the similarity index to generate the first risk index.

3. The intelligent protective helmet rapid response method based on emergency rescue scenario as claimed in claim 2, It is characterized in that Acquiring the preset scene feature based on the dynamic rescue scene includes: The emergency rescue scenario library includes multiple scenario samples, wherein any scenario sample includes a scenario destruction simulation sample, a scenario factor state sample and a preset risk index, wherein the preset risk index is calibrated by an expert system; A scene destruction simulation sample and a scene factor state sample of any scene sample are extracted from the multiple scene samples as the preset scene features.

4. The intelligent protective helmet rapid response method based on emergency rescue scenarios as claimed in claim 1, It is characterized in that The method further comprises: The first smart protective helmet communicates data with the rescue command center through a first communication channel, and when the first communication channel sends a communication interruption reminder signal, the risk indicator sequence and interruption location information sent by the first smart protective helmet to the rescue command center are obtained; Perform risk prediction in a preset time zone based on the risk indicator sequence to generate a risk prediction time series; An emergency response decision is generated based on the interruption location information and the risk prediction time sequence.

5. Intelligent protective helmet rapid response system based on emergency rescue scenarios, It is characterized in that The system is used to implement the intelligent protective helmet rapid response method based on emergency rescue scenarios according to any one of claims 1 to 4, and the system includes: An intelligent sensor module acquisition module, the intelligent sensor module acquisition module is used to acquire the intelligent sensor module of the first intelligent protective helmet, wherein the intelligent sensor module includes an environment monitoring module, a vital sign monitoring module, an image monitoring module and a positioning module; A data stream collection module, wherein the data stream collection module is used to collect environmental data streams, vital sign data streams, image data streams and position data streams through the intelligent sensor module when the first rescuer wears the first intelligent protective helmet; An emergency rescue scene mapping module, which performs emergency rescue scene mapping based on the environmental data stream, the vital sign data stream, the image data stream and the position data stream to obtain a dynamic rescue scene; A first risk indicator generation module, the first risk indicator generation module is used to traverse the preset scene features in the emergency rescue scene library to perform similarity comparison based on the dynamic rescue scene, and generate a first risk indicator; The first warning response information generating module performs a warning response based on the first risk indicator, generates first warning response information and sends it to the rescue command center.

Citation Information

Patent Citations

  • Movable modular intelligent fire-fighting on-duty guarantee equipment and related products

    CN109920099A

  • 5G fire rescue method and fire rescue helmet

    CN112107808A

  • Power operation safety monitoring and intelligent risk identification method based on Internet of Things

    CN113986006A

  • Geological disaster monitoring and early warning method based on big data and image recognition

    CN114783149A

  • Fire-fighting emergency rescue simulation training system based on VR

    CN118253060A