Help-seeking interaction system using two-dimensional code multi-dimensional positioning
Through the QR code multi-dimensional positioning and help-seeking interactive system, combined with the disaster detection model and Bayesian network, the problem of inefficient rescue of trapped people in tunnel fires is solved, and the accurate identification and rescue plan for fire locations and disasters is achieved. Optimization improves rescue efficiency and safety.
Patent Information
- Application Number
- CN202410904212.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology cannot effectively solve the problem of rescue of trapped people in tunnel fire scenarios, especially in determining the location of the fire and the fire situation when rescuers arrive, resulting in inefficient rescue.
A QR code multi-dimensional positioning and help-seeking interactive system is designed, including a call task generation unit, a rescue personnel information collection unit, a disaster detection and identification unit, a disaster situation scenario deduction unit and a rescue task execution unit, and a rescue task execution unit are located through a QR code multi-dimensional positioning. , disaster detection model and Bayesian network technologies can realize the identification and deduction of fire location and disaster scale, and optimize the rescue plan and process.
The work efficiency of the rescue mission is improved, the safety of the lives of trapped people is ensured, and the accuracy and efficiency of the rescue process are improved through accurate dispatch of recruitment tasks, disaster identification and rescue plans.
Smart Images

Figure CN119067151A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of help-seeking interaction, and in particular to a help-seeking interaction system utilizing multi-dimensional positioning of a QR code. Background Art
[0002] In recent years, my country's highway network construction has made great progress, and my country's highway tunnel construction has also flourished. However, the number of safety accidents in my country's tunnels during operation and service has increased. The Chinese patent application number 201910947444.9 discloses "a wireless intelligent positioning assistance alarm public service system, including wireless intelligent terminals, centralized control centers, command centers and help centers, etc.; the centralized control center, command center and help center in the system can be combined and applied with each other as needed, or they can be used separately. The alarmer can use the relevant APP to initiate an alarm signal with one click on the wireless intelligent terminal, and send it to the relevant command center through the wireless communication network. The command center can accurately locate the alarmer through GPIS or other real-time map platforms, and conduct audio and video intercom with the alarmer, and quickly arrange for police handling according to the alarm situation. The wireless intelligent terminal can also use SOS to ask for help or alarm with one click.";
[0003] This prior art only solves the existing public communication service system, which is quite outdated and completely unable to meet the current social needs for efficient development. It does not take into account the need to determine the location of the tunnel fire and the fire situation when the rescuers arrive when rescuing trapped people in a tunnel fire scenario, so as to facilitate the rescuers to quickly carry out rescue tasks when they arrive, thereby improving the efficiency of rescuing trapped people. Summary of the invention
[0004] The purpose of the present invention is to provide a multi-dimensional positioning assistance interaction system using QR codes to solve the problems raised in the above background technology.
[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a QR code multi-dimensional positioning help interaction system, including a call for help task generation unit, taking the QR code multi-dimensional positioning help interaction system in a tunnel fire scene as an example, the call for help task generation unit is used for trapped personnel in the tunnel fire scene to log in to the call for help software through a mobile terminal and submit relevant call for help demand information, and then generate a call for help task based on the submitted call for help demand information;
[0006] A rescuer information collection unit, the rescuer information collection unit is used to collect and record relevant information of the rescuers who have registered the rescue calling software, and regularly update the relevant information of the rescuers;
[0007] A disaster detection and identification unit, which is used to distribute the generated rescue mission based on the real-time relevant information of the rescue personnel, deploy relevant disaster detection equipment in the tunnel fire scene, and perform disaster detection and identification operations based on the relevant disaster data obtained through the constructed disaster detection model;
[0008] A disaster scenario deduction unit, which is used to process the obtained relevant disaster data, and obtain relevant disaster data after the development of the tunnel fire based on the processed relevant disaster data through a constructed disaster deduction model, thereby realizing the deduction of the situation and consequences of the tunnel fire disaster;
[0009] A rescue mission execution unit is used to select a rescue plan based on the tunnel fire disaster simulation results after the rescue personnel arrive, and to execute the selected rescue plan and record the rescue action process.
[0010] Preferably, the rescue task generation unit includes a user login module, a demand information submission module and a rescue task construction module. The user login module is used for trapped persons in a tunnel fire scenario to log in to the rescue software through a mobile terminal. The demand information submission module is used for the trapped persons to submit rescue demand information by scanning the rescue QR code in the rescue software after successfully logging in to the rescue software, including information on the number of trapped persons, trapped location information, captured surrounding video image information and trapped time information. The rescue task construction module is used for the rescue software to generate a corresponding rescue task based on the rescue demand information submitted by the trapped persons.
[0011] Preferably, the rescuer information collection unit includes a rescuer information recording module and a rescuer information updating module. The rescuer information recording module is used to collect and record relevant information of the rescuers who have registered with the rescue calling software, including rescuer scale information, rescuer execution rescue mission type information, rescuer execution mission number information, rescuer location information and rescuer work status information. The database of the rescue calling software stores previous rescuer execution records and rescue plans under different circumstances in various rescue scenarios. The rescuer information updating module is used to regularly update relevant information of the rescuers and data stored in the database. Rescuers can also update information by submitting the information that needs to be modified by themselves and verifying it through the rescue calling software administrator.
[0012] Preferably, the disaster detection and identification unit includes a rescue task distribution module and a disaster detection equipment deployment module. The rescue task distribution module is used to perform rescue task distribution operations on the generated rescue tasks based on the real-time relevant information of the rescue personnel, and generate the estimated time for the rescue personnel to arrive at the tunnel fire site. The disaster detection equipment deployment module is used to deploy relevant disaster detection equipment in the tunnel fire scene. The specific operation is to arrange vehicle detectors in front of each tunnel, obtain the number of vehicles between each tunnel through the detection data, and then obtain the total number of vehicles in the tunnel, and arrange ultrasonic wind speed and direction detectors at the entrance of the tunnel to obtain real-time wind speed and wind direction data in the tunnel, and place multiple temperature sensors at fixed intervals at the tunnel vault position, so that the detection area of the temperature sensor covers the entire tunnel area, so as to obtain the ambient temperature of each area of the tunnel and the tunnel vault temperature, and arrange cameras in front of each tunnel to collect fire images and vehicle images in each tunnel.
[0013] Preferably, the disaster detection and identification unit includes a disaster detection model construction module, which is used to construct a disaster detection model based on YOLO V5s, and specifically includes four parts: an input end, a backbone network, a neck network, and an output end. The input end is used to perform data enhancement operations on the input image through a Mosaic method to achieve the method of small targets in the input image, thereby improving the detection capability of the disaster detection model for small targets. The backbone network is used to perform feature extraction operations on the input image, split the high-resolution image in the input image into multiple low-resolution images, expand the input image in the RGB three-channel image, and achieve a two-fold downsampling operation while ensuring that the image information is not lost. The split image is convolved to obtain a feature map, and the feature map is subjected to a maximum pooling layer operation, and images with different receptive fields are synthesized to obtain a final feature map, thereby improving the recognition accuracy of the disaster detection model for small targets. The neck network is used to integrate and fuse the obtained feature maps, and the output end is used to convert the processed feature maps into the required detection results.
[0014] Preferably, the disaster detection and identification unit also includes a disaster detection model training module, which is used to obtain tunnel fire images and images of different types of vehicles in the tunnel through a search engine, and construct a tunnel fire data set based on this. When the number of samples in the tunnel fire data set is insufficient, the data enhancement operation of the tunnel fire data set can be achieved by flipping the acquired images at multiple angles, and the tunnel fire images and tunnel vehicle images in the tunnel fire data set are labeled through the LabelImg tool. The disaster detection model is trained through the tunnel fire data set, and the trained disaster detection model is verified for accuracy. Fire identification and accident vehicle identification operations are performed based on the acquired fire images, vehicle images and images submitted by trapped persons through the disaster detection model that meets the model accuracy requirements, thereby determining the fire location.
[0015] Preferably, the disaster scenario deduction unit includes a disaster data processing module and a disaster deduction model construction module. The disaster data processing module is used to process the obtained relevant disaster data. The specific operation is to calculate the fire heat release rate caused by the accident vehicle through a fire heat release algorithm. The specific calculation formula is as follows:
[0016]
[0017] In the formula, Q * represents the fire heat release rate, ΔT max represents the tunnel vault temperature, H f It represents the position from the fire location to the tunnel vault, and based on the calculation result and the type of accident vehicle, the calculation result is compared with the maximum heat release rate of combustion corresponding to the type of accident vehicle. When 0% < the ratio result ≤ 25%, it is judged as a small fire scale, when 25% < the ratio result ≤ 50%, it is judged as a medium fire scale, and when 50% < the ratio result ≤ 100%, it is judged as a large fire scale. The disaster simulation model construction module is used to construct a disaster simulation model based on a Bayesian network, and five factors, namely, tunnel ambient temperature, tunnel real-time wind speed, accident vehicle type, fire scale and number of vehicles in the tunnel, are selected as inputs of the disaster simulation model.
[0018] Preferably, the disaster scenario simulation unit further includes a simulation result output module and a disaster scale judgment module. The simulation result output module is used to obtain the fire scale at the estimated time when rescue personnel arrive at the fire site in the tunnel based on the input tunnel ambient temperature, tunnel real-time wind speed, accident vehicle type, fire scale and number of vehicles in the tunnel through the disaster simulation model. The disaster scale judgment module is used to judge the severity of the disaster from two aspects, namely, the smoke recirculation length and the smoke layer thickness, based on the fire scale result output by the disaster simulation model. Specifically, the smoke recirculation length is calculated through the smoke recirculation algorithm. The specific calculation formula is as follows:
[0019]
[0020] In the formula, represents the average hydraulic diameter of the tunnel, L represents the smoke return length, V * It represents the dimensionless tunnel ambient wind speed, and divides the smoke reflow length into four levels: none, [0m, 50m), [50m, 100m) and above 100m, which correspond to smoke reflow length level one to level four from low to high, and then the smoke reflow length level is obtained based on the calculation result, and the smoke layer thickness is calculated by the smoke layer thickness algorithm. The specific calculation formula is as follows:
[0021]
[0022] Where d represents the thickness of the smoke layer, m p represents the smoke mass floating rate, B / H represents the aspect ratio of the tunnel, and the smoke layer thickness is divided into four levels: none, [0m, 1.5m), [1.5m, 3m) and above 3m, corresponding to smoke layer thickness levels one to four from low to high, respectively. Then, the smoke layer thickness level is obtained based on the calculation results, and the smoke reflow length level and the smoke layer thickness level are averaged to obtain the disaster severity level when the rescue personnel arrive.
[0023] Preferably, the rescue mission execution unit includes a rescue plan selection module, a rescue plan execution module and a rescue mission recording module. The rescue plan selection module is used to call the rescue plan of the corresponding disaster severity level in the tunnel fire scenario from the database of the rescue calling software based on the disaster severity level obtained in the disaster scale judgment module after the rescue personnel arrive, and when the disaster severity level is not an integer, the tunnel fire situation corresponding to the disaster severity level is regarded as the tunnel fire situation corresponding to the disaster severity level of an adjacent and higher integer level for processing. The rescue plan execution module is used for the rescue personnel to perform the rescue mission based on the called rescue plan, so as to rescue the trapped persons. The rescue mission recording module is used for the rescue personnel to record the whole process of the rescue mission and upload it to the rescue calling software.
[0024] Compared with the prior art, the beneficial effects of the present invention at least include: the present invention proposes a multi-dimensional positioning help-seeking interactive system using a QR code, wherein a call for help task generation unit is provided to generate a call for help task based on information submitted by the trapped person, and the relevant information of the rescuer is collected and regularly updated by the provided rescuer information collection unit, and the rescuer can also update the information by submitting the information that needs to be modified by himself and verifying it by the call for help software manager, thereby ensuring the accuracy of the subsequent distribution of the call for help task, and the detection and identification of the fire is realized by the provided disaster detection and identification unit, thereby determining the accident vehicle and the fire location, and the fire scale is deduced by the provided disaster scenario deduction unit when the rescuer arrives at the tunnel fire site at the estimated time, and the severity of the disaster is determined based on the deduction result, thereby facilitating the rescuer to quickly determine the rescue plan and perform the rescue task when the rescuer arrives, thereby improving the work efficiency of the rescue task and ensuring the life safety of the trapped person, and determining the rescue plan and performing the rescue work by the provided rescue task execution unit, and recording the entire rescue work process. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention;
[0026] Figure 2 This is a schematic diagram of the process of generating a rescue task unit according to an embodiment of the present invention;
[0027] Figure 3 This is a schematic diagram of the process of the rescuer information collection unit according to an embodiment of the present invention;
[0028] Figure 4 This is a schematic diagram of the process of the disaster detection and identification unit according to an embodiment of the present invention;
[0029] Figure 5 This is a schematic diagram of the process flow of a disaster scenario deduction unit according to an embodiment of the present invention;
[0030] Figure 6 The figure is a flowchart of a rescue mission execution unit according to an embodiment of the present invention.
[0031] In the figure: 100, rescue task generation unit; 101, user login module; 102, demand information submission module; 103, rescue task construction module; 200, rescue personnel information collection unit; 201, rescue personnel information recording module; 202, rescue personnel information updating module; 300, disaster detection and identification unit; 301, rescue task distribution module; 302, disaster detection equipment deployment module; 303, disaster detection model construction module; 304, disaster detection model training module; 400, disaster scenario simulation unit; 401, disaster data processing module; 402, disaster simulation model construction module; 403, simulation result output module; 404, disaster scale judgment module; 500, rescue task execution unit; 501, rescue plan selection module; 502, rescue plan execution module; 503, rescue task recording module. DETAILED DESCRIPTION
[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0033] See also Figure 1-6 The present invention provides a technical solution: a multi-dimensional positioning help-seeking interactive system using a QR code, including a help-seeking task generation unit 100. Taking the multi-dimensional positioning help-seeking interactive system using a QR code in a tunnel fire scene as an example, the help-seeking task generation unit 100 is used for trapped personnel in the tunnel fire scene to log in to the help-seeking software through a mobile terminal and submit relevant help-seeking demand information, and then generate a help-seeking task based on the submitted help-seeking demand information;
[0034] The rescuer information collection unit 200 is used to collect and record the relevant information of the rescuers who have registered the rescue calling software, and regularly update the relevant information of the rescuers;
[0035] The disaster detection and identification unit 300 is used to distribute the generated rescue mission based on the real-time relevant information of the rescue personnel, deploy relevant disaster detection equipment in the tunnel fire scene, and perform disaster detection and identification operations based on the relevant disaster data obtained through the constructed disaster detection model;
[0036] A disaster scenario deduction unit 400 is used to process the obtained relevant disaster data, and obtain relevant disaster data after the development of the tunnel fire based on the processed relevant disaster data through the constructed disaster deduction model, thereby realizing the deduction of the situation and consequences of the tunnel fire disaster;
[0037] The rescue mission execution unit 500 is used to select a rescue plan based on the tunnel fire disaster simulation results after the rescue personnel arrive, and execute the selected rescue plan and record the rescue action process.
[0038] The rescue task generation unit 100 includes a user login module 101, a demand information submission module 102 and a rescue task construction module 103. The user login module 101 is used for the trapped personnel in the tunnel fire scene to log in to the rescue software through the mobile terminal. The demand information submission module 102 is used for the trapped personnel to submit the rescue demand information by scanning the rescue QR code in the rescue software after successfully logging in to the rescue software, including the number of trapped personnel, trapped location information, captured surrounding video image information and trapped time information. The rescue task construction module 103 is used for the rescue software to generate a corresponding rescue task based on the rescue demand information submitted by the trapped personnel;
[0039] The rescuer information collection unit 200 includes a rescuer information recording module 201 and a rescuer information updating module 202. The rescuer information recording module 201 is used to collect and record relevant information of the rescuers who have registered the rescue calling software, including rescuer size information, rescuer execution rescue mission type information, rescuer execution mission quantity information, rescuer location information and rescuer work status information. The database of the rescue calling software stores the rescue mission execution records of the previous rescuers and rescue plans under various rescue scenarios. The rescuer information updating module 202 is used to regularly update the relevant information of the rescuers and the data stored in the database. The rescuers can also update the information by submitting the information that needs to be modified by themselves and verifying it by the rescue calling software administrator.
[0040] The disaster detection and identification unit 300 includes a rescue task distribution module 301 and a disaster detection equipment deployment module 302. The rescue task distribution module 301 is used to perform rescue task distribution operations on the generated rescue tasks based on the real-time relevant information of the rescue personnel, and generate the estimated time for the rescue personnel to arrive at the tunnel fire site. The disaster detection equipment deployment module 302 is used to deploy relevant disaster detection equipment in the tunnel fire scene. The specific operation is to respectively arrange vehicle detectors in front of each tunnel, obtain the number of vehicles between each tunnel through the detection data, and then obtain the total number of vehicles in the tunnel, and arrange ultrasonic wind speed and wind direction detectors at the entrance of the tunnel to obtain real-time wind speed and wind direction data in the tunnel, and place multiple temperature sensors at fixed intervals at the tunnel vault position, so that the detection area of the temperature sensor covers the entire tunnel area, so as to obtain the ambient temperature of each area of the tunnel and the tunnel vault temperature, and respectively arrange cameras in front of each tunnel to collect fire images and vehicle images in each tunnel.
[0041] The disaster detection and identification unit 300 includes a disaster detection model construction module 303, which is used to construct a disaster detection model based on YOLO V5s, and specifically includes four parts: an input end, a backbone network, a neck network, and an output end. The input end is used to perform data enhancement operations on the input image through a Mosaic method to achieve a method for small targets in the input image, thereby improving the detection capability of the disaster detection model for small targets. The backbone network is used to perform feature extraction operations on the input image, split the high-resolution image in the input image into multiple low-resolution images, expand the image of the input image in the RGB three channels, and achieve a two-fold downsampling operation without losing image information. The split image is convoluted to obtain a feature map, and the feature map is subjected to a maximum pooling layer operation, and images with different receptive fields are synthesized to obtain a final feature map, thereby improving the recognition accuracy of the disaster detection model for small targets. The neck network is used to integrate and fuse the obtained feature maps, and the output end is used to convert the processed feature maps into the required detection results.
[0042] The disaster detection and identification unit 300 also includes a disaster detection model training module 304, which is used to obtain tunnel fire images and images of different types of vehicles in the tunnel through a search engine, and to construct a tunnel fire data set based on this. When the number of samples in the tunnel fire data set is insufficient, the data enhancement operation of the tunnel fire data set can be realized by flipping the acquired images at multiple angles, and the tunnel fire images and tunnel vehicle images in the tunnel fire data set are labeled through the LabelImg tool, and the disaster detection model is trained through the tunnel fire data set, and the trained disaster detection model is verified for accuracy, and the disaster detection model that meets the model accuracy requirements is used to perform fire identification and accident vehicle identification operations based on the acquired fire images, vehicle images, and images submitted by trapped persons, thereby determining the fire location;
[0043] The disaster scenario simulation unit 400 includes a disaster data processing module 401 and a disaster simulation model building module 402. The disaster data processing module 401 is used to process the obtained relevant disaster data. The specific operation is to calculate the fire heat release rate caused by the accident vehicle through a fire heat release algorithm. The specific calculation formula is as follows:
[0044]
[0045] In the formula, Q * represents the fire heat release rate, ΔT max represents the tunnel vault temperature, H f Indicates the position from the fire location to the tunnel vault, and based on the calculation result and the type of accident vehicle, takes the ratio of the calculation result to the maximum heat release rate of combustion corresponding to the type of accident vehicle, when 0%<ratio result≤25%, it is judged as a small fire scale, when 25%<ratio result≤50%, it is judged as a medium fire scale, when 50%<ratio result≤100%, it is judged as a large fire scale, the disaster deduction model construction module 402 is used to construct a disaster deduction model based on the Bayesian network, and selects five factors of tunnel ambient temperature, tunnel real-time wind speed, accident vehicle type, fire scale and number of vehicles in the tunnel as inputs of the disaster deduction model;
[0046] The disaster scenario simulation unit 400 also includes a simulation result output module 403 and a disaster scale judgment module 404. The simulation result output module 403 is used to obtain the fire scale at the estimated time when the rescue personnel arrive at the fire site in the tunnel based on the input tunnel environment temperature, tunnel real-time wind speed, accident vehicle type, fire scale and number of vehicles in the tunnel through the disaster simulation model. The disaster scale judgment module 404 is used to judge the severity of the disaster from two aspects of smoke recirculation length and smoke layer thickness based on the fire scale result output by the disaster simulation model, specifically, the smoke recirculation length is calculated through the smoke recirculation algorithm, and the specific calculation formula is as follows:
[0047]
[0048] In the formula, represents the average hydraulic diameter of the tunnel, L represents the smoke return length, V * It represents the dimensionless tunnel ambient wind speed, and divides the smoke reflow length into four levels: none, [0m, 50m), [50m, 100m) and above 100m, which correspond to smoke reflow length level one to level four from low to high, and then the smoke reflow length level is obtained based on the calculation result, and the smoke layer thickness is calculated by the smoke layer thickness algorithm. The specific calculation formula is as follows:
[0049]
[0050] Where d represents the thickness of the smoke layer, m p represents the smoke mass floating rate, B / H represents the aspect ratio of the tunnel, and the smoke layer thickness is divided into four levels: none, [0m, 1.5m), [1.5m, 3m) and above 3m, which correspond to smoke layer thickness levels one to four from low to high, respectively. Then, the smoke layer thickness level is obtained based on the calculation results, and the smoke reflow length level and the smoke layer thickness level are averaged to obtain the disaster severity level when the rescuers arrive.
[0051] The rescue mission execution unit 500 includes a rescue plan selection module 501, a rescue plan execution module 502 and a rescue mission recording module 503. The rescue plan selection module 501 is used to call the rescue plan of the corresponding disaster severity level in the tunnel fire scenario from the database of the rescue calling software based on the disaster severity level obtained in the disaster scale judgment module 404 after the rescue personnel arrive, and when the disaster severity level is not an integer, the tunnel fire situation corresponding to the disaster severity level is regarded as the tunnel fire situation corresponding to the disaster severity level of an adjacent and higher integer level for processing. The rescue plan execution module 502 is used for the rescue personnel to perform the rescue mission based on the called rescue plan, so as to rescue the trapped persons. The rescue mission recording module 503 is used for the rescue personnel to record the whole process of the rescue mission and upload it to the rescue calling software.
[0052] Working principle: Through the user login module 101, the trapped personnel in the tunnel fire scene log in to the rescue software, and the rescue demand information is submitted through the demand information submission module 102. The rescue task is generated through the rescue task construction module 103. The relevant information of the rescue personnel registered in the rescue software is collected and recorded through the rescue personnel information recording module 201. The relevant information of the rescue personnel and the data stored in the database are regularly updated through the rescue personnel information update module 202. The rescue task distribution module 301 distributes the generated rescue task based on the real-time relevant information of the rescue personnel. The relevant disaster detection equipment is deployed in the tunnel fire scene through the disaster detection equipment deployment module 302. The disaster detection model construction module 303 builds a disaster detection model. The detection model training module 304 trains the disaster detection model, and performs fire identification and accident vehicle identification operations to determine the location of the fire. The disaster data processing module 401 processes the obtained relevant disaster data, and the disaster simulation model construction module 402 constructs a disaster simulation model. The simulation result output module 403 uses the disaster simulation model to obtain the scale of the fire when the rescue personnel are expected to arrive at the tunnel fire site. The disaster scale judgment module 404 judges the severity of the disaster, and the rescue plan selection module 501 calls the rescue plan of the corresponding disaster severity level in the tunnel fire scenario. The rescue plan execution module 502 carries out the rescue task based on the called rescue plan, and the rescue task recording module 503 records the entire process of this rescue task.
[0053] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0054] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A multi-dimensional positioning assistance interactive system using QR codes, characterized by: It comprises a rescue task generation unit (100). Taking a multi-dimensional positioning help-seeking interactive system using a QR code in a tunnel fire scene as an example, the rescue task generation unit (100) is used for trapped persons in the tunnel fire scene to log in to the rescue software through a mobile terminal and submit relevant rescue demand information, thereby generating a rescue task based on the submitted rescue demand information; A rescuer information collection unit (200), the rescuer information collection unit (200) is used to collect and record relevant information of rescuers who have registered the rescue calling software, and to regularly update the relevant information of the rescuers; A disaster detection and identification unit (300), the disaster detection and identification unit (300) being used to perform a call for help task distribution operation on the generated call for help task based on the real-time relevant information of the rescue personnel, and to deploy relevant disaster detection equipment in the tunnel fire scene, and to perform a disaster detection and identification operation based on the relevant disaster data obtained through the constructed disaster detection model; A disaster scenario deduction unit (400), the disaster scenario deduction unit (400) is used to process the obtained relevant disaster data, and obtain relevant disaster data after the development of the tunnel fire based on the processed relevant disaster data through a constructed disaster deduction model, thereby realizing the deduction of the situation and consequences of the tunnel fire disaster; A rescue mission execution unit (500) is used to select a rescue plan based on the tunnel fire disaster simulation results after the rescue personnel arrive, and execute the selected rescue plan, and record the rescue action process.
2. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The rescue task generation unit (100) comprises a user login module (101), a demand information submission module (102) and a rescue task construction module (103). The user login module (101) is used by trapped persons in a tunnel fire scenario to log in to a rescue software via a mobile terminal. The demand information submission module (102) is used by trapped persons to submit rescue demand information by scanning a rescue QR code in the rescue software after successfully logging in to the rescue software, including information on the number of trapped persons, trapped location information, captured surrounding video image information and trapped time information. The rescue task construction module (103) is used by the rescue software to generate a corresponding rescue task based on the rescue demand information submitted by the trapped persons.
3. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The rescuer information collection unit (200) comprises a rescuer information recording module (201) and a rescuer information updating module (202). The rescuer information recording module (201) is used to collect and record relevant information of the rescuers who have registered the rescue calling software, including rescuer size information, rescuer execution rescue mission type information, rescuer execution mission quantity information, rescuer location information and rescuer work status information. The database of the rescue calling software stores previous rescuer execution records and rescue plans for various rescue scenarios. The rescuer information updating module (202) is used to regularly update relevant information of the rescuers and data stored in the database. The rescuers can also update the information by submitting the information that needs to be modified and verifying it by the rescue calling software administrator.
4. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The disaster detection and identification unit (300) comprises a rescue task distribution module (301) and a disaster detection equipment deployment module (302). The rescue task distribution module (301) is used to perform rescue task distribution operations on the generated rescue task based on the real-time relevant information of the rescue personnel, and generate the estimated time for the rescue personnel to arrive at the tunnel fire site. The disaster detection equipment deployment module (302) is used to deploy relevant disaster detection equipment in the tunnel fire scene. The specific operation is to respectively arrange vehicle detectors in front of each tunnel, obtain the number of vehicles between each tunnel through the detection data, and then obtain the total number of vehicles in the tunnel, and arrange ultrasonic wind speed and wind direction detectors at the tunnel entrance to obtain real-time wind speed and wind direction data in the tunnel, and place multiple temperature sensors at fixed intervals at the tunnel vault position so that the detection area of the temperature sensor covers the entire tunnel area, so as to obtain the ambient temperature of each area of the tunnel and the tunnel vault temperature, and respectively arrange cameras in front of each tunnel to collect fire images and vehicle images in each tunnel.
5. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The disaster detection and identification unit (300) comprises a disaster detection model construction module (303), which is used to construct a disaster detection model based on YOLO V5s, and specifically comprises four parts: an input end, a backbone network, a neck network and an output end, wherein the input end is used to perform a data enhancement operation on the input image through a Mosaic method, so as to realize the method of small targets in the input image, thereby improving the detection capability of the disaster detection model for small targets, wherein the backbone network is used to perform a feature extraction operation on the input image, by splitting a high-resolution image in the input image into multiple low-resolution images, expanding the image of the input image in the RGB three channels, and realizing a two-fold downsampling operation while ensuring that the image information is not lost, and performing a convolution operation on the split image to obtain a feature map, and performing a maximum pooling layer operation on the feature map, and synthesizing images with different receptive fields to obtain a final feature map, thereby improving the recognition accuracy of the disaster detection model for small targets, wherein the neck network is used to integrate and fuse the obtained feature maps, and wherein the output end is used to convert the processed feature maps into the required detection results.
6. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The disaster detection and identification unit (300) further comprises a disaster detection model training module (304), which is used to obtain tunnel fire images and images of different types of vehicles in the tunnel through a search engine, and to construct a tunnel fire data set based on the images. When the number of samples in the tunnel fire data set is insufficient, data enhancement operations on the tunnel fire data set can be implemented by flipping the acquired images at multiple angles, and the tunnel fire images and tunnel vehicle images in the tunnel fire data set are labeled through the LabelImg tool. The disaster detection model is trained through the tunnel fire data set, and the trained disaster detection model is verified for accuracy. The disaster detection model that meets the model accuracy requirements is used to perform fire identification and accident vehicle identification operations based on the acquired fire images, vehicle images, and images submitted by trapped persons, thereby determining the fire location.
7. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The disaster scenario deduction unit (400) comprises a disaster data processing module (401) and a disaster deduction model construction module (402). The disaster data processing module (401) is used to process the obtained relevant disaster data. The specific operation is to calculate the fire heat release rate caused by the accident vehicle through a fire heat release algorithm, and based on the calculation result and the type of the accident vehicle, take the ratio of the calculation result to the maximum combustion heat release rate corresponding to the type of the accident vehicle. When 0% < the ratio result ≤ 25%, it is judged to be a small fire scale; when 25% < the ratio result ≤ 50%, it is judged to be a medium fire scale; when 50% < the ratio result ≤ 100%, it is judged to be a large fire scale. The disaster deduction model construction module (402) is used to construct a disaster deduction model based on a Bayesian network, and selects five factors, namely, tunnel environment temperature, tunnel real-time wind speed, accident vehicle type, fire scale and number of vehicles in the tunnel, as inputs of the disaster deduction model.
8. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 1, characterized in that: The disaster scenario simulation unit (400) further comprises a simulation result output module (403) and a disaster scale judgment module (404). The simulation result output module (403) is used to obtain the fire scale at the estimated time when rescue personnel arrive at the fire site in the tunnel based on the input tunnel environment temperature, tunnel real-time wind speed, accident vehicle type, fire scale and number of vehicles in the tunnel through a disaster simulation model. The disaster scale judgment module (404) is used to judge the severity of the disaster from two aspects, namely, smoke recirculation length and smoke layer thickness, based on the fire scale result output by the disaster simulation model. Specifically, the smoke recirculation length is calculated through a smoke recirculation algorithm, and the smoke layer thickness is calculated. The smoke reflow length is divided into four levels: none, [0m, 50m), [50m, 100m) and above 100m, which correspond to smoke reflow length levels one to four from low to high, respectively. The smoke reflow length level is then obtained based on the calculation results. The smoke layer thickness is calculated by the smoke layer thickness algorithm, and the smoke layer thickness is divided into four levels: none, [0m, 1.5m), [1.5m, 3m) and above 3m, which correspond to smoke layer thickness levels one to four from low to high, respectively. The smoke layer thickness level is then obtained based on the calculation results, and the smoke reflow length level and the smoke layer thickness level are averaged to obtain the disaster severity level when the rescue personnel arrive.
9. The interactive system for multi-dimensional positioning assistance using QR codes according to claim 8, characterized in that: The rescue task execution unit (500) comprises a rescue plan selection module (501), a rescue plan execution module (502) and a rescue task recording module (503). The rescue plan selection module (501) is used to call a rescue plan corresponding to the disaster severity level in the tunnel fire scenario from the database of the rescue calling software based on the disaster severity level obtained in the disaster scale judgment module (404) after the rescue personnel arrive, and when the disaster severity level is not an integer, the tunnel fire situation corresponding to the disaster severity level is regarded as the tunnel fire situation corresponding to the disaster severity level of an adjacent and higher integer level for processing. The rescue plan execution module (502) is used for the rescue personnel to perform the rescue task based on the called rescue plan, so as to rescue the trapped persons. The rescue task recording module (503) is used for the rescue personnel to record the whole process of the rescue task and upload it to the rescue calling software.
Citation Information
Patent Citations
Elevator rescue method and device, computer device and storage medium
CN109019222A
BIM-based tunnel fire emergency rescue method, terminal and storage medium
CN113947014A
Beidou intelligent emergency rescue command and dispatch system
CN118278577A
Disaster rescue system
JP2009073652A
Cited By
Improved deep learning target detection Mosaic sample enhancement method
CN120107547A
