Tunnel construction safety emergency method and device based on artificial intelligence

Abstract

The invention provides a tunnel construction safety emergency method and device based on artificial intelligence, and relates to the technical field of artificial intelligence. Firstly, position distribution information and safety emergency feature information of persons in a tunnel are obtained based on video image information; then, judging whether a field safety accident is found in the tunnel or not based on the environment information and / or the video image information, and identifying the type and the occurrence place of the field safety accident; then, selecting a target person with scene safety accident handling experience in the tunnel based on the type of the scene safety accident, the position distribution information of the persons in the tunnel and the safety emergency feature information of each person; and finally, notifying the target person to go to the site where the on-site safety accident occurs to carry out safety emergency processing. According to the mode, the target person with accident handling experience can timely arrive at the site to handle the accident, the handling efficiency of the tunnel site safety accident is improved, the tunnel construction risk is reduced, and the tunnel safety construction is ensured.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based tunnel construction safety emergency method and device. Background Art

[0002] Construction safety is the top priority of on-site construction. Tunnel construction is a relatively closed construction environment. When a construction safety accident occurs somewhere in the tunnel, how to quickly discover and promptly handle it becomes a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0003] In order to at least overcome the above-mentioned deficiencies in the prior art, in a first aspect, the present invention provides a tunnel construction safety emergency method based on artificial intelligence, which is applied to a computer device in a tunnel construction safety emergency system, wherein the tunnel construction safety emergency system further comprises an image acquisition device and an environmental parameter detection sensor that are communicatively connected to the computer device, wherein the image acquisition device and the environmental parameter detection sensor are arranged in the tunnel for respectively acquiring video image information and environmental information in the tunnel, and the method comprises: Acquire video image information acquired by the image acquisition device, obtain position distribution information of people in the tunnel and identity feature information of each person based on the video image information, and obtain safety emergency feature information of the person based on the identity feature information of the person; Acquire environmental information collected by the environmental parameter detection sensor, and determine whether an on-site safety accident is found in the tunnel based on the environmental information and / or the video image information; If it is determined that an on-site safety accident has occurred in the tunnel, identifying the type and location of the on-site safety accident; Selecting a target person in the tunnel who has experience in handling the on-site safety accident based on the type of the on-site safety accident, the location distribution information of the people in the tunnel, and the safety emergency feature information of each person; Notify the target person to go to the location of the on-site safety accident for safety emergency response.

[0004] In a possible implementation, the step of acquiring the video image information acquired by the image acquisition device, acquiring the position distribution information of the persons in the tunnel and the identity feature information of each person based on the video image information, and acquiring the safety emergency feature information of the person based on the identity feature information of the person includes: Acquire video image information collected by image acquisition devices arranged at different positions in the tunnel; Determine the position distribution information of the people in the tunnel based on the proportion of people in the video image information, the location where the image acquisition device is installed, and the shooting angle of the image acquisition device; Obtaining identity feature information of each person in the tunnel based on facial feature recognition of the person in the video image information; Based on the identity feature information of the person, safety emergency feature information of the person is obtained from a safety emergency feature database, wherein the safety emergency feature database records safety emergency feature information of different persons, and the safety emergency feature information includes course information of whether the person has participated in safety emergency training and the corresponding actual safety emergency operation capability index.

[0005] In a possible implementation, the step of obtaining environmental information collected by the environmental parameter detection sensor, and judging whether an on-site safety accident is found in the tunnel based on the environmental information and / or the video image information, includes: Acquiring environmental information collected by environmental parameter detection sensors arranged at different positions in the tunnel, wherein the environmental information includes oxygen content, temperature, humidity, volume or brightness; The environmental information and / or the video image information are input into a safety accident prediction model, and the safety accident prediction model is used to determine whether an on-site safety accident is found in the tunnel, wherein the safety accident prediction model determines whether an on-site safety accident is found in the tunnel based on abnormal environmental parameters in the environmental information and / or abnormal images in abnormal areas in the video image information.

[0006] In a possible implementation, if it is determined that an on-site safety accident occurs in the tunnel, the step of identifying the type and location of the on-site safety accident further includes: When the safety accident prediction model determines that an on-site safety accident occurs in the tunnel, the safety accident prediction model further determines the location of the on-site safety accident based on the location of the environmental parameter detection sensor that collects the abnormal environmental parameters and / or the location of the image acquisition device that collects the abnormal image; The safety accident prediction model also predicts the type of the on-site safety accident based on the abnormal environmental parameters and / or abnormal images, wherein the types include construction personnel accidents, construction equipment accidents and construction site accidents.

[0007] In a possible implementation, the step of selecting a target person in the tunnel who has experience in handling the on-site safety accident based on the type of the on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency feature information of each person includes: Based on the location of the on-site safety accident and the position distribution information of the people in the tunnel, the distance characteristics of each person from the location of the accident are calculated, and based on the type of the on-site safety accident and the safety emergency characteristic information of each person, the adaptability characteristics of each person in handling the on-site safety accident are calculated; The success rate feature of the person in the tunnel successfully handling the on-site safety accident is calculated based on the distance feature and the fitness feature; Selecting a person whose success rate in the success rate feature is ranked before a preset ranking as the target person, and generating route guidance information for the target person to the occurrence location; The step of notifying the target person to go to the location where the on-site safety accident occurred to carry out safety emergency treatment includes: A notification of safety emergency handling and route guidance information to the occurrence location are sent to the communication device carried by the target person.

[0008] In a possible implementation, after the step of notifying the target person to go to the location where the on-site safety accident occurred to carry out safety emergency processing, the method further includes: Receiving the video image information of the occurrence location acquired by the image acquisition device and / or the environmental information of the occurrence location acquired by the environmental parameter detection sensor; Based on the video image information and / or environmental information of the occurrence location when and after the on-site safety accident occurs, obtain the time spent by the target person to handle the on-site safety accident and the handling result of the on-site safety accident; Calculating a handling evaluation value of the on-site safety accident based on the time spent by the target person in handling the on-site safety accident and the handling result of the on-site safety accident, wherein the handling evaluation value represents the handling efficiency of the on-site safety accident by the target person; The safety emergency feature information of the target person is updated according to the processing evaluation value.

[0009] In a possible implementation, the step of obtaining the time spent by the target person in handling the on-site safety accident and the handling result of the on-site safety accident based on the video image information and / or environmental information of the occurrence location when and after the on-site safety accident occurs includes: When the result of handling the on-site safety accident is to eliminate the on-site safety accident, the time spent by the target person in handling the on-site safety accident is compared with the average handling time of the on-site safety accident of the same type obtained from the big database, and when the handling time is less than the average handling time, a positive handling evaluation value is obtained; When the processing result of the on-site safety accident is elimination of the on-site safety accident, and the processing time is equal to the average processing time, a maintenance processing evaluation value is obtained; When the result of handling the on-site safety accident is that the on-site safety accident is not eliminated or the handling time is longer than the average handling time, a negative handling evaluation value is obtained; The step of updating the safety emergency feature information of the target person according to the processing evaluation value comprises: When the processing evaluation value is a positive processing evaluation value, increasing the emergency characteristic value in the safety emergency characteristic information of the target person, wherein the magnitude of the emergency characteristic value is positively correlated with the on-site emergency processing capability; When the processing evaluation value is the maintaining processing evaluation value, maintaining the emergency feature value in the safety emergency feature information of the target person unchanged; When the processing evaluation value is a negative processing evaluation value, the emergency feature value in the safety emergency feature information of the target person is reduced.

[0010] In a possible implementation, a training sample set is created, the training sample set including sample parameters and sample tags, wherein the sample parameters include a combination of environmental parameters and / or video image information under various tunnel environments, and the sample tags include whether it is an on-site safety accident and the type of the on-site safety accident; Inputting the training samples in the training sample set into a neural network in sequence for training, and outputting a first training result of whether the training sample is an on-site safety accident and a second training result of what type of on-site safety accident the training sample is, wherein the neural network includes a judgment subnetwork and a prediction subnetwork; Calculating a training loss function value based on the first training result, the second training result, and the sample label, and adjusting network parameters of the judgment subnetwork and the prediction subnetwork when the loss function value is greater than a preset loss function value; Repeatedly input the training samples in the training sample set into the neural network for training until the loss function value is no greater than the preset loss function value, and the training is completed. The neural network obtained by the training is used as the safety accident prediction model.

[0011] In a second aspect, the present invention further provides an artificial intelligence-based tunnel construction safety emergency device, which is applied to a computer device in a tunnel construction safety emergency system, wherein the tunnel construction safety emergency system further comprises an image acquisition device and an environmental parameter detection sensor that are communicatively connected to the computer device, wherein the image acquisition device and the environmental parameter detection sensor are arranged in the tunnel for respectively acquiring video image information and environmental information in the tunnel, and the device comprises: A first acquisition module is used to acquire video image information acquired by the image acquisition device, obtain position distribution information of people in the tunnel and identity feature information of each person based on the video image information, and obtain safety emergency feature information of the person based on the identity feature information of the person; A second acquisition module is used to acquire environmental information collected by the environmental parameter detection sensor, and determine whether an on-site safety accident is found in the tunnel based on the environmental information and / or the video image information; A judgment module, for identifying the type and location of an on-site safety accident if it is determined that an on-site safety accident has occurred in the tunnel; A selection module, configured to select a target person in the tunnel who has experience in handling the on-site safety accident based on the type of the on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency feature information of each person; The notification module is used to notify the target person to go to the location where the on-site safety accident occurred to carry out safety emergency response.

[0012] In a third aspect, the present invention also provides a computer device, comprising a processor, a computer-readable storage medium and a communication interface, wherein the computer-readable storage medium, the communication interface and the processor are connected via a bus system, the computer-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the computer-readable storage medium to execute the artificial intelligence-based tunnel construction safety emergency response method described in any possible implementation method of the first aspect.

[0013] In a fourth aspect, the present invention further provides a computer-readable storage medium, in which instructions are stored, which, when executed, enable a computer to execute an artificial intelligence-based tunnel construction safety emergency response method in any possible implementation of the first aspect.

[0014] In the above scheme provided by the present invention, first, the position distribution information of the people in the tunnel and the safety emergency characteristic information of each person are obtained based on the video image information; then, based on the environmental information and / or the video image information, it is determined whether an on-site safety accident is found in the tunnel. If it is determined that an on-site safety accident occurs in the tunnel, the type and location of the on-site safety accident are identified; then, based on the type of on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency characteristic information of each person, a target person with experience in handling on-site safety accidents in the tunnel is selected; finally, the target person is notified to go to the location of the on-site safety accident for safety emergency handling. The above method can determine whether an on-site safety accident occurs through tunnel information, and select the target person through the safety emergency characteristic information of the people in the tunnel when an on-site safety accident occurs. In this way, the target person with experience in handling accidents can arrive at the scene in time to handle the accident, thereby improving the efficiency of handling on-site safety accidents in the tunnel, reducing the risk of tunnel construction, and ensuring safe construction of the tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A block diagram of a tunnel construction safety emergency system provided by an embodiment of the present invention; Figure 2 A schematic diagram of a process flow of a tunnel construction safety emergency method based on artificial intelligence provided by an embodiment of the present invention; Figure 3 for Figure 2 A schematic flow chart of sub-steps of step S10 in FIG. Figure 4 for Figure 2 A schematic flow chart of sub-steps of step S40 in FIG. Figure 5 A schematic diagram of another part of the flow chart of the tunnel construction safety emergency response method based on artificial intelligence provided by an embodiment of the present invention; Figure 6 A schematic diagram of the functional modules of an artificial intelligence-based tunnel construction safety emergency device provided in an embodiment of the present invention; Figure 7 A schematic diagram of the structural framework of a computer device for implementing the above-mentioned artificial intelligence-based tunnel construction safety emergency method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention is described in detail below with reference to the accompanying drawings. The specific operation method in the method embodiment can also be applied to the device embodiment or the system embodiment.

[0018] Before introducing the specific solution provided by this embodiment, the application scenario of the tunnel construction safety emergency system to which the specific solution is applicable is first introduced. Figure 1 In this embodiment, the tunnel construction safety emergency system includes a computer device 10, an image acquisition device 20 and an environmental parameter detection sensor 30 that are communicatively connected, wherein the image acquisition device 20 and the environmental parameter detection sensor 30 are installed at different positions in the tunnel to collect video image information and environmental information at different positions in the tunnel. The image acquisition device 20 and the environmental parameter detection sensor 30 respectively send the collected video image information and environmental information to the computer device 10, and the computer device 10 performs analysis and processing based on the above video image information and environmental information. The image acquisition device 20 includes a camera with adjustable image acquisition direction, and the environmental parameter detection sensor 30 includes an oxygen content detection sensor, a humidity detection sensor, a temperature detection sensor, a volume detection sensor, a brightness detection sensor, and the like.

[0019] Combine the following Figure 1 The application scenario shown is an exemplary description of the tunnel construction safety emergency method based on artificial intelligence provided by the embodiment of the present application. Figure 2 The artificial intelligence-based tunnel construction safety emergency method provided in the embodiment of the present application can be executed by the aforementioned computer device 10. In other embodiments, the order of some steps in the artificial intelligence-based tunnel construction safety emergency method in the embodiment of the present application can be exchanged with each other according to actual needs, or some of the steps can be omitted or deleted. The detailed steps of the artificial intelligence-based tunnel construction safety emergency method executed by the computer device 10 are introduced as follows.

[0020] Step S10, obtaining video image information collected by the image acquisition device 20, obtaining position distribution information of people in the tunnel and identity feature information of each person based on the video image information, and obtaining safety emergency feature information of the person based on the identity feature information of the person.

[0021] In this embodiment, the computer device 10 can obtain information about people entering the tunnel by analyzing video image information collected by the image acquisition device 20 set at different positions in the tunnel. The person information includes the position distribution information of the person and the identity feature information of each person, wherein the identity feature information includes the person's work number information, gender, age, type of work and length of service, etc. For example, the identity feature information matching the person can be obtained by recognizing the face of the person in the video image information. For example, the identity feature information can be represented by a string of numbers, such as 01245-01-43-03-07, wherein 01245 represents the work number, 01 represents gender (such as male), 43 represents age, 03 represents the type of work (such as a dump truck driver), and 07 represents length of service. In this embodiment, after obtaining the identity feature information of the person in the tunnel, the safety emergency feature information of the person can be obtained based on the identity feature information, wherein the safety emergency feature information includes feature parameters used to characterize the person's ability to respond to on-site safety accidents.

[0022] Step S20, obtaining environmental information collected by environmental parameter detection sensors, and determining whether an on-site safety accident is found in the tunnel based on the environmental information and / or video image information.

[0023] In this embodiment, it is possible to determine whether an on-site safety accident has occurred based on at least one of the environmental information and the video image information. For example, it is possible to determine whether someone has fainted at the construction site based on the recognition and analysis of the images in the video image information. It is also possible to determine whether a fire has occurred at the construction site based on the environmental information (for example, temperature, oxygen content, and brightness). It is also possible to determine whether other on-site safety accidents have occurred at the construction site (for example, an explosion at the construction site, etc.) in combination with the video image information and the environmental information.

[0024] Step S30: If it is determined that an on-site safety accident occurs in the tunnel, the type and location of the on-site safety accident are identified.

[0025] After determining that an on-site safety accident has occurred in the tunnel, the computer device 10 can identify the type of on-site safety accident based on abnormal images in the video image information and / or abnormal environmental parameters in the environmental information, and determine the location of the on-site safety accident based on the position of the image acquisition device that acquires the above-mentioned abnormal images and / or the position of the environmental parameter detection sensor that acquires the above-mentioned abnormal environmental parameters.

[0026] Step S40, selecting a target person in the tunnel who has experience in handling on-site safety accidents based on the type of on-site safety accidents, the location distribution information of the people in the tunnel, and the safety emergency feature information of each person.

[0027] In this embodiment, the number of target persons selected may be configured based on the type of on-site safety incident, and different types of on-site safety incidents correspond to different selected target persons.

[0028] Step S50, notifying the target person to go to the scene where the safety accident occurred to carry out safety emergency treatment.

[0029] After determining the target person, the computer device 10 may send a notification to the target person via the local communication network established in the tunnel.

[0030] The technical solution provided above first obtains the position distribution information of the people in the tunnel and the safety emergency characteristic information of each person based on the video image information; then, based on the environmental information and / or video image information, it is determined whether an on-site safety accident is found in the tunnel. If it is determined that an on-site safety accident occurs in the tunnel, the type and location of the on-site safety accident are identified; then, based on the type of on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency characteristic information of each person, a target person with experience in handling on-site safety accidents in the tunnel is selected; finally, the target person is notified to go to the location of the on-site safety accident for safety emergency handling. The above method can determine whether an on-site safety accident occurs through tunnel information, and select the target person through the safety emergency characteristic information of the people in the tunnel when an on-site safety accident occurs. In this way, the target person with experience in handling accidents can arrive at the scene in time to handle the accident, improve the efficiency of handling on-site safety accidents in the tunnel, reduce the risk of tunnel construction, and ensure safe construction of the tunnel.

[0031] For further information, please refer to Figure 3 , step S10 in this embodiment can be implemented in the following way.

[0032] Step S101, obtaining video image information collected by image acquisition devices 20 arranged at different positions in the tunnel.

[0033] In this embodiment, the image acquisition devices 20 set at different positions can be numbered. When obtaining video image information acquired by different image acquisition devices 20, numbers can be added to the video image information so that the computer device 10 can identify its source and the corresponding tunnel location.

[0034] Step S102, determining the position distribution information of the people in the tunnel based on the proportion of people in the video image information, the installation position of the image acquisition device 20 and the shooting angle of the image acquisition device.

[0035] In this embodiment, the distance between the person and the image acquisition device 20 can be calculated through the person in the video image information and the shooting angle of the image acquisition device 20, and then the position distribution information of the person in the tunnel can be calculated based on the distance and the installation position of the image acquisition device 20, and the position distribution information includes at least the position coordinates.

[0036] Step S103, obtaining identity feature information of each person in the tunnel based on facial feature recognition of the person in the video image information.

[0037] Based on face recognition technology, the identity feature information of each person in the tunnel can be obtained by recognizing the facial features of the person in the video image information. In this embodiment, face recognition can be performed using video image information that captures the front facial features of the person.

[0038] Step S104, obtaining the person's safety emergency feature information from a safety emergency feature database based on the person's identity feature information.

[0039] The safety emergency characteristic database records the safety emergency characteristic information of different people, and the safety emergency characteristic information includes whether the course information of safety emergency training has been participated in and the corresponding actual safety emergency operation ability index, wherein the actual safety emergency operation ability index is positively correlated with the number of safety emergency treatments participated in by the person in history, and the higher the actual safety emergency operation ability index is, the stronger the safety emergency treatment ability of the person is. It can be understood that in this embodiment, different actual safety emergency operation ability indexes can be corresponding to different safety emergency types, such as a corresponding actual safety emergency operation ability index for on-site personnel rescue, and a corresponding actual safety emergency operation ability index for on-site construction equipment repair.

[0040] In this embodiment, step S20 can be implemented in the following manner.

[0041] First, environmental information collected by environmental parameter detection sensors 30 arranged at different positions in the tunnel is obtained.

[0042] The environmental information includes but is not limited to oxygen content, temperature, humidity, volume or brightness, etc.

[0043] Secondly, the environmental information and / or video image information is input into the safety accident prediction model, and the safety accident prediction model is used to determine whether an on-site safety accident is found in the tunnel.

[0044] Among them, the safety accident prediction model determines whether an on-site safety accident occurs in the tunnel based on abnormal environmental parameters in the environmental information and / or abnormal images in abnormal areas in the video image information.

[0045] Exemplarily, the safety accident prediction model can determine whether an on-site safety accident occurs in the tunnel based on at least one of abnormal environmental parameters and abnormal images. For example, it can determine whether someone has fainted at the construction site based on the identification and analysis of abnormal images in the video image information. It can also determine whether a fire has occurred at the construction site based on abnormal environmental information (for example, abnormal temperature, abnormal oxygen content, and abnormal brightness). It can also determine whether other on-site safety accidents occur at the construction site (for example, an explosion at the construction site, etc.) in combination with abnormal images in the video image information and abnormal environmental parameters in the environmental information.

[0046] Further, in this embodiment, step S30 can be implemented in the following manner.

[0047] First, when the safety accident prediction model determines that an on-site safety accident has occurred in the tunnel, the safety accident prediction model also determines the location of the on-site safety accident based on the position of the environmental parameter detection sensor that collects abnormal environmental parameters and / or the position of the image acquisition device that collects abnormal images.

[0048] Next, the safety accident prediction model also predicts the type of on-site safety accident based on abnormal environmental parameters and / or abnormal images.

[0049] Among them, the types of on-site safety accidents include construction personnel accidents, construction equipment accidents and construction site accidents. In detail, construction personnel accidents may include construction personnel falling ill, construction personnel on-site conflicts, etc. Construction equipment failures include construction equipment failures themselves, collision failures between multiple construction equipment, etc. Construction site failures include explosions, excessive amounts of hazardous substances, etc.

[0050] For further information, please refer to Figure 4 , step S40 in this embodiment can be implemented in the following way.

[0051] Step S401, based on the location of the on-site safety accident and the position distribution information of the characters in the tunnel, the distance characteristics of each character from the location of the accident are calculated, and based on the type of the on-site safety accident and the safety emergency characteristic information of each character, the adaptability characteristics of each character in handling the on-site safety accident are calculated.

[0052] In this embodiment, the distance that each person needs to travel to reach the location of the on-site safety accident (i.e., the distance characteristics of each person from the location of the accident) can be calculated based on the location of the on-site safety accident, the location distribution information of the person in the tunnel, and the distribution of the tunnel. And the adaptability characteristics of each person in handling on-site safety accidents can be calculated based on the type of on-site safety accident and the actual safety emergency operation capability index of each person for different types of on-site safety accidents.

[0053] Step S402, calculating the success rate feature of the person in the tunnel successfully handling the on-site safety accident based on the distance feature and the fitness feature.

[0054] For example, the success rate feature may be negatively correlated with the distance feature, and positively correlated with the fitness feature. That is, the larger the distance feature (the farther the distance), the longer it takes to reach the location of the incident, which will reduce the success rate of emergency response, while the higher the fitness feature, the higher the success rate of emergency response to on-site safety accidents.

[0055] Step S403, selecting a person whose success rate in the success rate feature is ranked before a preset ranking as a target person, and generating route guidance information for the target person to the occurrence location.

[0056] Further, in this embodiment, step S50 can be implemented in the following manner.

[0057] A notification of safety emergency handling and route guidance information to the location of the incident are sent to the communication device carried by the target person.

[0058] For further information, please refer to Figure 5 After step S50, the tunnel construction safety emergency method based on artificial intelligence provided in this embodiment also includes the following steps.

[0059] Step S61: receiving video image information of the occurrence location acquired by an image acquisition device and / or environmental information of the occurrence location acquired by an environmental parameter detection sensor.

[0060] Step S62, based on the video image information and / or environmental information of the location when and after the on-site safety accident occurs, obtain the time spent by the target person to handle the on-site safety accident and the handling result of the on-site safety accident.

[0061] In this embodiment, the processing result of the on-site safety accident can be determined by the video image information and / or environmental information at the time and after the on-site safety accident occurs. For example, when a person faints, the fainted person can move through emergency treatment of the target person or the rescue process video image analysis determines that the rescue process meets the rescue specifications, then the processing result of the on-site safety accident is determined to be the elimination of the on-site safety accident. For another example, when a fire occurs, the environmental parameters are restored to the normal parameter range or the firefighting process is determined to meet the firefighting specifications through the rescue process video image analysis, then the processing result of the on-site safety accident is determined to be the elimination of the on-site safety accident.

[0062] Step S63, calculating an evaluation value of the handling of the on-site safety accident based on the time spent by the target person in handling the on-site safety accident and the handling result of the on-site safety accident.

[0063] Among them, the processing evaluation value represents the efficiency of the target person in handling on-site safety accidents.

[0064] In this embodiment, step S63 can be implemented in the following manner.

[0065] When the result of handling an on-site safety accident is the elimination of the on-site safety accident, the time spent by the target person in handling the on-site safety accident is compared with the average handling time of this type of on-site safety accident obtained from the big database. When the handling time is less than the average handling time, a positive handling evaluation value is obtained.

[0066] When the result of handling the on-site safety accident is to eliminate the on-site safety accident, and the handling time is equal to the average handling time, a maintenance handling evaluation value is obtained; When the result of handling an on-site safety accident is that the on-site safety accident is not eliminated or the handling time is longer than the average handling time, a negative handling evaluation value is obtained.

[0067] Step S64: updating the safety emergency feature information of the target person according to the processing evaluation value.

[0068] In this embodiment, step S64 can be implemented in the following manner.

[0069] When the processing evaluation value is a positive processing evaluation value, the emergency characteristic value in the safety emergency characteristic information of the target person is increased, wherein the magnitude of the emergency characteristic value is positively correlated with the on-site emergency processing capability; When the processing evaluation value is the maintaining processing evaluation value, the emergency feature value in the safety emergency feature information of the target person is maintained unchanged; When the processing evaluation value is a negative processing evaluation value, the emergency feature value in the safety emergency feature information of the target person is reduced, and the target person may be required to re-learn the relevant safety emergency training courses.

[0070] Furthermore, the artificial intelligence-based tunnel construction safety emergency response method adopted in this embodiment also includes a step of training a safety accident prediction model, which can be implemented in the following way.

[0071] First, create a training sample set.

[0072] The training sample set includes sample parameters and sample tags, wherein the sample parameters include a combination of environmental parameters and / or video image information under various tunnel environments, and the sample tags include whether it is an on-site safety accident and the type of the on-site safety accident; Next, the training samples in the training sample set are sequentially input into a neural network for training, and a first training result of whether the training sample is an on-site safety accident and a second training result of the type of on-site safety accident are output.

[0073] Wherein, the neural network includes a judgment subnetwork and a prediction subnetwork; Next, a training loss function value is calculated based on the first training result, the second training result and the sample label, and when the loss function value is greater than a preset loss function value, network parameters of the judgment subnetwork and the prediction subnetwork are adjusted.

[0074] Then, the training samples in the training sample set are repeatedly input into the neural network for training until the loss function value is no greater than the preset loss function value, and the training is completed. The neural network obtained by the training is used as the safety accident prediction model.

[0075] Please refer to Figure 6 , Figure 6 The functional module diagram of the tunnel construction safety emergency device 100 based on artificial intelligence provided in this embodiment, this embodiment can divide the functional modules of the tunnel construction safety emergency device 100 based on artificial intelligence according to the above method embodiment, that is, the following functional modules corresponding to the tunnel construction safety emergency device 100 based on artificial intelligence can be used to execute the above method embodiments. Among them, the tunnel construction safety emergency device 100 based on artificial intelligence can include a first acquisition module 110, a second acquisition module 120, a judgment module 130, a selection module 140 and a notification module 150. The functions of each functional module of the tunnel construction safety emergency device 100 based on artificial intelligence are described in detail below.

[0076] The first acquisition module 110 is used to acquire video image information acquired by the image acquisition device, obtain the position distribution information of the people in the tunnel and the identity feature information of each person based on the video image information, and obtain the safety emergency feature information of the person based on the identity feature information of the person.

[0077] In this embodiment, the first acquisition module 110 can obtain information about people entering the tunnel by analyzing video image information collected by image acquisition devices 20 set at different positions in the tunnel. The person information includes position distribution information of the person and identity feature information of each person, wherein the identity feature information includes the person's work number information, gender, age, type of work and length of service, etc. For example, the identity feature information matching the person can be obtained by recognizing the face of the person in the video image information. For example, the identity feature information can be represented by a string of numbers, for example, 01245-01-43-03-07, wherein 01245 represents the work number, 01 represents gender (such as male), 43 represents age, 03 represents the type of work (such as a dump truck driver), and 07 represents length of service. In this embodiment, after obtaining the identity feature information of the person in the tunnel, the safety emergency feature information of the person can be obtained based on the identity feature information, wherein the safety emergency feature information includes feature parameters used to characterize the person's ability to respond to on-site safety accidents.

[0078] The first acquisition module 110 may be used to execute the above step S10. The detailed implementation of the first acquisition module 110 may refer to the above detailed description of step S10.

[0079] The second acquisition module 120 is used to acquire environmental information collected by the environmental parameter detection sensor, and determine whether an on-site safety accident is found in the tunnel based on the environmental information and / or video image information.

[0080] In this embodiment, it is possible to determine whether an on-site safety accident has occurred based on at least one of the environmental information and the video image information. For example, it is possible to determine whether someone has fainted at the construction site based on the recognition and analysis of the images in the video image information. It is also possible to determine whether a fire has occurred at the construction site based on the environmental information (for example, temperature, oxygen content, and brightness). It is also possible to determine whether other on-site safety accidents have occurred at the construction site (for example, an explosion at the construction site, etc.) in combination with the video image information and the environmental information.

[0081] The second acquisition module 120 may be used to execute the above step S20. The detailed implementation of the second acquisition module 120 may refer to the detailed description of the above step S20.

[0082] The judgment module 130 is used to identify the type and location of the on-site safety accident if it is determined that an on-site safety accident has occurred in the tunnel.

[0083] After determining that an on-site safety accident has occurred in the tunnel, the computer device 10 can identify the type of on-site safety accident based on abnormal images in the video image information and / or abnormal environmental parameters in the environmental information, and determine the location of the on-site safety accident based on the position of the image acquisition device that acquires the above-mentioned abnormal images and / or the position of the environmental parameter detection sensor that acquires the above-mentioned abnormal environmental parameters.

[0084] The determination module 130 executes the above step S30 . The detailed implementation of the determination module 130 may refer to the above detailed description of step S30 .

[0085] The selection module 140 is used to select a target person in the tunnel who has experience in handling on-site safety accidents based on the type of on-site safety accidents, the location distribution information of the people in the tunnel, and the safety emergency feature information of each person.

[0086] In this embodiment, the number of target persons selected may be configured based on the type of on-site safety incident, and different types of on-site safety incidents correspond to different selected target persons.

[0087] The selection module 140 may be used to execute the above step S40. The detailed implementation of the selection module 140 may refer to the above detailed description of step S40. The notification module 150 is used to notify the target person to go to the scene where the safety accident occurred to carry out safety emergency processing.

[0088] After determining the target person, the notification module 150 may send a notification to the target person via the local communication network established in the tunnel.

[0089] The notification module 150 may be used to execute the above step S50. The detailed implementation of the notification module 150 may refer to the above detailed description of step S50.

[0090] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, all or part of them can be integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software (for example, open source software) called by processing elements. It can also be implemented in the form of hardware. Some modules can also be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the selected module 140 can be a separately established processing element, or it can be integrated in a chip of the above device for implementation. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The function of the selected module 140. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in a processor element or instructions in the form of software.

[0091] Please refer to Figure 7 , Figure 7 The hardware structure diagram of the computer device 10 provided in the embodiment of the present disclosure for implementing the above-mentioned tunnel construction safety emergency method based on artificial intelligence is shown. The computer device 10 can be implemented on a cloud server. Figure 7 As shown, the computer device 10 may include a processor 101 , a computer-readable storage medium 102 , a bus 103 , and a communication interface 104 .

[0092] In a specific implementation process, at least one processor 101 executes computer-executable instructions (e.g., Figure 6 ), so that the processor 101 can execute the tunnel construction safety emergency method based on artificial intelligence as described in the above method embodiment, wherein the processor 101, the computer-readable storage medium 102 and the communication interface 104 are connected via a bus 103, and the processor 101 can be used to control the sending and receiving actions of the communication interface 104.

[0093] The specific implementation process of the processor 101 can refer to the various method embodiments executed by the above-mentioned computer device 10. The implementation principles and technical effects are similar and will not be repeated in this embodiment.

[0094] The computer-readable storage medium 102 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0095] The bus 103 can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of the present invention is not limited to only one bus or one type of bus.

[0096] In addition, an embodiment of the present invention further provides a readable storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the above-mentioned artificial intelligence-based tunnel construction safety emergency method is implemented.

[0097] In summary, the technical solution provided by the embodiment of the present invention first obtains the position distribution information of the people in the tunnel and the safety emergency feature information of each person based on the video image information; then, based on the environmental information and / or the video image information, it is determined whether an on-site safety accident is found in the tunnel. If it is determined that an on-site safety accident occurs in the tunnel, the type and location of the on-site safety accident are identified; then, based on the type of on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency feature information of each person, a target person with experience in handling on-site safety accidents in the tunnel is selected; finally, the target person is notified to go to the location of the on-site safety accident for safety emergency handling. The above method can determine whether an on-site safety accident occurs through tunnel information, and select the target person through the safety emergency feature information of the people in the tunnel when an on-site safety accident occurs. In this way, the target person with experience in handling accidents can arrive at the scene in time to handle the accident, thereby improving the efficiency of handling on-site safety accidents in the tunnel, reducing the risk of tunnel construction, and ensuring safe construction of the tunnel.

[0098] The above is a description of a specific embodiment of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A tunnel construction safety emergency method based on artificial intelligence, characterized in that: A computer device used in a tunnel construction safety emergency system, wherein the tunnel construction safety emergency system further comprises an image acquisition device and an environmental parameter detection sensor that are communicatively connected to the computer device, wherein the image acquisition device and the environmental parameter detection sensor are arranged in the tunnel for respectively acquiring video image information and environmental information in the tunnel, and the method comprises: Acquire video image information acquired by the image acquisition device, obtain position distribution information of people in the tunnel and identity feature information of each person based on the video image information, and obtain safety emergency feature information of the person based on the identity feature information of the person; Acquire environmental information collected by the environmental parameter detection sensor, and determine whether an on-site safety accident is found in the tunnel based on the environmental information and / or the video image information; If it is determined that an on-site safety accident has occurred in the tunnel, identifying the type and location of the on-site safety accident; Selecting a target person in the tunnel who has experience in handling the on-site safety accident based on the type of the on-site safety accident, the location distribution information of the people in the tunnel, and the safety emergency feature information of each person; Notify the target person to go to the location of the on-site safety accident for safety emergency response.

2. The artificial intelligence-based tunnel construction safety emergency method according to claim 1, characterized in that: The step of acquiring the video image information acquired by the image acquisition device, acquiring the position distribution information of the persons in the tunnel and the identity feature information of each person based on the video image information, and acquiring the safety emergency feature information of the person based on the identity feature information of the person, comprises: Acquire video image information collected by image acquisition devices arranged at different positions in the tunnel; Determine the position distribution information of the people in the tunnel based on the proportion of people in the video image information, the location where the image acquisition device is installed, and the shooting angle of the image acquisition device; Obtaining identity feature information of each person in the tunnel based on facial feature recognition of the person in the video image information; Based on the identity feature information of the person, safety emergency feature information of the person is obtained from a safety emergency feature database, wherein the safety emergency feature database records safety emergency feature information of different persons, and the safety emergency feature information includes course information of whether the person has participated in safety emergency training and the corresponding actual safety emergency operation capability index.

3. The artificial intelligence-based tunnel construction safety emergency method according to claim 1, characterized in that: The step of obtaining the environmental information collected by the environmental parameter detection sensor, and judging whether an on-site safety accident is found in the tunnel based on the environmental information and / or the video image information, comprises: Acquiring environmental information collected by environmental parameter detection sensors arranged at different positions in the tunnel, wherein the environmental information includes oxygen content, temperature, humidity, volume or brightness; The environmental information and / or the video image information are input into a safety accident prediction model, and the safety accident prediction model is used to determine whether an on-site safety accident is found in the tunnel, wherein the safety accident prediction model determines whether an on-site safety accident is found in the tunnel based on abnormal environmental parameters in the environmental information and / or abnormal images in abnormal areas in the video image information.

4. The artificial intelligence-based tunnel construction safety emergency method according to claim 3, characterized in that: If it is determined that an on-site safety accident occurs in the tunnel, the step of identifying the type and location of the on-site safety accident also includes: When the safety accident prediction model determines that an on-site safety accident occurs in the tunnel, the safety accident prediction model further determines the location of the on-site safety accident based on the location of the environmental parameter detection sensor that collects the abnormal environmental parameters and / or the location of the image acquisition device that collects the abnormal image; The safety accident prediction model also predicts the type of the on-site safety accident based on the abnormal environmental parameters and / or abnormal images, wherein the types include construction personnel accidents, construction equipment accidents and construction site accidents.

5. The artificial intelligence-based tunnel construction safety emergency method according to claim 4, characterized in that: The step of selecting a target person in the tunnel who has experience in handling the on-site safety accident based on the type of the on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency feature information of each person includes: Based on the location of the on-site safety accident and the position distribution information of the people in the tunnel, the distance characteristics of each person from the location of the accident are calculated, and based on the type of the on-site safety accident and the safety emergency characteristic information of each person, the adaptability characteristics of each person in handling the on-site safety accident are calculated; The success rate feature of the person in the tunnel successfully handling the on-site safety accident is calculated based on the distance feature and the fitness feature; Selecting a person whose success rate in the success rate feature is ranked before a preset ranking as the target person, and generating route guidance information for the target person to the occurrence location; The step of notifying the target person to go to the location where the on-site safety accident occurred to carry out safety emergency treatment includes: A notification of safety emergency handling and route guidance information to the occurrence location are sent to the communication device carried by the target person.

6. The artificial intelligence-based tunnel construction safety emergency method according to claim 5, characterized in that: After the step of notifying the target person to go to the location where the on-site safety accident occurred to carry out safety emergency treatment, the method further includes: Receiving the video image information of the occurrence location acquired by the image acquisition device and / or the environmental information of the occurrence location acquired by the environmental parameter detection sensor; Based on the video image information and / or environmental information of the occurrence location when and after the on-site safety accident occurs, obtain the time spent by the target person to handle the on-site safety accident and the handling result of the on-site safety accident; Calculating a handling evaluation value of the on-site safety accident based on the time spent by the target person in handling the on-site safety accident and the handling result of the on-site safety accident, wherein the handling evaluation value represents the handling efficiency of the on-site safety accident by the target person; The safety emergency feature information of the target person is updated according to the processing evaluation value.

7. The artificial intelligence-based tunnel construction safety emergency method according to claim 6, characterized in that: The step of calculating the processing evaluation value of the on-site safety accident based on the time spent by the target person in processing the on-site safety accident and the processing result of the on-site safety accident includes: When the result of handling the on-site safety accident is to eliminate the on-site safety accident, the time spent by the target person in handling the on-site safety accident is compared with the average handling time of the on-site safety accident of the same type obtained from the big database, and when the handling time is less than the average handling time, a positive handling evaluation value is obtained; When the processing result of the on-site safety accident is elimination of the on-site safety accident, and the processing time is equal to the average processing time, a maintenance processing evaluation value is obtained; When the result of handling the on-site safety accident is that the on-site safety accident is not eliminated or the handling time is longer than the average handling time, a negative handling evaluation value is obtained; The step of updating the safety emergency feature information of the target person according to the processing evaluation value comprises: When the processing evaluation value is a positive processing evaluation value, increasing the emergency characteristic value in the safety emergency characteristic information of the target person, wherein the magnitude of the emergency characteristic value is positively correlated with the on-site emergency processing capability; When the processing evaluation value is the maintaining processing evaluation value, maintaining the emergency feature value in the safety emergency feature information of the target person unchanged; When the processing evaluation value is a negative processing evaluation value, the emergency feature value in the safety emergency feature information of the target person is reduced.

8. The artificial intelligence-based tunnel construction safety emergency method according to claim 5, characterized in that: The method further comprises the step of training a safety accident prediction model, the step comprising: Creating a training sample set, the training sample set including sample parameters and sample tags, wherein the sample parameters include a combination of environmental parameters and / or video image information under various tunnel environments, and the sample tags include whether it is an on-site safety accident and the type of the on-site safety accident; Inputting the training samples in the training sample set into a neural network in sequence for training, and outputting a first training result of whether the training sample is an on-site safety accident and a second training result of what type of on-site safety accident the training sample is, wherein the neural network includes a judgment subnetwork and a prediction subnetwork; Calculating a training loss function value based on the first training result, the second training result, and the sample label, and adjusting network parameters of the judgment subnetwork and the prediction subnetwork when the loss function value is greater than a preset loss function value; Repeatedly input the training samples in the training sample set into the neural network for training until the loss function value is no greater than the preset loss function value, and the training is completed. The neural network obtained by the training is used as the safety accident prediction model.

9. A tunnel construction safety emergency device based on artificial intelligence, characterized in that: A computer device used in a tunnel construction safety emergency system, wherein the tunnel construction safety emergency system further comprises an image acquisition device and an environmental parameter detection sensor that are communicatively connected to the computer device, wherein the image acquisition device and the environmental parameter detection sensor are arranged in the tunnel for respectively acquiring video image information and environmental information in the tunnel, wherein the device comprises: A first acquisition module is used to acquire video image information acquired by the image acquisition device, obtain position distribution information of people in the tunnel and identity feature information of each person based on the video image information, and obtain safety emergency feature information of the person based on the identity feature information of the person; A second acquisition module is used to acquire environmental information collected by the environmental parameter detection sensor, and determine whether an on-site safety accident is found in the tunnel based on the environmental information and / or the video image information; A judgment module, for identifying the type and location of an on-site safety accident if it is determined that an on-site safety accident has occurred in the tunnel; A selection module, configured to select a target person in the tunnel who has experience in handling the on-site safety accident based on the type of the on-site safety accident, the position distribution information of the people in the tunnel, and the safety emergency feature information of each person; The notification module is used to notify the target person to go to the location where the on-site safety accident occurred to carry out safety emergency response.

10. A computer device, characterized in that: The computer device includes a processor, a readable storage medium and a communication interface. The readable storage medium, the communication interface and the processor are connected via a bus system. The readable storage medium is used to store programs, instructions or codes. The processor is used to execute the programs, instructions or codes in the readable storage medium to execute the artificial intelligence-based tunnel construction safety emergency response method as described in any one of claims 1-8.

Images