Subway construction information identification method, device and equipment and storage medium

Abstract

The application relates to a subway construction information identification method, device and equipment and a storage medium. The method comprises the following steps: obtaining a target image collected from a subway protection area; inputting the target image into a construction behavior identification model for processing, so that the construction behavior identification model extracts features of the image, extracts construction behavior image features, and identifies target construction behavior according to the construction behavior image features; inputting the target image into a construction equipment identification model for processing, so that the construction equipment extracts features of the image, extracts construction equipment image features, and identifies target construction equipment according to the construction equipment image features; and determining a dangerous construction identification result corresponding to the subway protection area based on the target construction behavior and the target construction equipment. The method can improve the identification efficiency.

Description

Technical Field

[0001] This application relates to the field of industrial intelligent technology, and in particular to a method, apparatus, equipment and storage medium for identifying subway construction information. Background Technology

[0002] With socio-economic development and urban progress, major cities are gradually improving their transportation networks to alleviate traffic congestion. Subways, a reliable and weather-independent form of public transportation, are very popular among residents of large cities. However, the continuous construction of subway lines inevitably increases the mileage of subway tunnels. Because the general public is not entirely aware of the locations of subway tunnels, surface construction may jeopardize tunnel safety. Therefore, inspections above the tunnels have become a key focus for the safe operation of the subway.

[0003] Currently, construction inspections are mainly conducted manually, which requires a large number of personnel to patrol the protected areas related to the subway, resulting in low efficiency in construction identification. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for identifying subway construction information, in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a method for identifying subway construction information. The method includes: acquiring a target image collected from a subway protection zone; inputting the target image into a construction behavior recognition model for processing, such that the construction behavior recognition model extracts features from the image to obtain construction behavior image features, and identifies a target construction behavior based on the construction behavior image features; inputting the target image into a construction equipment recognition model for processing, such that the construction equipment extracts features from the image to obtain construction equipment image features, and identifies a target construction equipment based on the construction equipment image features; and determining a hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0006] In one embodiment, determining the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment includes: comparing the target construction equipment with hazardous construction equipment in a pre-set set of hazardous construction equipment; if the set of hazardous construction equipment includes the target construction equipment; and if the target construction behavior is the construction behavior corresponding to the target construction equipment, then it is determined that there is a hazardous construction source in the subway protection zone.

[0007] In one embodiment, determining the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment includes: determining a behavior degree identification model corresponding to the target construction behavior; inputting the target image into the behavior degree identification model for degree identification to identify the degree of construction behavior corresponding to the target construction behavior; determining the construction damage level based on the degree of construction behavior, the target construction behavior, and the target construction equipment; and when the construction damage level exceeds the damage level threshold, the hazardous construction identification result indicates that hazardous construction exists.

[0008] In one embodiment, the method further includes: if the hazardous construction identification result indicates that hazardous construction exists, then sending a hazard warning message to the terminal corresponding to the subway management user, wherein the hazard warning message is used to describe the degree of the construction behavior, the target construction behavior, and the target construction equipment.

[0009] In one embodiment, the method further includes: acquiring target speech corresponding to the subway protection zone, which is synchronously acquired with the target image; inputting the target speech into a speech recognition model, wherein the speech recognition model extracts the timbre features of the template speech, and identifies the speech output device corresponding to the target speech based on the timbre features; determining the speech output duration corresponding to the speech output device, and if the speech output duration is greater than a preset duration threshold, outputting an alarm prompt message corresponding to the speech output device.

[0010] In one embodiment, the method further includes: arranging the target construction behaviors identified by the construction behavior recognition model at each time point in chronological order to obtain a construction behavior sequence; obtaining a standard behavior sequence corresponding to standard construction; comparing the construction behavior sequence with the standard behavior sequence, and if the comparison is consistent, providing a behavior prompt message.

[0011] Secondly, this application also provides a subway construction information identification device. The device includes: a target image acquisition module for acquiring a target image collected from a subway protection zone; a construction behavior identification module for inputting the target image into a construction behavior identification model for processing, so that the construction behavior identification model extracts features from the image to obtain construction behavior image features, and identifies the target construction behavior based on the construction behavior image features; a construction equipment identification module for inputting the target image into a construction equipment identification model for processing, so that the construction equipment extracts features from the image to obtain construction equipment image features, and identifies the target construction equipment based on the construction equipment image features; and a hazardous construction identification result determination module for determining the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0012] In one embodiment, the hazardous construction identification result determination module includes a comparison unit for comparing the target construction equipment with hazardous construction equipment in a pre-set set of hazardous construction equipment; and a hazardous construction source determination unit for determining that a hazardous construction source exists in the subway protection zone if the set of hazardous construction equipment includes the target construction equipment and the target construction behavior is the construction behavior corresponding to the target construction equipment.

[0013] In one embodiment, the hazardous construction identification result determination module is configured to: determine the behavior degree identification model corresponding to the target construction behavior; input the target image into the behavior degree identification model for degree identification, and identify the construction behavior degree corresponding to the target construction behavior; determine the construction damage level based on the construction behavior degree, the target construction behavior, and the target construction equipment; and when the construction damage level exceeds the damage level threshold, the hazardous construction identification result indicates that hazardous construction exists.

[0014] In one embodiment, the device further includes a hazard sending module: if the hazard construction identification result indicates that there is hazard construction, then sending hazard warning information to the terminal corresponding to the subway management user, wherein the hazard warning information is used to describe the degree of the construction behavior, the target construction behavior, and the target construction equipment.

[0015] In one embodiment, the device further includes a speech recognition module for acquiring target speech corresponding to the subway protection zone, which is synchronously acquired with the target image; a speech output device acquisition module for inputting the target speech into a speech recognition model, wherein the speech recognition model extracts the timbre features of the template speech and identifies the speech output device corresponding to the target speech based on the timbre features; and an alarm prompt information output module for determining the speech output duration corresponding to the speech output device, wherein if the speech output duration exceeds a preset duration threshold, an alarm prompt information corresponding to the speech output device is output.

[0016] In one embodiment, the device further includes a construction sequence arrangement module, used to arrange the target construction behaviors identified by the construction behavior recognition model at each time point in chronological order to obtain a construction behavior sequence; a construction sequence acquisition module, used to acquire the standard behavior sequence corresponding to dangerous construction; and a construction sequence comparison module, used to compare the construction behavior sequence with the standard behavior sequence, and if the comparison is consistent, output behavior prompt information.

[0017] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: acquiring a target image collected from a subway protection zone; inputting the target image into a construction behavior recognition model for processing, so that the construction behavior recognition model extracts features from the image to obtain construction behavior image features, and identifies a target construction behavior based on the construction behavior image features; inputting the target image into a construction equipment recognition model for processing, so that the construction equipment extracts features from the image to obtain construction equipment image features, and identifies a target construction equipment based on the construction equipment image features; and determining a hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0018] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps: acquiring a target image collected from a subway protection zone; inputting the target image into a construction behavior recognition model for processing, such that the construction behavior recognition model extracts features from the image to obtain construction behavior image features, and identifies a target construction behavior based on the construction behavior image features; inputting the target image into a construction equipment recognition model for processing, such that the construction equipment extracts features from the image to obtain construction equipment image features, and identifies a target construction equipment based on the construction equipment image features; and determining a hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0019] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps: acquiring a target image collected from a subway protection zone; inputting the target image into a construction behavior recognition model for processing, such that the construction behavior recognition model extracts features from the image to obtain construction behavior image features, and identifies a target construction behavior based on the construction behavior image features; inputting the target image into a construction equipment recognition model for processing, such that the construction equipment extracts features from the image to obtain construction equipment image features, and identifies a target construction equipment based on the construction equipment image features; and determining a hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0020] The aforementioned subway construction information identification method, device, computer equipment, storage medium, and computer program product acquire target images collected from subway protection zones; input the target images into a construction behavior identification model for processing, enabling the model to extract features from the images, thereby identifying the target construction behavior; input the target images into a construction equipment identification model for processing, enabling the construction equipment to extract features from the images, thereby identifying the target construction equipment; and based on the target construction behavior and equipment, determine the hazardous construction identification result corresponding to the subway protection zone. Because images of subway protection zones can be acquired, and the corresponding construction behaviors and equipment can be identified based on the images and corresponding models, and hazard sources can be identified based on the identified construction behaviors and equipment, the hazardous construction sources can be quickly and accurately determined, improving identification efficiency. Attached Figure Description

[0021] Figure 1 This is an application environment diagram of the subway construction information identification method in one embodiment;

[0022] Figure 2 This is a flowchart illustrating a subway construction information identification method in one embodiment;

[0023] Figure 3 This is a flowchart illustrating the steps for determining the hazardous construction identification results corresponding to the subway protection zone for the target construction behavior and target construction equipment in one embodiment.

[0024] Figure 4 This is a flowchart illustrating the steps for determining the hazardous construction identification results corresponding to the subway protection zone based on the target construction behavior and the target construction equipment in one embodiment.

[0025] Figure 5 This is a flowchart illustrating a subway construction information identification method in one embodiment;

[0026] Figure 6 This is a flowchart illustrating a subway construction information identification method in one embodiment;

[0027] Figure 7 This is a structural block diagram of a subway construction information identification device in one embodiment;

[0028] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0030] This application provides a method for identifying subway construction information, which can be applied to, for example... Figure 1In the application environment shown, the camera 102 communicates with the server 104 via a network. A data storage system can store the data that the server 104 needs to process. The data storage system can be integrated onto the server 104 or placed on a cloud or other network server. The terminal device 106 can be used to receive information sent by the server 104. The server 104 acquires images from the camera 102 via the communication network. These images are used by the recognition model in the server 104 to extract features. The extracted features are used by the recognition model in the server 104 for feature recognition to obtain information about construction activities and equipment. Based on the construction activities and equipment, a judgment is made. If a hazard is found in the construction activity or equipment, the construction activity, equipment, and geographical location are sent to the terminal device 106 via the communication network to display relevant information to the user, enabling workers to quickly and accurately determine the dangers posed by construction hazards above the subway tunnel. The camera 102 can be, but is not limited to, a drone or a camera, and the server 104 can be a standalone server or a server cluster consisting of multiple servers. Terminal device 106 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. It is understood that the subway construction information identification method provided in this application can also be executed on other devices, such as in a terminal device.

[0031] In one embodiment, such as Figure 2 As shown, a method for identifying subway construction information is provided, which can be applied to... Figure 1 Taking the server in the example, the following steps are included:

[0032] Step S202: Obtain the target image collected from the subway protection area;

[0033] The metro protection zone is an area related to the safe operation of the metro. Construction within the metro protection zone may affect metro operations. For example, the metro protection zone can be the surface area above the metro, such as above a metro tunnel. The target image is an image obtained by photographing the metro protection zone. Target images can be obtained using cameras installed within the metro protection zone or using drones. No-fly zones utilize camera images to collect information on construction personnel, construction materials, and the site's appearance.

[0034] Specifically, camera equipment can be used to photograph the subway protection zone to obtain target images. The camera equipment can send the captured images to the server in real time. The server can then obtain the target images and analyze whether there are any dangerous construction activities or dangerous construction equipment in the subway protection zone.

[0035] In one embodiment, the target image can be an image from a video, such as an image extracted from a video captured by a drone, which may include at least one type of image, such as equipment or human figures, as the target image. Each target image can be identified to determine the corresponding hazardous construction activity and hazardous construction equipment.

[0036] Step S204: Input the target image into the construction behavior recognition model for processing, so that the construction behavior recognition model can extract features from the image, extract construction behavior image features, and identify the target construction behavior based on the construction behavior image features.

[0037] The construction behavior recognition model can extract features from images corresponding to construction behaviors and identify those behaviors based on the extracted features. This model can be an artificial intelligence model trained on a large number of images of this type. Construction behavior refers to actions performed within a subway protection zone, such as drilling, excavation, or pile driving. Feature extraction involves extracting specific properties of the construction behavior from the image; this model can be a neural network model.

[0038] Specifically, the construction behavior recognition model can include a feature extraction layer and a behavior classification layer. The server obtains the target image from the terminal. The server has a pre-trained construction behavior recognition model with artificial intelligence built in. The feature extraction layer extracts features from the input image. The obtained image features are then fed to the behavior classification layer. The behavior classification layer obtains the probability of the image corresponding to various construction behaviors based on the extracted features. The construction behavior with the highest probability and greater than the probability threshold is selected as the target construction behavior.

[0039] For example, a construction behavior recognition model can be used to identify various behaviors such as drilling, excavation, and pile driving. When a target image is input into the construction behavior recognition model, the model outputs the probability of the behavior in the image being drilling, excavation, and pile driving. Assuming that the probabilities of drilling, excavation, and pile driving are 0.8, 0.15, and 0.05, respectively, and assuming that the preset threshold is 0.6, then since 0.8 is greater than 0.6, the identified target construction behavior is drilling.

[0040] In one embodiment, construction behavior images can be pre-acquired, and the corresponding construction behaviors in these images can be manually labeled, with the labeled behaviors serving as behavior labels for the construction behavior images. The construction behavior images can then be input into a construction behavior recognition model to be trained. The model identifies the construction behavior images, obtaining the probabilities corresponding to the behavior labels. The model loss value is determined based on the magnitude of these probabilities; a higher probability results in a lower loss value. The model parameters can be adjusted in the direction of decreasing the loss value to obtain a trained construction behavior recognition model.

[0041] Step S206: Input the target image into the construction equipment recognition model for processing, so that the construction equipment can extract features from the image, extract the image features of the construction equipment, and identify the target construction equipment based on the image features of the construction equipment.

[0042] The construction equipment recognition model can extract features from images corresponding to construction equipment and identify the equipment based on these extracted features. This model is an artificial intelligence model trained on a large number of images of this type. Construction equipment refers to equipment used for construction within a subway protected area; for example, it can include at least one of the following: drilling rigs, excavators, or pile drivers. Feature extraction extracts specific properties of the construction equipment from the image; this model is a neural network model.

[0043] Specifically, the construction equipment recognition model can include a feature extraction layer and an equipment classification layer. The server obtains the target image from the terminal. The server has a pre-trained construction equipment recognition model with artificial intelligence built in. The feature extraction layer extracts features from the input image, and the obtained image features are further fed to the equipment classification layer. The equipment classification layer obtains the probability of the image corresponding to various construction equipment based on the extracted features, and selects the construction equipment with the highest probability and greater than the probability threshold as the target construction equipment.

[0044] For example, a construction equipment recognition model can be used to identify various equipment such as drilling rigs, excavators, and pile drivers. When a target image is input into the construction equipment recognition model, the model outputs the probability that the equipment in the image is a drilling rig, an excavator, or a pile driver. Assuming that the probabilities of the drilling rig, excavator, and pile driver are 0.7, 0.2, and 0.1, respectively, and assuming that the preset threshold is 0.6, then since 0.7 is greater than 0.6, the identified target construction equipment is a drilling rig.

[0045] In one embodiment, images of construction equipment can be pre-acquired, and the corresponding construction equipment in the images can be manually labeled, with the labeled construction equipment serving as equipment tags for the construction equipment images. The construction equipment images can be input into a construction equipment recognition model to be trained. The model identifies the construction equipment images, obtaining the probability corresponding to each equipment tag. The model loss value is determined based on the magnitude of the probability corresponding to each equipment tag; a higher probability results in a lower loss value. The model parameters can be adjusted in the direction of decreasing the loss value to obtain a trained construction equipment recognition model.

[0046] Step S208: Determine the hazardous construction identification results corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0047] The hazardous construction identification result refers to the identification result of whether the target construction behavior and target construction equipment pose a danger to the subway protection zone. The hazard identification result can be either "there is a danger" or "there is no danger." For example, if an excavator is digging above, then it is hazardous construction.

[0048] The target construction activity refers to any activity above the subway protection zone, which is not limited to drilling, excavation, or pile driving. This activity can be defined differently depending on the protection requirements. The target construction equipment refers to any equipment above the subway protection zone, which is not limited to drilling rigs, excavators, or pile drivers. This equipment can be defined differently depending on the protection requirements. The identification result indicates the impact of this construction activity or equipment on the subway tunnel.

[0049] Specifically, the target construction activity is compared with hazardous construction activities, and the target construction equipment is compared with hazardous construction equipment. This comparison can be between individual construction activities or individual construction equipment, or both. If they match, a hazard exists. If they do not match, no hazard exists. If a hazard exists, information such as the construction activity, construction equipment, specific location, and extent of damage is provided. If no hazard source is found, the identification process continues.

[0050] In one embodiment, the construction behavior recognition model detects drilling activity and compares it with preset dangerous construction behaviors. If the comparison results are the same, a signal is immediately sent to the alarm center. The alarm center connects to the terminal on the hand of the inspection personnel and sends information such as construction behavior, construction equipment, specific location, and degree of damage to notify the inspection personnel to handle the situation in a timely manner.

[0051] In the aforementioned method for identifying subway construction information, target images are acquired by collecting data from a subway protection zone. These images are then input into a construction behavior recognition model for processing, enabling the model to extract features from the images and identify the target construction behavior. The target images are then input into a construction equipment recognition model for processing, allowing the model to extract features from the images and identify the target construction equipment. Finally, the method determines the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment. Because images of the subway protection zone can be acquired, and the corresponding construction behavior and equipment can be identified based on the images and their corresponding models, and hazard sources can be identified based on the identified construction behavior and equipment, the hazardous construction sources can be quickly and accurately determined, improving identification efficiency.

[0052] In one embodiment, such as Figure 3 As shown, the hazardous construction identification results corresponding to the subway protection zone, based on the target construction behavior and target construction equipment, include:

[0053] Step S302: Compare the target construction equipment with the hazardous construction equipment in the pre-set set of hazardous construction equipment.

[0054] The "dangerous construction equipment" category includes pre-defined construction equipment that poses a danger to subway operation. The specific equipment can be selected based on needs; for example, dangerous construction equipment can be, but is not limited to, at least one of drilling rigs, excavators, or pile drivers. Non-dangerous construction equipment could include, for example, road leveling machines.

[0055] Specifically, a set of hazardous construction equipment is pre-set in the server. When a relevant machine enters the subway protection zone and carries out construction, the type of equipment is obtained based on the target image, and compared with the set of hazardous construction equipment to determine whether the feature recognition result is the same as that of the set of hazardous construction equipment.

[0056] In one embodiment, the subway control center pre-sets, but is not limited to, a set of drilling machines, excavators, and pile drivers. For example, if a drilling machine is identified above a subway tunnel, the output will be compared with the set mentioned above.

[0057] Step S304: If the set of hazardous construction equipment includes the target construction equipment, and the target construction behavior is the construction behavior corresponding to the target construction equipment, then it is determined that there is a hazardous construction source in the subway protection zone.

[0058] The target construction activity refers to the construction activity corresponding to the target construction equipment. Specifically, it refers to the activity performed by the target construction equipment; that is, in this image, the target construction equipment is performing relevant construction activities within the subway protection zone, such as an excavator digging or a pile driver driving piles. The subway protection zone is the area on the ground surface above the subway protection zone. A hazardous construction source is an activity involving hazardous construction equipment and hazardous construction activities.

[0059] Specifically, the construction equipment identification model identifies construction equipment, and the construction behavior identification model also identifies construction behavior. If these behaviors are related to the identified equipment, then it is determined whether the construction equipment is dangerous and whether the construction behavior is dangerous. If both are determined to be dangerous, then this is considered a dangerous construction source, and a signal is sent to the alarm center. For example, if a pile driver is placed above a subway tunnel and performs pile driving, this indicates that the target construction behavior corresponds to the target construction equipment. If the pile driver is above the subway tunnel but does not perform pile driving, then the combined condition of both is not met.

[0060] In one embodiment, construction is underway above a subway protection zone. The camera captures images that identify the construction equipment as a pile driver and the construction activity as pile driving. When both are identified, the server considers this a dangerous construction source and sends a signal to the alarm center.

[0061] In this embodiment, by jointly comparing the target construction equipment and the target construction behavior, it is possible to further determine that the construction is a dangerous construction, thereby improving the accuracy of identification.

[0062] In one embodiment, such as Figure 4 As shown, the hazardous construction identification results corresponding to the subway protection zone, based on the target construction behavior and target construction equipment, include:

[0063] Step S402: Determine the behavior degree recognition model corresponding to the target construction behavior.

[0064] The target construction behavior refers to the construction actions performed by the target construction equipment. The behavior degree recognition model can extract features from images to obtain construction behavior features that reflect the degree of construction. It can also identify the degree of the construction behavior based on the extracted features; for example, the digging degree of an excavator can be categorized as high, medium, or low. The behavior degree recognition model is an artificial intelligence model trained on a large number of images corresponding to this construction behavior and their corresponding behavior degree labels.

[0065] Specifically, the behavior degree recognition model has a behavior degree classification layer. The server obtains the target image, and the server has a pre-trained behavior degree recognition model with artificial intelligence built in. The obtained image features are further fed to the behavior degree classification layer. The device classification layer derives the corresponding probability for each behavior degree based on the image, and selects the construction behavior degree with the highest probability and greater than the probability threshold as the target behavior degree.

[0066] For example, different construction behaviors correspond to different behavior degree models. For instance, the behavior degree recognition model can include various models such as the model for recognizing the degree of drilling by a drilling rig, the degree of excavation by an excavator, and the degree of pile driving by a pile driver. The target image is input into the behavior degree recognition model corresponding to the target construction behavior, and the behavior degree recognition model outputs the probability of various degrees corresponding to the behavior in the image. Assuming that the probabilities of high, medium, and low degrees of drilling by a drilling rig are 0.9, 0.05, and 0.05, respectively, the recognized degree of the target construction behavior is high.

[0067] Step S404: Input the target image into the behavior degree recognition model for degree recognition, and identify the degree of construction behavior corresponding to the target construction behavior.

[0068] The degree of construction behavior refers to the extent to which the equipment's behavior reaches a certain level, which can be divided into three grades: high, medium, and low, with the corresponding destructiveness decreasing accordingly. For example, a high degree of construction behavior generally corresponds to high destructiveness.

[0069] Specifically, the target image is input into the behavior degree recognition model, which will judge the degree of construction behavior based on the image and obtain the probabilities corresponding to high, medium and low degrees. For example, if the behavior degree model identifies the excavator construction degree as high, medium and low with probabilities of 0.1, 0.1 and 0.8 respectively, then the construction degree is judged as low.

[0070] In one embodiment, there is complex construction above the subway protection zone, with construction machinery including pile drivers. The recognition model identifies the level of construction based on captured images. If the equipment is engaged in construction, the corresponding level of construction is indicated; otherwise, monitoring continues.

[0071] Step S406: Determine the level of construction damage based on the degree of construction behavior, the target construction behavior, and the target construction equipment.

[0072] The damage level is classified according to the destructiveness of the construction equipment to the subway protection zone. The destructiveness is divided into three levels: high, medium and low, which reflect the degree of damage caused by the construction equipment to the protection zone and the degree of harm to the subway operation.

[0073] The degree of construction behavior, construction behavior, and construction equipment corresponding to each level of construction damage are pre-set. That is, each level of damage is determined by the combination of the degree of behavior, construction behavior, and construction equipment. For example, the level corresponding to the behavior degree is low, the construction behavior is drilling, and the construction equipment is drilling equipment is set.

[0074] Specifically, the destructiveness of the construction area is judged based on the results of three identifications: the degree of construction activity in the construction area, the target construction activity, and the target construction equipment.

[0075] In one embodiment, the captured images are input into different models to obtain the degree of construction behavior, the target construction behavior, and the target construction equipment. The system then analyzes the data and determines the level of destructiveness, outputting both single-factor and combined destructiveness. Single-factor destructiveness refers to a determination based solely on one of the construction behavior or the construction equipment.

[0076] In one embodiment, the construction damage level can be determined by combining the degree of construction behavior corresponding to multiple target images, the target construction behavior, and the target construction equipment. For example, a behavior sequence corresponding to each damage level is pre-set. When the behavior identified from the target image is sorted by time and is the same as the behavior sequence, it is determined whether the degree of the behavior identified from the target image and the construction equipment are the same as the degree of the behavior and the construction equipment of the damage level corresponding to the behavior sequence. If they are the same, then it is determined to be that damage level.

[0077] Step S408: When the construction damage level exceeds the damage level threshold, the hazardous construction identification result is that hazardous construction exists.

[0078] The damage level threshold refers to the minimum value for construction damage. Values ​​above this threshold indicate destructive potential, while values ​​below it indicate potential damage requiring enhanced monitoring. Hazardous construction refers to construction that damages the subway protection zone; values ​​exceeding the damage level threshold classify it as hazardous construction.

[0079] Specifically, when the damage level exceeds a set threshold, the construction is considered hazardous. If the damage level is below the threshold but construction activity is still taking place, the construction will be closely monitored; once it exceeds the threshold, it will be considered hazardous. For example, the area corresponding to the construction activity can be designated as a key detection area, increasing the frequency of image capture in this area. Alternatively, a fixed-area detection command can be sent to a drone, instructing it to fly to the key detection area and take pictures. The server then uses the captured images to continue detecting construction activity and equipment, thereby improving detection efficiency.

[0080] In one embodiment, when there is road leveling construction above a subway protection zone, the image input into the model does not identify any target construction equipment, so the construction is not considered a dangerous construction.

[0081] In this embodiment, by using a construction degree identification model to identify the construction degree, it is possible to quickly determine whether the construction behavior is at a high, medium, or low level, providing an important reference for providing information on the degree of damage and repair.

[0082] In one embodiment, if the hazardous construction identification result indicates that hazardous construction exists, a hazard warning message is sent to the terminal corresponding to the subway management user. The hazard warning message describes the degree of construction behavior, the target construction behavior, and the target construction equipment.

[0083] Among them, subway management users refer to the inspection personnel of the subway protection zone, and the terminal refers to the equipment carried by the inspection personnel, which can display information such as the degree of construction behavior, the target construction behavior, and the target construction equipment.

[0084] Specifically, when the identification model detects a hazardous construction source, it sends information such as the extent of the construction activity, the target construction activity, and the target construction equipment to subway inspectors via the server, along with the specific location, and automatically assesses the hazard level of the hazardous construction source. For example, if a drilling rig is conducting drilling operations, the system will identify the hazardous construction source by inputting data from a camera terminal, immediately sending an alarm signal to the inspectors and marking the real-time location of the source.

[0085] In one embodiment, such as Figure 5 As shown, the method also includes:

[0086] Step S502: Obtain the target speech corresponding to the subway protected area, which is synchronously acquired with the target image.

[0087] The target speech refers to the speech collected simultaneously with the video collected above the subway protection zone. This speech is synchronized with the video and can assist in identifying drilling, excavation, and pile driving.

[0088] Specifically, while capturing video, the audio capture device is turned on. This device has an audio filtering effect, which can filter out sounds other than those emitted by the device, resulting in audio that is synchronized with the target image.

[0089] In one embodiment, the drone and camera are equipped with a voice acquisition device. When the voice acquisition device is turned on while shooting, the voice can be obtained in sync with the image.

[0090] Step S504: Input the target speech into the speech recognition model. The speech recognition model extracts the timbre features of the template speech and identifies the speech output device corresponding to the target speech based on the timbre features.

[0091] The speech recognition model can extract features from the speech corresponding to construction equipment, and can also identify the construction equipment based on the extracted speech features. Timbre features are characteristics that reflect the perceptual properties of sound; the speech recognition model is an artificial intelligence model trained on a large number of such speech samples. Construction speech refers to the sounds emitted by equipment used for construction in subway protected areas; for example, construction equipment may include at least one of the following: drilling rigs, excavators, or pile drivers. Feature extraction involves extracting the unique sound properties of the construction equipment from the speech.

[0092] Specifically, the speech recognition model can include a feature extraction layer and a device layer. The server obtains the target speech from the terminal. The server has a pre-trained speech recognition model with artificial intelligence built in. The feature extraction layer extracts features from the input speech. The obtained speech features are then fed to the device classification layer. The device classification layer obtains the probability that the speech corresponds to various construction equipment based on the extracted features. The construction equipment with the highest probability and greater than the probability threshold is selected as the target construction equipment.

[0093] For example, a speech recognition model can be used to identify various sounds such as drilling machines, excavators, and pile drivers. When the target speech is input into the speech recognition model, the speech recognition model outputs the probability that the device that produced the speech is a drilling machine, an excavator, or a pile driver.

[0094] Step S506: Determine the voice output duration corresponding to the voice output device. If the voice output duration exceeds the preset duration threshold, output the alarm prompt information corresponding to the voice output device.

[0095] Among them, the voice output device is a device that outputs sound. The duration threshold refers to a threshold set for the duration of voice output. Exceeding this threshold will trigger an alarm, while not exceeding this duration indicates normal operation.

[0096] Specifically, the speech output devices of the speech recognition model at various times can be obtained to determine the duration of speech output from each device. By comparing the output duration with a preset duration threshold, if the speech duration exceeds the threshold, it indicates that the construction equipment has been operating for a long time and may cause damage. If it is shorter than the threshold, it indicates that the construction equipment has not yet reached the point of causing damage.

[0097] In one embodiment, a voice recording device from a drone or camera captures a segment of sound. After the voice recognition model identifies the sound as an excavator digging, if the duration of the voice recording exceeds a set threshold, the excavator may have dug to a relatively deep location, potentially causing damage to the subway tunnel. If the duration is shorter than the threshold, it indicates that no damage has occurred, but work needs to be stopped immediately.

[0098] In this embodiment, by simultaneously acquiring audio while acquiring images, it is possible to assist in judging construction activities from the perspective of audio, and prevent dangerous construction activities from being missed due to incorrect image acquisition.

[0099] In one embodiment, such as Figure 6 As shown, the method also includes:

[0100] Step S602: Arrange the target construction behaviors identified by the construction behavior recognition model at each time point in chronological order to obtain a construction behavior sequence.

[0101] Among them, arranging construction activities in chronological order refers to arranging construction activities identified at different times in chronological order, while construction activity sequence refers to a sequence of construction activities arranged in chronological order.

[0102] Specifically, construction behaviors are obtained by inputting images taken at different times into the construction behavior recognition model. The recognized construction behaviors are arranged in chronological order, and the resulting sequence is a set of each construction behavior. Non-construction behaviors are not included in the arrangement.

[0103] In one embodiment, the excavator first drills a hole, then digs, and finally transports the excavated soil to a designated location for stockpiling. These steps are arranged in chronological order, resulting in the sequence: drilling, digging, and stockpiling.

[0104] Step S604: Obtain the standard behavior sequence corresponding to hazardous construction.

[0105] The standard behavior sequence refers to a sequence obtained from the standard construction behaviors in chronological order.

[0106] Specifically, the sequence obtained by following the standard construction order from first to last is input into the server to obtain a standard behavior sequence. This standard behavior sequence is a pre-set sequence that represents the order of dangerous construction actions.

[0107] In one embodiment, the standard sequence for an excavator to excavate is drilling, digging, and stacking, and this sequence is input into the server.

[0108] Step S606: Compare the construction behavior sequence with the standard behavior sequence. If the comparison is consistent, output the behavior prompt information.

[0109] Among them, the construction behavior sequence refers to a construction sequence in which the actual construction behaviors are arranged in chronological order from first to last.

[0110] Specifically, the construction behaviors identified by the construction behavior recognition model are arranged in chronological order to obtain the construction behavior sequence. This sequence is then compared with a standard construction behavior sequence. If they match, it indicates that the construction is hazardous.

[0111] In one embodiment, an excavator performs digging, and a construction behavior recognition model identifies the behaviors of drilling, digging, and stacking. The drilling, digging, and stacking are arranged in chronological order and then compared with a standard construction behavior sequence. If they are found to be consistent, a behavior prompt message is output.

[0112] In this embodiment, by comparing the construction behavior sequence with the standard behavior sequence, non-standard construction can be identified from the construction sequence, and prompt information can be output to the corresponding terminal to indicate that there is a danger in carrying out the behavior sequence.

[0113] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0114] Based on the same inventive concept, this application also provides a subway construction information identification device for implementing the subway construction information identification method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more subway construction information identification device embodiments provided below can be found in the limitations of the subway construction information identification method described above, and will not be repeated here.

[0115] In one embodiment, such as Figure 7As shown, a subway construction information recognition device is provided, including: an image acquisition module, a construction behavior recognition module, a construction equipment recognition module, and a hazardous construction recognition result determination module, wherein:

[0116] Image acquisition module 702 is used to acquire target images collected from the subway protection zone;

[0117] The construction behavior recognition module 704 is used to input the target image into the construction behavior recognition model for processing, so that the construction behavior recognition model can extract features from the image, extract construction behavior image features, and recognize the target construction behavior based on the construction behavior image features;

[0118] The construction equipment identification module 706 is used to input the target image into the construction equipment identification model for processing, so that the construction equipment can extract features from the image, extract the construction equipment image features, and identify the target construction equipment based on the construction equipment image features;

[0119] The hazardous construction identification result determination module 708 is used to determine the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment.

[0120] In one embodiment, the hazardous construction identification result determination module includes:

[0121] The comparison unit compares the target construction equipment with the hazardous construction equipment in a pre-set set of hazardous construction equipment.

[0122] The hazardous construction source determination unit is used to determine that a hazardous construction source exists in the subway protection zone if the set of hazardous construction equipment includes the target construction equipment and the target construction behavior is the construction behavior corresponding to the target construction equipment.

[0123] In one embodiment, the hazardous construction identification result determination module is used to: determine the behavior degree recognition model corresponding to the target construction behavior; input the target image into the behavior degree recognition model for degree recognition, and identify the degree of construction behavior corresponding to the target construction behavior; determine the construction damage level based on the degree of construction behavior, the target construction behavior, and the target construction equipment; when the construction damage level exceeds the damage level threshold, the hazardous construction identification result is that hazardous construction exists.

[0124] In one embodiment, the device further includes a hazard sending module: if the hazard construction identification result indicates that there is hazard construction, it sends a hazard warning message to the terminal corresponding to the subway management user. The hazard warning message describes the degree of construction behavior, the target construction behavior, and the target construction equipment.

[0125] In one embodiment, the device further includes a speech recognition module for acquiring target speech corresponding to the subway protection zone obtained synchronously with the target image; a speech output device acquisition module for inputting the target speech into a speech recognition model, wherein the speech recognition model extracts the timbre features of the template speech and identifies the speech output device corresponding to the target speech based on the timbre features; and an alarm prompt information output module for determining the speech output duration corresponding to the speech output device, wherein if the speech output duration exceeds a preset duration threshold, an alarm prompt information corresponding to the speech output device is output.

[0126] In one embodiment, the device further includes a construction sequence arrangement module, used to arrange the target construction behaviors identified by the construction behavior recognition model at each time point in chronological order to obtain a construction behavior sequence; a construction sequence acquisition module, used to acquire the standard behavior sequence corresponding to dangerous construction; and a construction sequence comparison module, used to compare the construction behavior sequence with the standard behavior sequence, and if the comparison is consistent, output behavior prompt information.

[0127] Each module in the aforementioned subway construction information identification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0128] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores data to be processed. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for identifying subway construction information.

[0129] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0130] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0131] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0132] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps described in the above method embodiments. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0133] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0134] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for identifying information on subway construction, characterized by, The method includes: Acquire target images collected from the subway protection zone; The target image is input into the construction behavior recognition model for processing, so that the construction behavior recognition model can extract features from the image, extract construction behavior image features, and identify the target construction behavior based on the construction behavior image features; The target image is input into the construction equipment recognition model for processing, so that the construction equipment can extract features from the image, obtain construction equipment image features, and identify the target construction equipment based on the construction equipment image features; Determining the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment includes: determining the behavior degree recognition model corresponding to the target construction behavior; inputting the target image into the behavior degree recognition model for degree recognition to identify the degree of construction behavior corresponding to the target construction behavior; determining the construction damage level based on the degree of construction behavior, the target construction behavior, and the target construction equipment; when the construction damage level exceeds the damage level threshold, the hazardous construction identification result indicates that hazardous construction exists, and a hazard warning message is fed back based on the hazardous construction identification result. The method further includes: The target construction behaviors identified by the construction behavior recognition model at each time point are arranged in chronological order to obtain a construction behavior sequence; the construction behavior sequence is obtained after filtering out invalid construction behaviors. Obtain the standard behavioral sequence corresponding to hazardous construction; The construction behavior sequence is compared with the standard behavior sequence. If the comparison is consistent, a behavior prompt message is output. The method further includes: Acquire the target speech corresponding to the subway protection zone, which is acquired synchronously with the target image; the target speech is obtained by filtering out sounds emitted by non-construction equipment in the subway protection zone. The target speech is input into a speech recognition model, which extracts the timbre features of the target speech and identifies the speech output device corresponding to the target speech based on the timbre features. Determine the voice output duration corresponding to the voice output device. If the voice output duration exceeds a preset duration threshold, output the alarm prompt information corresponding to the voice output device.

2. The method of claim 1, wherein, Based on the target construction behavior and the target construction equipment, the hazardous construction identification results corresponding to the subway protection zone include: The target construction equipment is compared with the hazardous construction equipment in a pre-set set of hazardous construction equipment. If the set of hazardous construction equipment includes the target construction equipment, and the target construction behavior is the construction behavior corresponding to the target construction equipment, then it is determined that there is a hazardous construction source in the subway protection zone.

3. The method of claim 1, wherein, The method further includes: If the hazardous construction identification result indicates that hazardous construction exists, a hazard warning message is sent to the terminal corresponding to the subway management user. The hazard warning message describes the degree of the construction behavior, the target construction behavior, and the target construction equipment.

4. A subway construction information identification device, characterized in that, The device includes: The image acquisition module is used to acquire target images collected from the subway protection zone; The construction behavior recognition module is used to input the target image into the construction behavior recognition model for processing, so that the construction behavior recognition model can extract features from the image, extract construction behavior image features, and recognize the target construction behavior based on the construction behavior image features; The construction equipment identification module is used to input the target image into the construction equipment identification model for processing, so that the construction equipment can extract features from the image, extract the construction equipment image features, and identify the target construction equipment based on the construction equipment image features; The hazardous construction identification result determination module is used to determine the hazardous construction identification result corresponding to the subway protection zone based on the target construction behavior and the target construction equipment. Specifically, the hazardous construction identification result determination module is used to determine the behavior degree identification model corresponding to the target construction behavior; input the target image into the behavior degree identification model for degree identification, and identify the construction behavior degree corresponding to the target construction behavior; determine the construction damage level based on the construction behavior degree, the target construction behavior, and the target construction equipment; when the construction damage level exceeds the damage level threshold, the hazardous construction identification result indicates that hazardous construction exists, and a hazard warning message is fed back based on the hazardous construction identification result; Also includes: The construction sequence arrangement module is used to arrange the target construction behaviors identified by the construction behavior recognition model at each time point in chronological order to obtain a construction behavior sequence; the construction behavior sequence is obtained after filtering out invalid construction behaviors; The construction sequence acquisition module is used to acquire the standard behavior sequence corresponding to hazardous construction. The construction sequence comparison module is used to compare the construction behavior sequence with the standard behavior sequence. If the comparison is consistent, a behavior prompt message is output. Also includes: A speech recognition module is used to acquire target speech corresponding to the subway protection zone, which is synchronously acquired with the target image; the target speech is obtained by filtering out sounds emitted by non-construction equipment in the subway protection zone; the target speech is input into a speech recognition model, which extracts the timbre features of the target speech and identifies the speech output device corresponding to the target speech based on the timbre features; the speech output duration corresponding to the speech output device is determined, and if the speech output duration is greater than a preset duration threshold, an alarm prompt message corresponding to the speech output device is output. 5.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-4 when the computer program is executed by the processor. When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

7. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images