Vehicle flow statistical method, device and equipment based on vibration signals and storage medium

Abstract

The invention provides a traffic flow statistical method, device and equipment based on a vibration signal and a storage medium, and belongs to the technical field of traffic flow statistics, and the method comprises the steps: obtaining a space-time vibration response diagram of a section region; the space-time vibration response diagram comprises a plurality of initial vehicle driving tracks; determining at least two to-be-selected driving tracks conforming to a signal attenuation rule and at least one first target driving track not conforming to the signal attenuation rule in the plurality of initial vehicle driving tracks, and determining a second target driving track with the maximum waveform amplitude in the at least two to-be-selected driving tracks; performing track straight line extraction on the first target driving track and the second target driving track to obtain a plurality of extraction tracks; and taking each extracted track as a vehicle object, and determining the traffic flow of the section area based on the vehicle object. According to the invention, the situation that one vehicle corresponds to multiple tracks is avoided, and the accuracy of the counted traffic flow is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle flow statistics, and in particular to a vehicle flow statistics method, device, equipment and storage medium based on vibration signals. Background Art

[0002] The speed of vehicles on highways is high and the density of vehicles is high. Once a traffic accident occurs, the consequences are often serious. Therefore, ensuring traffic safety is the primary task of highway operations. Accurate detection of traffic flow is crucial for timely detection of abnormal situations such as traffic congestion and potential accidents. Through reasonable traffic flow monitoring, traffic control and diversion measures can be taken in advance to reduce the risk of accidents and improve road safety. In addition, knowing and optimizing traffic flow distribution and reasonably arranging road resources can reduce vehicle delays on highways, improve transportation efficiency, and reduce transportation costs, which is of great significance to promoting economic development.

[0003] At present, the mainstream methods of vehicle flow detection include video detection, laser radar, etc. For video detection, when encountering complex traffic environments, such as bad weather and dense vehicles, large detection errors are likely to occur. For detection equipment such as laser radar, the cost is relatively high. If the vehicle flow statistics work is to be completed all the time, in all areas and in all weather conditions, the economic cost required is too high, which makes it impossible to implement it on a large scale. Grating array vibration sensing technology has developed rapidly in recent years. Compared with traditional electrical sensors, it has the characteristics of small size, anti-electromagnetic interference, high temperature resistance, high precision, low cost, and "all-time, all-domain and all-weather" vibration signal perception. It is very suitable for long-distance mobile object detection and is currently widely used in aerospace, smart ships, smart security, and smart transportation. By laying the vibration optical cable underground, long-distance vibration object recognition can be achieved, and it is also highly adaptable to full-time and full-domain vehicle flow detection applications. However, there are the following technical problems when detecting vehicle flow based on grating array vibration sensing technology: the vibration signal will propagate in the medium, that is, when detecting a vehicle based on a certain measuring point, other measuring points will form false trajectories due to the propagation of vehicle vibration, resulting in inaccurate vehicle flow statistics.

[0004] Therefore, there is an urgent need to provide a vehicle flow statistics method, device, equipment and storage medium based on vibration signals to eliminate the adverse effects of false trajectories on vehicle flow results and improve the accuracy of vehicle flow statistics results in cross-sectional areas. Summary of the invention

[0005] In view of this, it is necessary to provide a vehicle flow statistics method, device, equipment and storage medium based on vibration signals to solve the technical problem that the propagation characteristics of vibration signals are not considered in the prior art, resulting in inaccurate vehicle flow statistics results.

[0006] On the one hand, in order to solve the above technical problems, the present invention provides a vehicle flow statistics method based on vibration signals, comprising: Obtaining a spatiotemporal vibration response diagram of a cross-sectional area; the spatiotemporal vibration response diagram comprising a plurality of initial vehicle driving trajectories; Determine at least two candidate driving trajectories that conform to the signal attenuation law and at least one first target driving trajectory that does not conform to the signal attenuation law among the multiple initial vehicle driving trajectories, and determine a second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories; Extracting straight lines from the first target driving trajectory and the second target driving trajectory to obtain a plurality of extracted trajectories; Each of the extracted trajectories is regarded as a vehicle object, and the vehicle flow rate of the cross-sectional area is determined based on the vehicle object.

[0007] In a possible implementation, the step of obtaining a spatiotemporal vibration response diagram of a cross-sectional area includes: Acquire global vehicle vibration signals collected by grating array vibration sensing optical cables; Determine a target vehicle vibration signal in the global vehicle vibration signal based on the cross-sectional area; the target vehicle vibration signal includes a plurality of target vehicle vibration sub-signal sequences corresponding to each measuring point of the grating array vibration sensing optical cable, and the target vehicle vibration sub-signal sequence includes a plurality of target vehicle vibration sub-signals corresponding to a plurality of acquisition time points; The target vehicle vibration signals are sorted with the spatial positions of the measuring points as rows and the time as columns to obtain the spatiotemporal vibration response diagram.

[0008] In a possible implementation, the signal attenuation law is:

[0009] In the formula, is the average amplitude of the first candidate driving trajectory; is the distance between the first candidate driving trajectory and the second candidate driving trajectory; The distance from the first candidate driving trajectory is The average amplitude of the second candidate driving trajectory; is the attenuation coefficient; is the natural logarithm.

[0010] In a possible implementation manner, after acquiring the spatiotemporal vibration response diagram of the cross-sectional area, the method further includes: Acquire a fully trained image enhancement model, wherein the image enhancement model includes a generator and a discriminator, and the generator includes a feature extraction unit, a feature attention learning unit, and an enhancement unit; Performing feature extraction on the spatiotemporal vibration response graph based on the feature extraction unit to obtain initial image features; Based on the feature attention learning unit, channel attention learning and spatial attention learning are performed on the initial image features in sequence to obtain image attention features; Based on the enhancement unit, feature reconstruction is performed on the image attention feature to obtain an enhanced spatiotemporal vibration response map.

[0011] In a possible implementation, the feature attention learning unit includes a spatial attention learning layer and a channel attention learning layer; the step of sequentially performing channel attention learning and spatial attention learning on the initial image features based on the feature attention learning unit to obtain image attention features includes: Performing channel feature learning on the initial image features based on the channel attention learning layer to obtain channel attention features; Multiplying the channel attention feature with the initial image feature to obtain a channel image feature; Performing spatial feature learning on the channel image features based on the spatial attention learning layer to obtain spatial attention features; Multiply the spatial attention feature by the channel image feature to obtain the image attention feature.

[0012] In a possible implementation, the plurality of extracted trajectories include a first extracted trajectory at a current moment and a second extracted trajectory at a historical moment; before taking each of the extracted trajectories as a vehicle object, the method further includes: Acquire a first trajectory feature of the first extracted trajectory and a second trajectory feature of the second extracted trajectory; Determining feature loss values ​​of the first extracted trajectory and the second extracted trajectory based on the first trajectory feature and the second trajectory feature; When the feature loss value is less than a loss threshold, merging the first extracted trajectory and the second extracted trajectory to obtain a merged trajectory; Then, each of the extracted trajectories is taken as a vehicle object, specifically: The merged trajectory is used as the vehicle object.

[0013] In a possible implementation, the first trajectory feature and the second trajectory feature both include speed, energy, lane, and position; and determining the feature loss values ​​of the first extracted trajectory and the second extracted trajectory based on the first trajectory feature and the second trajectory feature includes: Determining a speed loss value, an energy loss value, a lane loss value, and a position loss value for the first extracted trajectory and the second extracted trajectory; The speed loss value, the energy loss value, the lane loss value and the position loss value are weighted based on preset weights to obtain the characteristic loss value.

[0014] On the other hand, the present invention also provides a vehicle flow statistics device based on vibration signals, comprising: A time-space vibration response map acquisition unit, used to acquire a time-space vibration response map of a cross-sectional area; the time-space vibration response map includes a plurality of initial vehicle driving trajectories; A driving trajectory judgment unit is used to determine at least two candidate driving trajectories that conform to the signal attenuation law and at least one first target driving trajectory that does not conform to the signal attenuation law among the multiple initial vehicle driving trajectories; and determine a second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories; A trajectory straight line extraction unit, used for performing trajectory straight line extraction on the first target driving trajectory and the second target driving trajectory to obtain a plurality of extracted trajectories; The vehicle flow determination unit is used to take each of the extracted trajectories as a vehicle object and determine the vehicle flow of the cross-sectional area based on the vehicle object.

[0015] On the other hand, the present invention also provides an electronic device, including a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the vehicle flow statistics method based on vibration signals described in any one of the possible implementations above.

[0016] On the other hand, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can implement the steps of the vehicle flow statistics method based on vibration signals described in any of the above possible implementation methods.

[0017] The beneficial effect of the present invention is that the vehicle flow statistics method based on vibration signals provided by the present invention determines multiple initial vehicle driving trajectories in the spatiotemporal vibration response diagram based on the signal attenuation law, and obtains at least two candidate driving trajectories that conform to the signal attenuation law. The characteristic that the energy of the vibration signal will attenuate due to scattering and absorption in the medium is utilized to avoid the situation where one vehicle corresponds to multiple trajectories, thereby improving the accuracy of the determined second target form trajectory, and thus improving the accuracy of the statistical vehicle flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic flow chart of an embodiment of a vehicle flow statistics method based on vibration signals provided by the present invention; Figure 2 This is a comparison diagram of the trajectory clarity before and after the straight line extraction of the present invention; Figure 3 This is a flow chart of an embodiment of step S101 of the present invention; Figure 4 A schematic diagram of a flow chart of an embodiment of enhancing a spatiotemporal vibration response diagram provided by the present invention; Figure 5 A schematic diagram of the structure of an embodiment of the image enhancement model provided by the present invention; Figure 6 A schematic diagram of the structure of an embodiment of a feature attention learning unit provided by the present invention; Figure 7 A schematic diagram of the structure of an embodiment of matching and merging extracted trajectories provided by the present invention; Figure 8 is a flow chart of an embodiment of step S702 of the present invention; Fig. 9 A schematic diagram of the structure of an embodiment of a vehicle flow counting device based on vibration signals provided by the present invention; Fig.10 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0021] It should be understood that the schematic drawings are not drawn to scale. The flowchart used in the present invention shows the operations implemented according to some embodiments of the present invention. It should be understood that the operations of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present invention, can add one or more other operations to the flowchart, and can also remove one or more operations from the flowchart. Some of the block diagrams shown in the accompanying drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor systems and / or microcontroller systems.

[0022] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0023] The present invention provides a vehicle flow statistics method, device, equipment and storage medium based on vibration signals, which are described below respectively.

[0024] Figure 1 A flow chart of an embodiment of a vehicle flow statistics method based on vibration signals provided by the present invention is shown in FIG. Figure 1 As shown, the vehicle flow statistics method based on vibration signals includes: S101, obtaining a spatiotemporal vibration response diagram of a cross-sectional area; the spatiotemporal vibration response diagram includes a plurality of initial vehicle driving trajectories; S102, determining at least two candidate driving trajectories that conform to the signal attenuation law and at least one first target driving trajectory that does not conform to the signal attenuation law from among the multiple initial vehicle driving trajectories, and determining a second target driving trajectory with the largest waveform amplitude from among the at least two candidate driving trajectories; S103, extracting straight lines from the first target driving trajectory and the second target driving trajectory to obtain a plurality of extracted trajectories; S104: Taking each extracted trajectory as a vehicle object, and determining the traffic volume in the cross-sectional area based on the vehicle object.

[0025] The method of obtaining the spatiotemporal vibration response diagram of the cross-sectional area in step S101 may be: obtaining the spatiotemporal vibration response diagram of the entire area, intercepting the spatiotemporal vibration response diagram of the entire area based on the cross-sectional area, and then obtaining the spatiotemporal vibration response diagram of the cross-sectional area. Alternatively, the method may be: directly obtaining the spatiotemporal vibration response diagram of the cross-sectional area.

[0026] It should be understood that the at least two candidate driving trajectories that meet the signal attenuation law in step S102 are multiple trajectories caused by one vehicle, and the multiple trajectories include a real trajectory and at least one false trajectory generated due to vibration propagation characteristics, and the trajectory with the largest waveform amplitude is the real trajectory. Therefore, only the second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories is retained, and the other candidate driving trajectories are eliminated.

[0027] The trajectory line extraction method in step S103 may be probabilistic Hough extraction.

[0028] Under normal circumstances, vehicles usually travel at a stable constant speed or accelerate during driving, and vehicle trajectories appear as clear curves with stable and continuous characteristics. Therefore, the straight line detection method is very suitable for trajectory extraction.

[0029] By extracting straight lines from the first target driving trajectory and the second target driving trajectory, the trajectory clarity can be improved. Figure 2 As shown, Figure 2 The left side shows the driving trajectory before straight line extraction. The white straight line is the driving trajectory before straight line extraction. Figure 2 The right side shows the driving trajectory after straight line extraction. The red solid line is the extracted driving trajectory. Figure 2 It can be seen that the clarity of the extracted trajectory is improved by extracting the trajectory straight line.

[0030] Specifically, step S104 includes: determining the total number of vehicles in the cross-sectional area within the statistical time based on the vehicle objects, and this total number is the vehicle flow.

[0031] Compared with the prior art, the vehicle flow statistics method based on vibration signals provided in the embodiment of the present invention determines multiple initial vehicle driving trajectories in the spatiotemporal vibration response diagram based on the signal attenuation law to obtain at least two candidate driving trajectories that conform to the signal attenuation law. The vibration signal propagates in the medium and its energy is attenuated due to scattering and absorption, which avoids the situation where one vehicle corresponds to multiple trajectories, improves the accuracy of the determined second target form trajectory, and further improves the accuracy of the statistical vehicle flow.

[0032] In a specific embodiment of the present invention, Figure 3 As shown, step S101 includes: S301, obtaining a global vehicle vibration signal collected by a grating array vibration sensing optical cable; S302, determining a target vehicle vibration signal in the global vehicle vibration signal based on the cross-sectional area; the target vehicle vibration signal includes a plurality of target vehicle vibration sub-signal sequences corresponding to each measuring point of the grating array vibration sensing optical cable, and the target vehicle vibration sub-signal sequence includes a plurality of target vehicle vibration sub-signals corresponding to a plurality of acquisition time points; S303 , sorting the vibration signals of the target vehicle with the spatial positions of the measuring points as rows and the time as columns to obtain a spatiotemporal vibration response diagram.

[0033] Specifically, the grating array vibration sensing optical cable is composed of a large number of vibration sensing gratings, and each vibration sensing grating serves as a measuring point.

[0034] In a specific embodiment of the present invention, the signal attenuation law is:

[0035] In the formula, is the average amplitude of the first candidate driving trajectory; is the distance between the first candidate driving trajectory and the second candidate driving trajectory; The distance from the first candidate driving trajectory is The average amplitude of the second candidate driving trajectory; is the attenuation coefficient; is the natural logarithm.

[0036] Due to the propagation attenuation characteristics of the vibration signal, the initial vehicle driving trajectory in the spatiotemporal vibration response diagram has breakpoints or fluctuations, that is, the image quality of the initial vehicle driving trajectory is poor. In order to improve the accuracy of the vehicle flow statistics, it is necessary to ensure the image quality of the spatiotemporal vibration response diagram. To achieve this goal, in some embodiments of the present invention, such as Figure 4 and Figure 5 As shown, after step S101, the following further includes: S401, obtaining a fully trained image enhancement model, the image enhancement model including a generator and a discriminator, the generator including a feature extraction unit, a feature attention learning unit and an enhancement unit; S402, extracting features from the spatiotemporal vibration response graph based on a feature extraction unit to obtain initial image features; S403, performing channel attention learning and spatial attention learning on the initial image features in sequence based on the feature attention learning unit to obtain image attention features; S404: reconstruct the image attention features based on the enhancement unit to obtain an enhanced spatiotemporal vibration response map.

[0037] Among them, the discriminator is only used in the training stage of the image enhancement model. The training goal of the image enhancement model is to alternately train the generator and the discriminator so that the generator gradually learns to generate image data close to the real enhanced spatiotemporal vibration response map, so that the discriminator cannot distinguish whether the image data generated by the generator is real data or image data generated by the generator.

[0038] The feature extraction unit can be an existing mature backbone network, such as ResNet and Unet, etc. The enhancement unit is composed of multiple convolutional layers and activation function layers.

[0039] The embodiment of the present invention enhances the spatiotemporal vibration response map through an image enhancement model, thereby improving the clarity of the enhanced spatiotemporal vibration response map, and further improving the accuracy of vehicle flow statistics using the enhanced spatiotemporal vibration response map.

[0040] In addition, the embodiment of the present invention sets a feature attention learning unit to perform channel attention learning and spatial attention learning on the initial image features in sequence, which more effectively characterizes the image features from the two dimensions of channel and space, and greatly improves the clarity of the enhanced spatiotemporal vibration response map.

[0041] In a specific embodiment of the present invention, Figure 6 As shown, the feature attention learning unit includes a spatial attention learning layer and a channel attention learning layer, and step S403 is specifically as follows: Based on the channel attention learning layer, channel feature learning is performed on the initial image features to obtain channel attention features; Multiply the channel attention feature with the initial image feature to obtain the channel image feature; Based on the spatial attention learning layer, spatial feature learning is performed on the channel image features to obtain spatial attention features; Multiply the spatial attention feature with the channel image feature to obtain the image attention feature.

[0042] Among them, the channel attention learning layer and the spatial attention learning layer are existing mature architectures and will not be described in detail here.

[0043] After step S102, only the false trajectories caused by vehicle vibration are eliminated, and some noise trajectories still exist in the spatiotemporal vibration response diagram. In addition, when performing trajectory straight line extraction on the first target driving trajectory and the second target driving trajectory, abnormal noise straight lines are also introduced. In order to eliminate the above noise trajectories and further improve the accuracy of the vehicle object, in some embodiments of the present invention, the multiple extracted trajectories include a first extracted trajectory at the current moment and a second extracted trajectory at the historical moment; wherein the current moment is an adjacent sampling moment of the historical moment.

[0044] Since the vehicle driving process is relatively stable, that is, the trajectory characteristics of the same vehicle object at different times are not much different, while the trajectory characteristics of the noise trajectory are quite different from the vehicle trajectory. Based on this characteristic, in some embodiments of the present invention, such as Figure 7 As shown, before step S104, the vehicle flow statistics method based on vibration signals further includes: S701, obtaining a first trajectory feature of a first extracted trajectory and a second trajectory feature of a second extracted trajectory; S702, determining feature loss values ​​of the first extracted trajectory and the second extracted trajectory based on the first trajectory feature and the second trajectory feature; S703: When the feature loss value is less than the loss threshold, merge the first extracted trajectory and the second extracted trajectory to obtain a merged trajectory.

[0045] Then, in step S104, each extracted trajectory is taken as a vehicle object, specifically: Merge the track as a vehicle object.

[0046] It should be understood that: when the feature loss value is greater than or equal to the loss threshold, the first extracted trajectory and the second extracted trajectory do not belong to the driving trajectory corresponding to the same vehicle.

[0047] The embodiment of the present invention utilizes the driving stability of the vehicle to track and match the extracted trajectories at different times, thereby avoiding misjudging the features of the same vehicle at multiple different times as two vehicle objects, improving the accuracy of the determined vehicle objects, and further improving the accuracy of the traffic flow statistics.

[0048] Moreover, by matching trajectory features, noise trajectories can be eliminated, further ensuring the accuracy of the determined vehicle objects.

[0049] It should be understood that the loss threshold can be set or adjusted according to the actual application scenario and is not specifically limited here.

[0050] To ensure the accuracy of the feature loss value, in a specific embodiment of the present invention, the first trajectory feature and the second trajectory feature both include speed, energy, lane and position; Figure 8 As shown, step S702 includes: S801, determining a speed loss value, an energy loss value, a lane loss value, and a position loss value of a first extracted trajectory and a second extracted trajectory; S802: weighting the speed loss value, the energy loss value, the lane loss value, and the position loss value based on preset weights to obtain a feature loss value.

[0051] The embodiment of the present invention determines the feature loss value through four physical features of speed, energy, lane and position, thereby avoiding the problem of inaccurate trajectory tracking matching caused by single-dimensional physical features, ensuring the accuracy and reliability of trajectory tracking matching, and improving the accuracy of the determined vehicle object.

[0052] Among them, the expression of the feature loss value is as follows:

[0053] In the formula, is the feature loss value; The trajectory features of the first extracted trajectory at the current moment; is the trajectory feature of the second extracted trajectory of the historical moment; Feature weights.

[0054] In summary, the vehicle flow statistics method based on vibration signals proposed in the embodiment of the present invention realizes long-distance, all-weather, and real-time vibration signal collection based on grating array vibration sensor optical cables, thereby realizing long-distance, all-weather, and real-time vehicle flow statistics. In addition, by setting signal attenuation rules, image enhancement, and trajectory tracking matching, the accuracy of identified vehicle objects is improved, thereby realizing accurate vehicle flow statistics, providing strong support for subsequent vehicle path planning, reasonable vehicle diversion, and other events.

[0055] In order to better implement the vehicle flow statistics method based on vibration signals in the embodiment of the present invention, on the basis of the vehicle flow statistics method based on vibration signals, the embodiment of the present invention also provides a vehicle flow statistics device based on vibration signals, such as Fig. 9 As shown, the vehicle flow counting device 900 based on vibration signals includes: The time-space vibration response map acquisition unit 901 is used to acquire the time-space vibration response map of the cross-section area; the time-space vibration response map includes multiple initial vehicle driving trajectories; The driving trajectory judgment unit 902 is used to determine at least two candidate driving trajectories that meet the signal attenuation law and at least one first target driving trajectory that does not meet the signal attenuation law among the multiple initial vehicle driving trajectories; and determine a second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories; A trajectory straight line extraction unit 903, used to extract trajectory straight lines from the first target driving trajectory and the second target driving trajectory to obtain multiple extracted trajectories; The vehicle flow determination unit 904 is used to take each extracted trajectory as a vehicle object and determine the vehicle flow of the cross-sectional area based on the vehicle object.

[0056] The vehicle flow statistics device 900 based on vibration signals provided in the above embodiment can implement the technical solution described in the above embodiment of the vehicle flow statistics method based on vibration signals. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above embodiment of the vehicle flow statistics method based on vibration signals, which will not be repeated here.

[0057] like Fig.10 As shown, the present invention also provides an electronic device 1000. The electronic device 1000 includes a processor 1001, a memory 1002 and a display 1003. Fig.10 Only some components of the electronic device 1000 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0058] In some embodiments, the processor 1001 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 1002, such as the vehicle flow statistics method based on vibration signals in the present invention.

[0059] In some embodiments of the present invention, processor 1001 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, processor 1001 may be local or remote. In some embodiments, processor 1001 may be implemented in a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0060] In some embodiments, the memory 1002 may be an internal storage unit of the electronic device 1000, such as a hard disk or memory of the electronic device 1000. In other embodiments, the memory 1002 may also be an external storage device of the electronic device 1000, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 1000.

[0061] Furthermore, the memory 1002 may include both an internal storage unit of the electronic device 1000 and an external storage device. The memory 1002 is used to store application software installed in the electronic device 1000 and various data.

[0062] In some embodiments, the display 1003 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 1003 is used to display information of the electronic device 1000 and to display a visual user interface. The components 1001-1003 of the electronic device 1000 communicate with each other via a system bus.

[0063] In some embodiments of the present invention, when the processor 1001 executes the vehicle flow statistics program based on the vibration signal in the memory 1002, the following steps may be implemented: Obtaining a spatiotemporal vibration response diagram of a cross-sectional area; the spatiotemporal vibration response diagram includes a plurality of initial vehicle driving trajectories; Determine at least two candidate driving trajectories that conform to the signal attenuation law and at least one first target driving trajectory that does not conform to the signal attenuation law among the multiple initial vehicle driving trajectories, and determine a second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories; Extracting straight lines from the first target driving trajectory and the second target driving trajectory to obtain a plurality of extracted trajectories; Each extracted trajectory is regarded as a vehicle object, and the traffic volume in the cross-sectional area is determined based on the vehicle object.

[0064] It should be understood that: when the processor 1001 executes the vehicle flow statistics program based on vibration signals in the memory 1002, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the corresponding method embodiment above.

[0065] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 1000 mentioned, and the electronic device 1000 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices. It should also be understood that in some other embodiments of the present invention, the electronic device 1000 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (such as a touch panel).

[0066] Correspondingly, an embodiment of the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps or functions of the vehicle flow statistics method based on vibration signals provided in the above-mentioned method embodiments can be implemented.

[0067] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0068] The above is a detailed introduction to the vehicle flow statistics method, device, equipment and storage medium based on vibration signals provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A vehicle flow statistics method based on vibration signals, characterized in that: include: Obtaining a spatiotemporal vibration response diagram of a cross-sectional area; the spatiotemporal vibration response diagram comprising a plurality of initial vehicle travel trajectories; Determine at least two candidate driving trajectories that conform to the signal attenuation law and at least one first target driving trajectory that does not conform to the signal attenuation law among the multiple initial vehicle driving trajectories, and determine a second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories; Extracting straight lines from the first target driving trajectory and the second target driving trajectory to obtain a plurality of extracted trajectories; Each of the extracted trajectories is regarded as a vehicle object, and the vehicle flow rate of the cross-sectional area is determined based on the vehicle object.

2. The vehicle flow statistics method based on vibration signals according to claim 1 is characterized in that: The step of obtaining a spatiotemporal vibration response diagram of a cross-sectional area includes: Acquire global vehicle vibration signals collected by grating array vibration sensing optical cables; Determine a target vehicle vibration signal in the global vehicle vibration signal based on the cross-sectional area; the target vehicle vibration signal includes a plurality of target vehicle vibration sub-signal sequences corresponding to each measuring point of the grating array vibration sensing optical cable, and the target vehicle vibration sub-signal sequence includes a plurality of target vehicle vibration sub-signals corresponding to a plurality of acquisition time points; The target vehicle vibration signals are sorted with the spatial positions of the measuring points as rows and the time as columns to obtain the spatiotemporal vibration response diagram.

3. The vehicle flow statistics method based on vibration signals according to claim 1, characterized in that: The signal attenuation law is: In the formula, is the average amplitude of the first candidate driving trajectory; is the distance between the first candidate driving trajectory and the second candidate driving trajectory; The distance from the first candidate driving trajectory is The average amplitude of the second candidate driving trajectory; is the attenuation coefficient; is the natural logarithm.

4. The vehicle flow statistics method based on vibration signals according to claim 1, characterized in that: After obtaining the spatiotemporal vibration response diagram of the cross-sectional area, the method further includes: Acquire a fully trained image enhancement model, wherein the image enhancement model includes a generator and a discriminator, and the generator includes a feature extraction unit, a feature attention learning unit, and an enhancement unit; Performing feature extraction on the spatiotemporal vibration response graph based on the feature extraction unit to obtain initial image features; Based on the feature attention learning unit, channel attention learning and spatial attention learning are performed on the initial image features in sequence to obtain image attention features; Based on the enhancement unit, feature reconstruction is performed on the image attention feature to obtain an enhanced spatiotemporal vibration response map.

5. The vehicle flow statistics method based on vibration signals according to claim 4 is characterized in that: The feature attention learning unit includes a spatial attention learning layer and a channel attention learning layer; the step of sequentially performing channel attention learning and spatial attention learning on the initial image features based on the feature attention learning unit to obtain image attention features includes: Performing channel feature learning on the initial image features based on the channel attention learning layer to obtain channel attention features; Multiplying the channel attention feature with the initial image feature to obtain a channel image feature; Performing spatial feature learning on the channel image features based on the spatial attention learning layer to obtain spatial attention features; Multiply the spatial attention feature by the channel image feature to obtain the image attention feature.

6. The vehicle flow statistics method based on vibration signals according to claim 1, characterized in that: The plurality of extracted trajectories include a first extracted trajectory at a current moment and a second extracted trajectory at a historical moment; before taking each of the extracted trajectories as a vehicle object, the method further includes: Acquire a first trajectory feature of the first extracted trajectory and a second trajectory feature of the second extracted trajectory; Determining feature loss values ​​of the first extracted trajectory and the second extracted trajectory based on the first trajectory feature and the second trajectory feature; When the feature loss value is less than a loss threshold, merging the first extracted trajectory and the second extracted trajectory to obtain a merged trajectory; Then, each of the extracted trajectories is taken as a vehicle object, specifically: The merged trajectory is used as the vehicle object.

7. The vehicle flow statistics method based on vibration signals according to claim 6 is characterized in that: The first trajectory feature and the second trajectory feature both include speed, energy, lane, and position; and determining the feature loss values ​​of the first extracted trajectory and the second extracted trajectory based on the first trajectory feature and the second trajectory feature includes: Determining a speed loss value, an energy loss value, a lane loss value, and a position loss value for the first extracted trajectory and the second extracted trajectory; The speed loss value, the energy loss value, the lane loss value and the position loss value are weighted based on preset weights to obtain the characteristic loss value.

8. A vehicle flow statistics device based on vibration signals, characterized in that: include: A time-space vibration response map acquisition unit, used to acquire a time-space vibration response map of a cross-sectional area; the time-space vibration response map includes a plurality of initial vehicle driving trajectories; A driving trajectory judgment unit is used to determine at least two candidate driving trajectories that conform to the signal attenuation law and at least one first target driving trajectory that does not conform to the signal attenuation law among the multiple initial vehicle driving trajectories; and determine a second target driving trajectory with the largest waveform amplitude among the at least two candidate driving trajectories; A trajectory straight line extraction unit, used for performing trajectory straight line extraction on the first target driving trajectory and the second target driving trajectory to obtain a plurality of extracted trajectories; The vehicle flow determination unit is used to take each of the extracted trajectories as a vehicle object and determine the vehicle flow of the cross-sectional area based on the vehicle object.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the vehicle flow statistics method based on vibration signals as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the vehicle flow statistics method based on vibration signals as described in any one of claims 1 to 7.

Images