Model training method, false terminal identification method and device, electronic device

Abstract

The embodiment of the present invention discloses a model training method, a false terminal identification method and device, and an electronic device. The model training method includes: obtaining first data of multiple vehicle-mounted terminals, and determining the labeled data corresponding to the first data; the first data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle; the labeled data is determined by the first data, and / or the labeled data is pre-set data; the initial classification model is trained based on the multiple first data and the labeled data corresponding to the first data to obtain a terminal identification model; the labeled data represents the type of the first data as a real data type or a false data type.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle networking, and in particular to a model training method, a false terminal identification method and device, and an electronic device. Background Art

[0002] With the rapid development of smart cars, the demand for connected technologies is also gradually increasing. Currently, there are problems with fake vehicle terminal access (V2X, Vehicle to everything) platforms defrauding V2X data such as traffic light information.

[0003] Regarding the identification of fake terminals, traditional authentication and other security mechanisms can be used to identify whether the terminal is a real terminal or a fake terminal. However, there is still a situation where the real terminal reports false messages to obtain data from the V2X platform. Because the analysis of the V2X platform relies on the data reported by the terminal, it is still necessary to distinguish and eliminate this part of the data. If the terminal continues to report false data, the platform needs to stop the service of such terminals. Based on this, there is currently no effective solution for how to identify the real terminal that reports false messages. Summary of the invention

[0004] In order to solve the existing technical problems, the embodiments of the present invention provide a model training method, a false terminal identification method and device, and an electronic device.

[0005] To achieve the above object, the technical solution of the embodiment of the present invention is implemented as follows:

[0006] An embodiment of the present invention provides a model training method, the method comprising:

[0007] Obtaining first data sent by a plurality of vehicle-mounted terminals, and determining annotated data corresponding to the first data; the first data is data representing a driving state of a vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle; the annotated data is determined by the first data, and / or the annotated data is pre-set data; the annotated data represents that the type of the first data is a real data type or a false data type;

[0008] An initial classification model is trained based on the plurality of first data and the labeled data corresponding to the first data to obtain a terminal recognition model.

[0009] In the above solution, the first data at least includes the location information of the vehicle; determining the labeled data by using the first data includes:

[0010] Determine trajectory data based on the obtained multiple location information of the vehicle;

[0011] Determine the map data corresponding to the location information of the vehicle, and obtain the road data corresponding to the location information of the vehicle from the map data;

[0012] The similarity between the trajectory data and the road data is determined, and based on the similarity, the first data is determined to be a real data type or a false data type.

[0013] In the above solution, the first data includes at least one of the following: location information of the vehicle and speed information of the vehicle; determining the labeled data through the first data includes:

[0014] collecting speed information of the vehicle, determining a first distribution feature of the vehicle within a first preset time range based on the speed information of the vehicle, and determining whether the first data is a real data type or a false data type based on the first distribution feature; and / or,

[0015] Determine the distance moved by the vehicle based on the location information of the vehicle, determine the acceleration information of the vehicle based on the distance moved by the vehicle and the speed information of the vehicle, determine a second distribution feature of the vehicle within a second preset time range based on the acceleration information of the vehicle, and determine whether the first data is a true data type or a false data type based on the second distribution feature.

[0016] In the above solution, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle;

[0017] The training of the initial classification model based on the plurality of first data and the labeled data corresponding to the first data includes:

[0018] The plurality of first data are normalized, and an initial classification model is trained based on the plurality of first data after the normalization and the corresponding labeled data.

[0019] The embodiment of the present invention further provides a method for identifying a fake terminal, the method comprising:

[0020] Obtaining second data sent by the vehicle-mounted terminal, where the second data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle;

[0021] The type of the second data is determined based on a terminal recognition model, and whether the vehicle-mounted terminal is a false terminal is determined based on the type of the second data; wherein the terminal recognition model is obtained using the model training method described in an embodiment of the present invention.

[0022] In the above scheme, the method further includes: in the case where the vehicle-mounted terminal is determined to be a false terminal, performing corresponding processing according to the duration of time that the vehicle-mounted terminal is determined to be a false terminal.

[0023] In the above solution, the corresponding processing is performed according to the duration of time during which the vehicle-mounted terminal is determined to be a false terminal, including:

[0024] When the duration for which the vehicle-mounted terminal is determined to be a false terminal reaches a first preset threshold, outputting a warning prompt message;

[0025] When the duration for which the vehicle-mounted terminal is determined to be a false terminal reaches a second preset threshold, stopping pushing data to the vehicle-mounted terminal;

[0026] When the duration for which the vehicle-mounted terminal is determined to be a false terminal reaches a third preset threshold, recording the identification of the vehicle-mounted terminal in a blacklist;

[0027] The first preset threshold is smaller than the second preset threshold, and the second preset threshold is smaller than the third preset threshold.

[0028] The embodiment of the present invention further provides a model training device, which includes: a first acquisition unit, a determination unit and a training unit; wherein:

[0029] The first acquisition unit is used to obtain first data of multiple vehicle terminals; the first data is data representing the driving state of the vehicle obtained by the vehicle terminal during the driving of the vehicle;

[0030] The determining unit is used to determine the annotated data corresponding to the first data; the annotated data is determined by the first data, and / or the annotated data is pre-set data; the annotated data represents the type of the first data as a real data type or a false data type;

[0031] The training unit is used to train the initial classification model based on the multiple first data and the labeled data corresponding to the first data to obtain the terminal recognition model.

[0032] In the above solution, the first data at least includes location information of the vehicle;

[0033] The determination unit is used to determine trajectory data based on multiple location information of the vehicle; determine map data corresponding to the location information of the vehicle, and obtain road data corresponding to the location information of the vehicle from the map data; determine the similarity between the trajectory data and the road data, and determine whether the first data is a real data type or a false data type based on the similarity.

[0034] In the above solution, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle;

[0035] The determining unit is used to collect speed information of the vehicle, determine a first distribution feature of the vehicle within a first preset time range based on the speed information of the vehicle, and determine whether the first data is a real data type or a false data type based on the first distribution feature; and / or,

[0036] Determine the distance moved by the vehicle based on the location information of the vehicle, determine the acceleration information of the vehicle based on the distance moved by the vehicle and the speed information of the vehicle, determine a second distribution feature of the vehicle within a second preset time range based on the acceleration information of the vehicle, and determine whether the first data is a true data type or a false data type based on the second distribution feature.

[0037] In the above solution, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle;

[0038] The device further comprises a preprocessing unit, configured to perform normalization processing on the plurality of the first data;

[0039] The training unit is used to train the initial classification model based on the normalized plurality of first data and the corresponding labeled data.

[0040] The embodiment of the present invention further provides a false terminal identification device, the device comprising a second acquisition unit and a classification unit; wherein,

[0041] The second acquisition unit is used to obtain second data sent by the vehicle-mounted terminal, where the second data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle;

[0042] The classification unit is used to determine the type of the second data based on a terminal recognition model, and determine whether the vehicle-mounted terminal is a false terminal based on the type of the second data; wherein the terminal recognition model is obtained using the model training method described in an embodiment of the present invention.

[0043] In the above solution, the device further includes a processing unit for, when the vehicle-mounted terminal is determined to be a false terminal, performing corresponding processing according to the duration for which the vehicle-mounted terminal is determined to be a false terminal.

[0044] In the above scheme, when the duration of the vehicle-mounted terminal being determined as a false terminal reaches a first preset threshold, a warning prompt message is output; when the duration of the vehicle-mounted terminal being determined as a false terminal reaches a second preset threshold, data push to the vehicle-mounted terminal is stopped; when the duration of the vehicle-mounted terminal being determined as a false terminal reaches a third preset threshold, the identification of the vehicle-mounted terminal is recorded in a blacklist; wherein, the first preset threshold is less than the second preset threshold, and the second preset threshold is less than the third preset threshold.

[0045] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the model training method described in the embodiment of the present invention; or, when executed by a processor, implements the steps of the false terminal identification method described in the embodiment of the present invention.

[0046] An embodiment of the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the model training method described in the embodiment of the present invention are implemented; or, when the processor executes the program, the steps of the model training method described in the embodiment of the present invention are implemented.

[0047] The model training method, false terminal identification method and device, and electronic device provided by the embodiment of the present invention include: obtaining first data of multiple vehicle terminals, determining the labeled data corresponding to the first data; the first data is the data characterizing the driving state of the vehicle obtained by the vehicle terminal during the driving of the vehicle; the labeled data is determined by the first data, and / or the labeled data is pre-set data; the labeled data characterizes the type of the first data as a real data type or a false data type; based on the multiple first data and the labeled data corresponding to the first data, the initial classification model is trained to obtain a terminal identification model. The technical solution of the embodiment of the present invention is adopted to establish a terminal identification model through the first data reported by the vehicle terminal, that is, to establish a model according to the law of the first data reported by the vehicle terminal, so as to identify the authenticity of the data reported by the vehicle terminal based on the terminal identification model, avoid the behavior of the real terminal reporting false data to defraud the V2X platform data, and ensure the data security and validity of the V2X platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a flow chart of a model training method according to an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of determining trajectory data in a model training method according to an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of a flow chart of a false terminal identification method according to an embodiment of the present invention;

[0051] Figure 4 A schematic diagram of the structure of a model training device according to an embodiment of the present invention;

[0052] Figure 5 A schematic diagram of another structure of the model training device according to an embodiment of the present invention;

[0053] Figure 6 A schematic diagram of a structure of a false terminal identification device according to an embodiment of the present invention;

[0054] Figure 7 It is another schematic diagram of the structure of the false terminal identification device according to an embodiment of the present invention;

[0055] Figure 8 The figure is a schematic diagram of the hardware structure of the electronic device according to the embodiment of the present invention. DETAILED DESCRIPTION

[0056] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] An embodiment of the present invention provides a model training method. Figure 1 Schematic diagram of the model training method according to an embodiment of the present invention; Figure 1 As shown, the method includes:

[0058] Step 101: obtaining first data of a plurality of vehicle-mounted terminals, and determining annotated data corresponding to the first data; the first data is data representing a driving state of a vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle; the annotated data is determined by the first data, and / or the annotated data is pre-set data; the annotated data represents that the type of the first data is a real data type or a false data type;

[0059] Step 102: Train an initial classification model based on the first data and the labeled data corresponding to the first data to obtain a terminal recognition model.

[0060] The model training method of this embodiment is applied to a model training device; the model training device can be located in any electronic device, such as a personal computer (PC).

[0061] In this embodiment, the vehicle-mounted terminal is located in the vehicle, and the vehicle-mounted terminal moves with the movement of the vehicle. The vehicle-mounted terminal can obtain first data representing the state of the vehicle; the first data may include at least one of the following: location information of the vehicle, speed information of the vehicle, etc. In other embodiments, the first data may also include a terminal identification (ID). In practical applications, a positioning module may be provided in the vehicle, and the location information of the vehicle may be obtained through the positioning module. As an example, the positioning module may be implemented by a global positioning system (GPS, Global Positioning System) module. In addition, a speed sensor may also be provided in the vehicle, and its own speed information may be obtained through the speed sensor; wherein the speed information may include the magnitude and direction of the speed. Then the vehicle-mounted terminal may establish a communication connection with the positioning module, and obtain the location information of the vehicle collected by the positioning module based on the communication connection; and / or, the vehicle-mounted terminal may establish a communication connection with the speed sensor, and obtain the speed information of the vehicle collected by the speed sensor based on the communication connection.

[0062] In this embodiment, the model training device can obtain the first data of multiple vehicle terminals. In addition, the model training device can obtain the first data of multiple vehicle terminals at a preset frequency. For example, for a vehicle terminal, N first data are obtained within a T time period.

[0063] As an implementation mode, the model training device may establish a wireless connection with the vehicle-mounted terminal, and obtain the first data collected by the vehicle-mounted terminal based on the wireless connection. The vehicle-mounted terminal may send the first data to the model training device at a preset frequency, so that the model training device obtains the first data sent by the vehicle-mounted device at the preset frequency. As an example, the preset frequency may be 1 Hz, 5 Hz, 10 Hz, etc., and of course may be other values, which are not limited in this embodiment.

[0064] As another implementation, the vehicle-mounted terminal may also store the collected first data, and the model training device may obtain the first data stored in the vehicle-mounted terminal through a wired connection or an external mobile storage device. The vehicle-mounted terminal may send the first data to the model training device at a preset frequency, so that the model training device obtains the first data of the vehicle-mounted device that meets the preset frequency.

[0065] In this embodiment, the labeled data corresponding to the first data characterizes that the type of the first data is a real data type or a false data type; the method for determining the labeled data may include the following methods: determining through the first data, and / or, determining through a pre-set method. It can be understood that in this embodiment, as an implementation method, the labeled data corresponding to the first data can be determined by manual labeling. For example, it is set that the first data reported by a certain vehicle-mounted terminal are all real data types, then it is determined that the labeled data corresponding to the first data of the vehicle-mounted terminal are all real data types. As another implementation method, the regularity of the first data can be analyzed and processed, and the labeled data corresponding to the first data can be determined according to the analysis and processing results.

[0066] In an optional embodiment of the present invention, the first data includes at least location information of a vehicle; determining the annotation data through the first data includes: determining trajectory data based on multiple location information of the vehicle; determining map data corresponding to the location information of the vehicle, and obtaining road data corresponding to the location information of the vehicle from the map data; determining the similarity between the trajectory data and the road data, and determining whether the first data is a real data type or a false data type based on the similarity.

[0067] In this embodiment, determining the trajectory data based on the obtained multiple position information of the vehicle includes: obtaining multiple position information within a period of time, and determining the trajectory data based on the multiple position information.

[0068] Specifically, when the location information is obtained through a GPS module in the vehicle, the location information can be represented by longitude and latitude coordinates. The longitude and latitude coordinates are first converted to a rectangular coordinate system to obtain rectangular coordinates representing the vehicle's position, and a scatter plot corresponding to multiple location information within the time range (for example, five minutes) is obtained. A fitting curve is then obtained by fitting, and the fitting curve is the trajectory data of the vehicle within the time range. In actual applications, the conversion between geographic coordinates and rectangular coordinates can be performed by calling the projfwd function. A scatter plot representing multiple location information can be obtained by calling the scatter function. Curve fitting can be performed by calling the plotfit function. Refer to Figure 2 As shown, the GPS data of a period of time is converted into rectangular coordinates as shown in Figure 2 As shown in the data points, some of these data points are erroneous data points, and most of the data points are correct data points; a fitting curve is obtained by curve fitting and the fitting curve is the trajectory data.

[0069] Furthermore, the vehicle's location information (such as the starting location information and the ending location information within the time range) is used to determine the road data where the vehicle is located by calling the map data, and the road data is directly obtained from the map data. The similarity between the vehicle's trajectory data and the road data is compared.

[0070] As an implementation manner, determining the similarity between the trajectory data and the road data includes: determining a Euclidean distance between the trajectory data and the road data, and determining the similarity between the trajectory data and the road data based on the Euclidean distance.

[0071] In this embodiment, the similarity between the trajectory data and the road data is determined based on calculating the Euclidean distance. The determined Euclidean distance is shown in the following expression (1):

[0072]

[0073] Wherein, y1 and y2 represent the ordinates of the points with the same abscissa in the trajectory data and the road data, respectively; n represents the number of data points in the trajectory data and the road data. Of course, in other embodiments, the similarity between the trajectory data and the road data can also be determined based on the abscissas of the points with the same ordinates. In this embodiment, the Euclidean distance can also be determined by using expression (1). The difference is that y1 and y2 represent the abscissas of the points with the same ordinates in the trajectory data and the road data, respectively.

[0074] In practical applications, the ordinate in the corresponding road data that satisfies the abscissa can be determined based on the abscissa of each data point in the trajectory data, or the abscissa in the corresponding road data that satisfies the ordinate can be determined based on the ordinate of each data point in the trajectory data, and then the Euclidean distance between the trajectory data and the road data can be determined according to the above expression (1).

[0075] In this embodiment, determining whether the first data is a real data type or a false data type based on the similarity includes: when the similarity is greater than a first threshold, determining that the first data is a real data type; when the similarity is less than a second threshold, determining that the first data is a false data type; wherein the second threshold is less than or equal to the first threshold.

[0076] As another embodiment, determining the similarity between the trajectory data and the road data includes: determining a first coefficient based on the vertical coordinates in the trajectory data and the road data that satisfy the same horizontal coordinate, or the horizontal coordinates that satisfy the same vertical coordinate; the first coefficient characterizes the correlation between the trajectory data and the road data; and determining whether the first data is a real data type or a false data type based on the first coefficient.

[0077] Among them, the first coefficient can be a Pearson correlation coefficient (Pearson Correlation Coefficient); the larger the absolute value of the Pearson correlation coefficient, the stronger the correlation between the trajectory data and the road data. Conversely, the smaller the absolute value of the Pearson correlation coefficient, the weaker the correlation between the trajectory data and the road data.

[0078] In this embodiment, the similarity between the trajectory data and the road data is determined by calculating the Pearson correlation coefficient. The determined Pearson correlation coefficient is shown in the following expression (2):

[0079]

[0080] Where Y1 and Y2 represent the ordinates of the points with the same abscissa in the trajectory data and the road data, respectively; cov(Y1, Y2) represents the covariance of Y1 and Y2; σ Y1 Represents the standard deviation of the ordinate of the points in the trajectory data; σ Y2 represents the standard deviation of the ordinate of the point in the road data that has the same abscissa as the point in the trajectory data; E represents the mathematical expectation; u Y1 represents the true value of Y1; u Y2 Of course, in other embodiments, the similarity between the trajectory data and the road data can also be determined based on the horizontal coordinates of the points with the same vertical coordinates. In this embodiment, the Pearson correlation coefficient can also be determined by using expression (2). The difference is that Y1 and Y2 represent the horizontal coordinates of the points with the same vertical coordinates in the trajectory data and the road data, respectively.

[0081] In actual applications, the ordinate in the corresponding road data that satisfies the abscissa can be determined based on the abscissa of each data point in the trajectory data, or the abscissa in the corresponding road data that satisfies the ordinate can be determined based on the ordinate of each data point in the trajectory data, and then the Pearson correlation coefficient of the trajectory data and the road data can be determined according to the above expression (2).

[0082] In this embodiment, determining whether the first data is a real data type or a false data type based on the similarity includes: when the absolute value of the first coefficient is greater than a third threshold, determining that the first data is a real data type; when the absolute value of the first coefficient is less than a fourth threshold, determining that the first data is a false data type; wherein the fourth threshold is less than or equal to the third threshold.

[0083] In an optional embodiment of the present invention, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle; determining the labeled data through the first data includes: counting the speed information of the vehicle, determining a first distribution feature of the vehicle within a first preset time range based on the speed information of the vehicle, and determining whether the first data is a real data type or a false data type based on the first distribution feature; and / or determining a distance moved by the vehicle based on the location information of the vehicle, determining the acceleration information of the vehicle based on the distance moved by the vehicle and the speed information of the vehicle, determining a second distribution feature of the vehicle within a second preset time range based on the acceleration information of the vehicle, and determining whether the first data is a real data type or a false data type based on the second distribution feature.

[0084] In this embodiment, by counting the first data and determining the distribution characteristics of the first data, it is determined whether the first data is a real data type or a false data type.

[0085] As an implementation mode, if the speed information corresponding to the vehicle terminal remains unchanged within a period of time, or the speed information corresponding to the vehicle terminal changes but the location information of the vehicle remains unchanged, the first data can be determined to be a false data type. For example, the speed information corresponding to the same vehicle terminal identification for a period of time is always a certain value. Considering that the vehicle speed will not remain at a fixed speed for a period of time during driving, the first data can be determined to be a false data type.

[0086] As another implementation, if the change in the speed information and / or acceleration information corresponding to the vehicle-mounted terminal within a period of time exceeds the fifth threshold or is less than the sixth threshold, the first data can be determined to be a false data type; the sixth threshold is less than the fifth threshold. For example, if the change in the speed and / or acceleration corresponding to the same vehicle-mounted terminal identification within a period of time is too large or too small, such as the speed change and / or acceleration change in a short period of time exceeds 50%, the first data can be determined to be a false data type.

[0087] Based on the determination of the false data type and the real data type, combined with the method of presetting the labeled data, a data set for model training is generated, wherein the data set includes a plurality of first data and the labeled data corresponding to each first data. The initial classification model is trained based on the data in the data set to obtain a terminal recognition model.

[0088] In practical applications, the initial classification model is input into the data set, and the initial classification model can specifically be a machine learning network model or a deep learning network model. The parameters in the initial classification model are adjusted in a supervised learning manner, and the terminal recognition model is obtained when the convergence conditions are met.

[0089] In an optional embodiment of the present invention, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle; the training of the initial classification model based on the multiple first data and the labeled data corresponding to the first data includes: normalizing the multiple first data, and training the initial classification model based on the multiple first data after normalization and the corresponding labeled data.

[0090] In this embodiment, before training based on the first data and the corresponding labeled data, since the measurement units of the values ​​are different, in order to eliminate the dimension effect, the first data can be normalized, specifically, each type of first data is normalized.

[0091] Specifically, the normalization process for a type of first data may be performed as shown in the following expression (3):

[0092]

[0093] Wherein, fi' represents the normalization processing result of a type of first data; max|f| represents the maximum value in a type of first data; min|f| represents the minimum value in a type of first data; and fi represents the i-th first data in a type of first data. Wherein, the type of first data may represent the speed information of multiple vehicles and the acceleration information of multiple vehicles; further, the type of first data may also include the proportion of speed values ​​in the speed information of vehicles lower than a first preset threshold (for example, lower than 10 kilometers per hour), the proportion of speed values ​​higher than a second preset threshold (for example, higher than 120 kilometers per hour), the proportion of acceleration values ​​higher than a third preset threshold (for example, higher than 5m / s 2 ) etc. The first type of data may also include the similarity between the trajectory data and the road data.

[0094] In an optional embodiment of the present invention, the method may also include: obtaining new first data of the vehicle-mounted terminal and labeled data corresponding to the first data, and training the terminal recognition model based on the new first data and the corresponding labeled data to update the terminal recognition model.

[0095] The terminal recognition model of this embodiment can support expansion and automatic updating. When new data for model training is generated, the terminal recognition model can be automatically updated.

[0096] By adopting the technical solution of the embodiment of the present invention, a terminal recognition model is established through the first data reported by the vehicle-mounted terminal, that is, a model is established according to the regularity of the first data reported by the vehicle-mounted terminal, so as to identify the authenticity of the data reported by the vehicle-mounted terminal based on the terminal recognition model, avoid the behavior of the real terminal reporting false data to defraud the V2X platform data, and ensure the data security and validity of the V2X platform.

[0097] The embodiment of the present invention also provides a false terminal identification method. Figure 3 FIG. 1 is a flow chart of a method for identifying a fake terminal according to an embodiment of the present invention; Figure 3 As shown, the method includes:

[0098] Step 201: obtaining second data sent by the vehicle-mounted terminal, where the second data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle;

[0099] Step 202: Determine the type of the second data based on a terminal recognition model, and determine whether the vehicle-mounted terminal is a false terminal based on the type of the second data; wherein the terminal recognition model is obtained using the model training method described in the aforementioned embodiment of the present invention.

[0100] The false terminal identification method of this embodiment can be applied to a false terminal identification device; the false terminal identification device can be located in a V2X platform. The terminal identification model obtained by the aforementioned model training method can be integrated into the false terminal identification device or the V2X platform, so that after the V2X platform obtains the second data sent by the vehicle-mounted terminal, it is determined based on the terminal identification model whether the second data is a real data type or a false data type; if the second data is determined to be a false data type, the corresponding vehicle-mounted terminal can be determined to be a false terminal.

[0101] In an optional embodiment of the present invention, the method further includes: in the case where the vehicle-mounted terminal is determined to be a false terminal, performing corresponding processing according to the duration of time during which the vehicle-mounted terminal is determined to be a false terminal.

[0102] In this embodiment, in order to avoid misjudgment of false terminals and improve recognition accuracy, in this embodiment, corresponding processing can be performed based on the duration of the vehicle-mounted terminal being determined as a false terminal (which can also be understood as the number of times the vehicle-mounted terminal is determined as a false terminal).

[0103] For example, for the same vehicle terminal ID, three second data of different time ranges are randomly selected and input into the terminal identification model for determination. If the number of times the vehicle terminal is determined to be a false terminal exceeds a preset threshold, for example, the preset threshold is 60%, if the vehicle terminal is determined to be a false terminal based on the second data of two different time ranges input into the terminal identification model, it can be considered that the vehicle terminal is indeed a false terminal, and the second data reported by the vehicle terminal is false data. The corresponding processing may be to delete the data corresponding to the vehicle terminal in the V2X platform.

[0104] In an optional embodiment of the present invention, corresponding processing is performed according to the duration of the vehicle-mounted terminal being determined as a false terminal, including: when the duration of the vehicle-mounted terminal being determined as a false terminal reaches a first preset threshold, outputting a warning prompt message; when the duration of the vehicle-mounted terminal being determined as a false terminal reaches a second preset threshold, terminating the push of data to the vehicle-mounted terminal; when the duration of the vehicle-mounted terminal being determined as a false terminal reaches a third preset threshold, recording the identification of the vehicle-mounted terminal in a blacklist; wherein the first preset threshold is less than the second preset threshold, and the second preset threshold is less than the third preset threshold.

[0105] For example, if the vehicle-mounted terminal is judged as a false terminal for three consecutive days, the V2X platform will push information to prompt an alert; if it is judged as a false terminal for five consecutive days, the V2X platform can terminate the service, that is, stop pushing data; if it is judged as a false terminal for one consecutive week, the V2X platform can write the vehicle-mounted terminal ID into the blacklist.

[0106] The embodiment of the present invention also provides a model training device. Figure 4 Schematic diagram of a structure of a model training device according to an embodiment of the present invention; Figure 4 As shown, the device includes: a first acquisition unit 31, a determination unit 32 and a training unit 33; wherein,

[0107] The first acquisition unit 31 is used to obtain first data of multiple vehicle terminals; the first data is data representing the driving state of the vehicle obtained by the vehicle terminal during the driving of the vehicle;

[0108] The determining unit 32 is used to determine the annotated data corresponding to the first data; the annotated data is determined by the first data, and / or the annotated data is pre-set data; the annotated data indicates that the type of the first data is a real data type or a false data type;

[0109] The training unit 33 is used to train the initial classification model based on the first data and the labeled data corresponding to the first data to obtain a terminal recognition model.

[0110] In an optional embodiment of the present invention, the first data at least includes location information of the vehicle;

[0111] The determination unit 32 is used to determine trajectory data based on multiple location information of the vehicle; determine map data corresponding to the location information of the vehicle, and obtain road data corresponding to the location information of the vehicle from the map data; determine the similarity between the trajectory data and the road data, and determine whether the first data is a real data type or a false data type based on the similarity.

[0112] In an optional embodiment of the present invention, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle;

[0113] The determining unit 32 is configured to collect speed information of the vehicle, determine a first distribution feature of the vehicle within a first preset time range based on the speed information of the vehicle, and determine whether the first data is a real data type or a false data type based on the first distribution feature; and / or,

[0114] Determine the distance moved by the vehicle based on the location information of the vehicle, determine the acceleration information of the vehicle based on the distance moved by the vehicle and the speed information of the vehicle, determine a second distribution feature of the vehicle within a second preset time range based on the acceleration information of the vehicle, and determine whether the first data is a true data type or a false data type based on the second distribution feature.

[0115] In an optional embodiment of the present invention, the first data includes at least one of the following: location information of the vehicle, speed information of the vehicle; Figure 5 As shown, the device further includes a preprocessing unit 34, which is used to perform normalization processing on the plurality of the first data;

[0116] The training unit 33 is used to train the initial classification model based on the normalized plurality of first data and the corresponding labeled data.

[0117] In the embodiment of the present invention, the first acquisition unit 31, the determination unit 32, the training unit 33 and the preprocessing unit 34 in the device can all be implemented by a central processing unit (CPU), a digital signal processor (DSP), a microcontroller unit (MCU) or a programmable gate array (FPGA) in practical applications.

[0118] It should be noted that: the model training device provided in the above embodiment only uses the division of the above program modules as an example when performing model training. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device is divided into different program modules to complete all or part of the processing described above. In addition, the model training device provided in the above embodiment belongs to the same concept as the model training method embodiment. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0119] The embodiment of the present invention also provides a false terminal identification device. Figure 6 FIG. 1 is a schematic diagram of the structure of a false terminal identification device according to an embodiment of the present invention; Figure 6 As shown, the device includes a second acquisition unit 41 and a classification unit 42; wherein,

[0120] The second acquisition unit 41 is used to obtain second data sent by the vehicle-mounted terminal, where the second data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle;

[0121] The classification unit 42 is used to determine the type of the second data based on a terminal recognition model, and determine whether the vehicle-mounted terminal is a false terminal based on the type of the second data; wherein the terminal recognition model is obtained using the model training method described in an embodiment of the present invention.

[0122] In an optional embodiment of the present invention, Figure 7 As shown, the device further includes a processing unit 43 for, when the vehicle-mounted terminal is determined to be a false terminal, performing corresponding processing according to the duration for which the vehicle-mounted terminal is determined to be a false terminal.

[0123] In an optional embodiment of the present invention, the processing unit 43 is used to output a warning prompt message when the duration time that the vehicle-mounted terminal is determined as a false terminal reaches a first preset threshold; to stop pushing data to the vehicle-mounted terminal when the duration time that the vehicle-mounted terminal is determined as a false terminal reaches a second preset threshold; and to record the identification of the vehicle-mounted terminal in a blacklist when the duration time that the vehicle-mounted terminal is determined as a false terminal reaches a third preset threshold; wherein the first preset threshold is less than the second preset threshold, and the second preset threshold is less than the third preset threshold.

[0124] In an embodiment of the present invention, the classification unit 42 and the processing unit 43 in the device can be implemented by a CPU, a DSP, an MCU or an FPGA in practical applications; the second acquisition unit 41 in the device can be implemented by a communication module (including: a basic communication kit, an operating system, a communication module, a standardized interface and protocol, etc.) and a transceiver antenna in practical applications.

[0125] It should be noted that: the false terminal identification device provided in the above embodiment only uses the division of the above program modules as an example when performing false terminal identification. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device is divided into different program modules to complete all or part of the processing described above. In addition, the false terminal identification device provided in the above embodiment and the false terminal identification method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0126] An embodiment of the present invention further provides an electronic device. Figure 8 FIG. 1 is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Figure 8 As shown, the electronic device includes a memory 52, a processor 51, and a computer program stored in the memory 52 and executable on the processor 51. When the processor 51 executes the program, the steps of the model training method described in the embodiment of the present invention are implemented; or, when the processor 51 executes the program, the steps of the model training method described in the embodiment of the present invention are implemented.

[0127] Optionally, the electronic device may further include a communication interface 53. The various components in the electronic device are coupled together via a bus system 54. It is understood that the bus system 54 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 54 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 8 Various buses are labeled as bus system 54 .

[0128] It can be understood that the memory 52 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAM bus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 52 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0129] The method disclosed in the above embodiment of the present invention can be applied to the processor 51, or implemented by the processor 51. The processor 51 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 51 or the instruction in the form of software. The above processor 51 can be a general processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 51 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor to execute, or the hardware and software modules in the decoding processor can be combined to execute. The software module can be located in a storage medium, which is located in the memory 52. ​​The processor 51 reads the information in the memory 52 and completes the steps of the above method in combination with its hardware.

[0130] In an exemplary embodiment, the electronic device may be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD), field programmable gate array (FPGA), general processor, controller, microcontroller (MCU), microprocessor, or other electronic components to execute the aforementioned method.

[0131] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the model training method described in the embodiment of the present invention; or, when executed by a processor, implements the steps of the false terminal identification method described in the embodiment of the present invention.

[0132] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0133] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0134] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0135] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium, which, when executed, executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, disks or optical disks.

[0136] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0137] The methods disclosed in several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0138] The features disclosed in several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0139] The features disclosed in several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0140] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A model training method, characterized in that: The method comprises: Obtaining first data sent by a plurality of vehicle-mounted terminals, and determining annotated data corresponding to the first data; the first data is data representing a driving state of a vehicle obtained by the vehicle-mounted terminal in a process of accompanying the vehicle in driving; the first data includes at least one of the location information of the vehicle and the speed information of the vehicle; the annotated data is determined by the first data, and / or the annotated data is pre-set data; the annotated data represents that the type of the first data is a real data type or a false data type; the pre-set data indicates that the regularity of the first data reported by the vehicle-mounted terminal is first analyzed and processed, and the annotated data corresponding to the first data is determined according to the analysis and processing result; An initial classification model is trained based on the plurality of first data and the labeled data corresponding to the first data to obtain a terminal recognition model.

2. The method according to claim 1, characterized in that The first data at least includes location information of the vehicle; Determining the labeled data by using the first data includes: Determine trajectory data based on the obtained multiple location information of the vehicle; Determine the map data corresponding to the location information of the vehicle, and obtain the road data corresponding to the location information of the vehicle from the map data; The similarity between the trajectory data and the road data is determined, and based on the similarity, the first data is determined to be a real data type or a false data type.

3. The method according to claim 1, characterized in that The first data includes at least one of the following: location information of the vehicle and speed information of the vehicle; determining the labeled data through the first data includes: collecting speed information of the vehicle, determining a first distribution feature of the vehicle within a first preset time range based on the speed information of the vehicle, and determining whether the first data is a real data type or a false data type based on the first distribution feature; and / or, Determine the distance moved by the vehicle based on the location information of the vehicle, determine the acceleration information of the vehicle based on the distance moved by the vehicle and the speed information of the vehicle, determine a second distribution feature of the vehicle within a second preset time range based on the acceleration information of the vehicle, and determine whether the first data is a true data type or a false data type based on the second distribution feature.

4. The method according to any one of claims 1 to 3, characterized in that: The first data includes at least one of the following: location information of the vehicle, speed information of the vehicle; The training of the initial classification model based on the plurality of first data and the labeled data corresponding to the first data includes: The plurality of first data are normalized, and an initial classification model is trained based on the plurality of first data after the normalization and the corresponding labeled data.

5. A method for identifying a fake terminal, characterized in that: The method comprises: Obtaining second data sent by the vehicle-mounted terminal, where the second data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle; Determine the type of the second data based on a terminal recognition model, and determine whether the vehicle-mounted terminal is a false terminal based on the type of the second data; wherein the terminal recognition model is obtained using the model training method described in any one of claims 1 to 4.

6. The method according to claim 5, characterized in that The method further includes: when it is determined that the vehicle-mounted terminal is a false terminal, performing corresponding processing according to a duration for which the vehicle-mounted terminal is determined to be a false terminal.

7. The method according to claim 6, characterized in that The performing corresponding processing according to the duration of time during which the vehicle-mounted terminal is determined to be a false terminal includes: When the duration for which the vehicle-mounted terminal is determined to be a false terminal reaches a first preset threshold, outputting a warning prompt message; When the duration for which the vehicle-mounted terminal is determined to be a false terminal reaches a second preset threshold, stopping pushing data to the vehicle-mounted terminal; When the duration for which the vehicle-mounted terminal is determined to be a false terminal reaches a third preset threshold, recording the identification of the vehicle-mounted terminal in a blacklist; The first preset threshold is smaller than the second preset threshold, and the second preset threshold is smaller than the third preset threshold.

8. A model training device, characterized in that: The device comprises: a first acquisition unit, a determination unit and a training unit; wherein, The first acquisition unit is used to obtain first data of multiple vehicle terminals; the first data is data representing the driving state of the vehicle obtained by the vehicle terminal during the driving of the vehicle; the first data includes at least one of the location information of the vehicle and the speed information of the vehicle; The determining unit is used to determine the annotated data corresponding to the first data; the annotated data is determined by the first data, and / or the annotated data is pre-set data; the annotated data indicates that the type of the first data is a real data type or a false data type; the pre-set data indicates that the regularity of the first data reported by the vehicle-mounted terminal is first analyzed and processed, and the annotated data corresponding to the first data is determined according to the analysis and processing result; The training unit is used to train the initial classification model based on the multiple first data and the labeled data corresponding to the first data to obtain the terminal recognition model.

9. The device according to claim 8, characterized in that The first data at least includes location information of the vehicle; The determination unit is used to determine trajectory data based on multiple location information of the vehicle; determine map data corresponding to the location information of the vehicle, and obtain road data corresponding to the location information of the vehicle from the map data; determine the similarity between the trajectory data and the road data, and determine whether the first data is a real data type or a false data type based on the similarity.

10. The device according to claim 8, characterized in that The first data includes at least one of the following: location information of the vehicle, speed information of the vehicle; The determining unit is used to collect speed information of the vehicle, determine a first distribution feature of the vehicle within a first preset time range based on the speed information of the vehicle, and determine whether the first data is a real data type or a false data type based on the first distribution feature; and / or, Determine the distance moved by the vehicle based on the location information of the vehicle, determine the acceleration information of the vehicle based on the distance moved by the vehicle and the speed information of the vehicle, determine a second distribution feature of the vehicle within a second preset time range based on the acceleration information of the vehicle, and determine whether the first data is a true data type or a false data type based on the second distribution feature.

11. The device according to any one of claims 8 to 10, characterized in that The first data includes at least one of the following: location information of the vehicle, speed information of the vehicle; The device further comprises a preprocessing unit, configured to perform normalization processing on the plurality of the first data; The training unit is used to train the initial classification model based on the normalized plurality of first data and the corresponding labeled data.

12. A false terminal identification device, characterized in that: The device comprises a second acquisition unit and a classification unit; wherein, The second acquisition unit is used to obtain second data sent by the vehicle-mounted terminal, where the second data is data representing the driving state of the vehicle obtained by the vehicle-mounted terminal during the driving of the vehicle; The classification unit is used to determine the type of the second data based on a terminal recognition model, and determine whether the vehicle-mounted terminal is a false terminal based on the type of the second data; wherein the terminal recognition model is obtained using the model training method described in any one of claims 1 to 4.

13. The device according to claim 12, characterized in that The device further includes a processing unit for, when determining that the vehicle-mounted terminal is a false terminal, performing corresponding processing according to a duration for which the vehicle-mounted terminal is determined to be a false terminal.

14. The device according to claim 13, characterized in that The processing unit is used to output a warning prompt message when the duration time that the vehicle-mounted terminal is determined as a false terminal reaches a first preset threshold; to stop pushing data to the vehicle-mounted terminal when the duration time that the vehicle-mounted terminal is determined as a false terminal reaches a second preset threshold; and to record the identification of the vehicle-mounted terminal in a blacklist when the duration time that the vehicle-mounted terminal is determined as a false terminal reaches a third preset threshold; wherein the first preset threshold is less than the second preset threshold, and the second preset threshold is less than the third preset threshold.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the model training method described in any one of claims 1 to 4 are implemented; or, when the program is executed by a processor, the steps of the false terminal identification method described in any one of claims 5 to 7 are implemented.

16. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the model training method described in any one of claims 1 to 4 are implemented; or, when the processor executes the program, the steps of the false terminal identification method described in any one of claims 5 to 7 are implemented.

Images