Method and apparatus for identifying target behavior and electronic device

Abstract

The present disclosure relates to a method and apparatus for identifying a target behavior, an electronic device, and a computer-readable storage medium. The method for identifying a target behavior includes: obtaining a vehicle coordinate stream of a current order; extracting a road condition classification feature of the current order based on the vehicle coordinate stream; inputting the road condition classification feature into a pre-trained road condition classification model to obtain a first recognition result; and if the first recognition result is a first preset value, it is recognized that the driver corresponding to the current order has a target behavior. In the embodiments of the present disclosure, by extracting the road condition classification feature of the current order based on the vehicle coordinate stream of the current order and inputting the road condition classification feature into the pre-trained road condition classification model to identify whether the driver corresponding to the current order has a target behavior, since the road condition classification model fully considers the influence of road conditions on the vehicle driving speed, the accuracy of recognition is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technologies, and in particular, to a method and apparatus for identifying a target behavior, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the development of intelligent devices and mobile Internet technologies, the popularity of ride-hailing software has brought great convenience to people's travel. Passengers can send orders through the ride-hailing software, and the ride-hailing software sends the orders to the background server. The background server distributes the orders to the drivers within a predetermined range around the passenger. After receiving the order, the driver can respond and accept the order.

[0003] However, after accepting an order, in order to select orders, the driver will deliberately delay the time when picking up the passenger, so as to let the user cancel the order. In addition, delaying the driving during the drop-off process will cause the user to incur higher costs. Therefore, how to identify delayed driving is a problem that needs to be solved.

[0004] Currently, delayed driving can be identified by the driving speed of the vehicle. For example, if the driving speed of the vehicle is lower than a preset threshold, it is identified that the vehicle is driving with delay. However, this identification method only identifies through the driving speed, and the identification dimension is single, and the accuracy is poor. Summary of the Invention

[0005] In view of this, the present application provides a method and apparatus for identifying a target behavior, an electronic device, and a computer-readable storage medium.

[0006] Specifically, the present application is implemented through the following technical solutions:

[0007] According to a first aspect of an embodiment of the present disclosure, there is provided a method for identifying a target behavior, the method including:

[0008] Obtain a vehicle coordinate stream of a current order;

[0009] Extract a road condition classification feature of the current order based on the vehicle coordinate stream;

[0010] Input the road condition classification feature into a pre-trained road condition classification model to obtain a first recognition result;

[0011] If the first recognition result is a first preset value, it is recognized that the driver corresponding to the current order has the target behavior.

[0012] In an embodiment, the method further includes:

[0013] If the first recognition result is a second preset value, extract a consumption duration classification feature of the current order based on the vehicle coordinate stream;

[0014] Input the classified consumption duration feature into a pre-trained consumption duration classification model to obtain a second recognition result;

[0015] If the second recognition result is the first preset value, it is recognized that the driver corresponding to the current order has the target behavior.

[0016] In one embodiment, the extraction of the road condition classification feature of the current order based on the vehicle coordinate stream includes:

[0017] Extract the target road section based on the vehicle coordinate stream;

[0018] Obtain the road condition level of each target road section;

[0019] Classify the target road sections according to the road condition level to obtain the number of target road sections included in each road condition level;

[0020] Take the number of target road sections included in each road condition level as the road condition classification feature of the current order.

[0021] In one embodiment, the extraction of the consumption duration classification feature of the current order based on the vehicle coordinate stream includes:

[0022] Extract the target road section based on the vehicle coordinate stream and calculate the consumption duration of each target road section;

[0023] Determine the road sections other than the target road sections in the vehicle coordinate stream as non-target road sections and calculate the consumption duration of each non-target road section;

[0024] Classify the target road sections and the non-target road sections respectively according to the consumption duration to obtain the number of target road sections in different consumption duration ranges and the number of non-target road sections in different consumption duration ranges;

[0025] Take the number of target road sections in different consumption duration ranges and the number of non-target road sections in different consumption duration ranges as the consumption duration classification feature of the current order.

[0026] In one embodiment, the extraction of the target road section based on the vehicle coordinate stream includes:

[0027] Calculate the distance between adjacent two coordinate points in chronological order;

[0028] Calculate the time interval between adjacent two coordinate points;

[0029] Based on the distance and the time interval, calculate the driving speed between adjacent two coordinate points;

[0030] Determine the target section based on consecutive coordinate points where the driving speed is lower than the normal driving speed under the corresponding road condition level.

[0031] In one embodiment, the method further includes:

[0032] Collect the vehicle coordinate stream of historical orders;

[0033] Extract the road condition classification features of the historical orders based on the vehicle coordinate stream of the historical orders;

[0034] Train the road condition classification model based on the road condition classification features of the historical orders.

[0035] In one embodiment, the method further includes:

[0036] Collect the vehicle coordinate stream of historical orders;

[0037] Extract the consumption duration classification features of the historical orders based on the vehicle coordinate stream of the historical orders;

[0038] Train the consumption duration classification model based on the consumption duration classification features of the historical orders.

[0039] According to the second aspect of the embodiments of the present disclosure, there is provided an identification device for a target behavior, the device includes:

[0040] An acquisition module, configured to acquire the vehicle coordinate stream of the current order;

[0041] An extraction module, configured to extract the road condition classification features of the current order based on the vehicle coordinate stream acquired by the acquisition module;

[0042] A first input module, configured to input the road condition classification features extracted by the extraction module into a pre-trained road condition classification model to obtain a first recognition result;

[0043] A first recognition module, configured to, if the first recognition result obtained by the first input module is a first preset value, recognize that the driver corresponding to the current order has the target behavior.

[0044] In one embodiment, the device further includes:

[0045] A determination extraction module, configured to, if the first recognition result is a second preset value, extract the consumption duration classification features of the current order based on the vehicle coordinate stream;

[0046] A second input module, configured to input the consumption duration classification features extracted by the determination extraction module into a pre-trained consumption duration classification model to obtain a second recognition result;

[0047] A second recognition module, configured to recognize that the driver corresponding to the current order has the target behavior if the second recognition result is the first preset value.

[0048] According to a third aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium storing a computer program for executing the above-mentioned method for recognizing target behaviors.

[0049] According to a fourth aspect of the embodiments of the present disclosure, there is provided an electronic device including a processor, a memory, and a computer program stored on the memory and executable on the processor, where the processor implements the above-mentioned method for recognizing target behaviors when executing the computer program.

[0050] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0051] By extracting the road condition classification features of the current order based on the vehicle coordinate stream of the current order and inputting the road condition classification features into a pre-trained road condition classification model to recognize whether the driver corresponding to the current order has the target behavior, since the road condition classification model fully considers the influence of road conditions on the vehicle driving speed, the recognition accuracy is improved.

[0052] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present invention and used together with the specification to explain the principles of the present invention.

[0054] Figure 1 is a flowchart of a method for recognizing target behaviors shown in an exemplary embodiment of the present application.

[0055] Figure 2 is a flowchart of a method for extracting road condition classification features of the current order shown in an exemplary embodiment of the present application.

[0056] Figure 3 is a flowchart of another method for recognizing target behaviors shown in an exemplary embodiment of the present application.

[0057] Figure 4 is a flowchart of a method for extracting consumption duration classification features of the current order shown in an exemplary embodiment of the present application.

[0058] Figure 5 is a flowchart of a method for training a road condition classification model shown in an exemplary embodiment of the present application.

[0059] Figure 6 It is a flowchart for training a consumption duration classification model shown in an exemplary embodiment of the present application.

[0060] Figure 7 It is a hardware structure diagram of an electronic device where an identification device for a target behavior shown in an exemplary embodiment of the present application is located.

[0061] Figure 8 It is a block diagram of an identification device for a target behavior shown in an exemplary embodiment of the present application. Detailed implementation manners

[0062] Here, exemplary embodiments will be described in detail, and examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0063] The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0064] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to a determination".

[0065] Figure 1 It is an identification method for a target behavior shown in an exemplary embodiment of the present application. As Figure 1 shown, the method includes:

[0066] Step S101, obtaining a vehicle coordinate stream of the current order.

[0067] Wherein, the vehicle coordinate stream refers to a series of coordinate points passed by the vehicle during driving. This series of coordinate points is in time order and corresponds to the road conditions at that time.

[0068] Step S102: Extract the road condition classification features of the current order based on the vehicle coordinate stream.

[0069] In this embodiment, the road conditions can be classified into different levels according to the road smoothness. Under different road condition levels, the normal driving speeds of vehicles are different. Among them, the corresponding relationship between the road condition levels and the normal driving speeds of vehicles can be shown in Table 1:

[0070] Table 1 Corresponding relationship between road condition levels and normal driving speeds of vehicles

[0071] Level 1 Level 2 …… Level i <![CDATA[v1]]> <![CDATA[v2]]> …… <![CDATA[v i >

[0072] In this embodiment, by classifying the road conditions and setting the normal driving speeds of vehicles under different road condition levels, the situation of slow vehicle driving caused by objective factors is excluded, thereby improving the accuracy of the original data.

[0073] Among them, as Figure 2 shown, extracting the road condition classification features of the current order may include:

[0074] Step S1021: Extract the target road section based on the vehicle coordinate stream.

[0075] Among them, the target road section may be a procrastinating road section.

[0076] In this embodiment, the distance between two adjacent coordinate points can be calculated in chronological order, and the time interval between two adjacent coordinate points can be calculated. Then, based on the calculated distance and time interval, the driving speed between two adjacent coordinate points can be calculated. Finally, based on the consecutive coordinate points whose driving speed is lower than the normal driving speed under the corresponding road condition level, the target road section is determined.

[0077] For example, the distance between two adjacent coordinate points can be calculated using formula (1), and formula (1) is:

[0078]

[0079] Among them, S represents the distance between two adjacent coordinate points, longtitude1 and latitude1 respectively represent the longitude and latitude of the previous coordinate point, and longtitude2 and latitude2 respectively represent the longitude and latitude of the subsequent coordinate point.

[0080] For example, the driving speed between two adjacent coordinate points can be calculated using formula (2), and formula (2) is:

[0081] v = S / t

[0082] Among them, v represents the driving speed between two adjacent coordinate points, S represents the distance between two adjacent coordinate points, and t represents the time interval between two adjacent coordinate points.

[0083] After calculating the driving speed between two adjacent coordinate points, the normal speed under the corresponding road condition level can be found according to Table 1. If the calculated driving speed is less than the normal speed under the corresponding road condition level, it is determined that the road section between these two adjacent coordinate points is a delay section.

[0084] By using the above method, all the delay sections between two adjacent coordinate points can be determined.

[0085] Step S1022: Obtain the road condition level of each target road section.

[0086] In this embodiment, the road condition level corresponding to each target road section can be obtained.

[0087] Step S1023: Classify the target road sections according to the road condition level to obtain the number of target road sections included in each road condition level.

[0088] For example, if the number of target road sections extracted based on the vehicle coordinate stream of the current order is 10, and these 10 target road sections correspond to different road condition levels, count the number of target road sections included in each road condition level. For example, the number of target road sections included in road condition level 1 is 3, the number of target road sections included in road condition level 2 is 4, and the number of target road sections included in road condition level 3 is 3.

[0089] Step S1024: Use the number of target road sections included in each road condition level as the road condition classification feature of the current order.

[0090] In this embodiment, the number of target road sections included in each obtained road condition level is used as the road condition classification feature of the current order. Continuing with the above example, the number of target road sections included in road condition level 1 is 3, the number of target road sections included in road condition level 2 is 4, and the number of target road sections included in road condition level 3 is 3 are used as the road condition classification feature of the current order.

[0091] In this embodiment, by extracting the road condition classification feature of the current order based on the vehicle coordinate stream, the influence of road conditions on the vehicle driving speed is considered, which is beneficial to improving the accuracy of recognition.

[0092] Step S103: Input the road condition classification feature into a pre-trained road condition classification model to obtain a first recognition result.

[0093] Step S104: If the first recognition result is a first preset value, it is recognized that the driver corresponding to the current order has a target behavior.

[0094] For example, if the first preset value represents procrastination behavior, it is considered that the driver corresponding to the current order has procrastination behavior.

[0095] In the above embodiment, by extracting the road condition classification feature of the current order based on the vehicle coordinate stream of the current order and inputting the road condition classification feature into the pre-trained road condition classification model to identify whether the driver corresponding to the current order has the target behavior, since the road condition classification model fully considers the influence of road conditions on the vehicle driving speed, the accuracy of the identification is improved.

[0096] Optionally, as Figure 3 shown, after the above step S103, the method may further include:

[0097] Step S105, if the first recognition result is the second preset value, extract the consumption duration classification feature of the current order based on the vehicle coordinate stream.

[0098] For example, if the second preset value represents non-procrastination behavior, extract the consumption duration classification feature of the current order based on the vehicle coordinate stream.

[0099] Among them, as Figure 4 shown, extracting the consumption duration classification feature of the current order may include:

[0100] Step S1051, extract the target road sections based on the vehicle coordinate stream and calculate the consumption duration of each target road section.

[0101] Among them, the method for extracting the target road sections based on the vehicle coordinate stream can refer to the relevant description in step S1021 and will not be elaborated here.

[0102] Step S1052, determine the road sections other than the target road sections in the vehicle coordinate stream as non-target road sections and calculate the consumption duration of each non-target road section.

[0103] Step S1053, classify the target road sections and non-target road sections respectively according to the consumption duration to obtain the number of target road sections in different consumption duration ranges and the number of non-target road sections in different consumption duration ranges.

[0104] For example, it can be obtained that the number of target road sections in the range of 0 to t11 is a1, the number of target road sections in the range of t11 to t12 is a2, the number of target road sections in the range of t12 to t13 is a3, etc., and the number of target road sections in the range of 0 to t21 is b1, the number of target road sections in the range of t21 to t22 is b2, the number of target road sections in the range of t22 to t23 is b3, etc.

[0105] Step S1054: Use the number of target road segments in different consumption duration ranges and the number of non-target road segments in different consumption duration ranges as the consumption duration classification features of the current order.

[0106] In this embodiment, by using the number of target road segments in different consumption duration ranges and the number of non-target road segments in different consumption duration ranges as the consumption duration classification features of the current order, the consumption duration classification features can include the driving behavior of the corresponding driver throughout the order, and can more comprehensively reflect the behavior of the corresponding driver.

[0107] Step S106: Input the consumption duration classification features into a pre-trained consumption duration classification model to obtain a second recognition result.

[0108] Step S107: If the second recognition result is a first preset value, it is recognized that the driver corresponding to the current order has a target behavior.

[0109] In the above embodiment, after initially not determining that the driver corresponding to the current order has a procrastination behavior, the consumption duration classification features of the current order are extracted based on the vehicle coordinate stream, and whether the driver corresponding to the current order has a target behavior is recognized based on the consumption duration classification features of the current order, that is, the target behavior is comprehensively recognized through the road condition classification model and the consumption duration classification model, and the recognition result will be more accurate.

[0110] Optionally, before step S103, the method may further include training a road condition classification model, as Figure 5 shown, the process of training the road condition classification model includes:

[0111] Step S501: Collect the vehicle coordinate streams of historical orders.

[0112] Step S502: Based on the vehicle coordinate streams of historical orders, extract the road condition classification features of historical orders.

[0113] Among them, the method of extracting the road condition classification features of historical orders is similar to the method of extracting the road condition classification features of the current order in step S102 above, and will not be elaborated here.

[0114] Step S503: Train a road condition classification model based on the road condition classification features of historical orders.

[0115] Among them, algorithms such as neural networks or classification trees can be used but are not limited to training the road condition classification features of historical orders to obtain a road condition classification model.

[0116] In the above embodiments, by using the vehicle coordinate stream of historical orders, the road condition classification features of historical orders are extracted, and a road condition classification model is trained based on the road condition classification features of historical orders, so that the trained road condition classification model fully considers the influence of road conditions on vehicle driving speed, which is beneficial to improving the recognition accuracy.

[0117] Optionally, before step S106, the method may further include training a consumption duration classification model, as Figure 6 shown, training a consumption duration classification model may include:

[0118] Step S601, collect the vehicle coordinate stream of historical orders.

[0119] Step S602, based on the vehicle coordinate stream of historical orders, extract the consumption duration classification features of historical orders.

[0120] Among them, the method for extracting the consumption duration classification features of historical orders is similar to the method for extracting the consumption duration classification features of the current order in step S105, which will not be elaborated here.

[0121] Step S603, train a consumption duration classification model based on the consumption duration classification features of historical orders.

[0122] Among them, algorithms such as neural networks or classification trees can be used but are not limited to training the consumption duration classification features of historical orders to obtain a consumption duration classification model.

[0123] In the above embodiments, by using the vehicle coordinate stream of historical orders, the consumption duration classification features of historical orders are extracted, and a consumption duration classification model is trained based on the consumption duration classification features of historical orders, so that the trained consumption duration classification model can include the driving behavior of the corresponding driver throughout the order, improving the data comprehensiveness of the consumption duration classification model, and thus being beneficial to improving the recognition accuracy.

[0124] Corresponding to the embodiments of the method for identifying target behaviors described above, the present application also provides embodiments of an apparatus for identifying target behaviors.

[0125] The embodiments of the apparatus for identifying target behaviors in the present application can be applied to an electronic device. Among them, the electronic device can be a server. The apparatus embodiments can be implemented by software, or by hardware, or by a combination of software and hardware. As Figure 7 shown, it is a hardware structure diagram of an electronic device where the apparatus 700 for identifying target behaviors in the present application is located. The electronic device includes a processor 710, a memory 720, and a computer program stored in the memory 720 and executable on the processor 710. When the processor 710 executes the computer program, the above method for identifying target behaviors is implemented. In addition to Figure 7In addition to the processor 710 and the memory 720 shown, in an embodiment, the electronic device where the apparatus is located generally may further include other hardware according to the actual function of identifying the target behavior, which will not be elaborated herein.

[0126] Figure 8 is a block diagram of an apparatus for identifying a target behavior shown in an exemplary embodiment of the present application. As Figure 8 shown, the identification apparatus includes: an acquisition module 81, an extraction module 82, a first input module 83, and a first identification module 84.

[0127] The acquisition module 81 is configured to acquire the vehicle coordinate stream of the current order.

[0128] Wherein, the vehicle coordinate stream refers to a series of coordinate points passed by the vehicle during driving, and this series of coordinate points is in order according to time and corresponds to the road conditions at that time.

[0129] The extraction module 82 is configured to extract the road condition classification feature of the current order based on the vehicle coordinate stream acquired by the acquisition module 81.

[0130] In this embodiment, the road conditions can be divided into different levels according to the road smoothness. Under different road condition levels, the normal driving speed of the vehicle is different. Among them, the corresponding relationship between the road condition level and the normal driving speed of the vehicle can be shown in Table 1.

[0131] In this embodiment, by classifying the road conditions and setting the normal driving speed of the vehicle under different road condition levels, the situation of slow vehicle driving caused by objective factors is excluded, thereby improving the accuracy of the original data.

[0132] Among them, the process of extracting the road condition classification feature of the current order can be as Figure 2 shown, which will not be elaborated herein.

[0133] The first input module 83 is configured to input the road condition classification feature extracted by the extraction module 82 into a pre-trained road condition classification model to obtain a first identification result.

[0134] The first identification module 84 is configured to, if the first identification result obtained by the first input module 83 is a first preset value, identify that the driver corresponding to the current order has a target behavior.

[0135] For example, if the first preset value represents a procrastination behavior, it is considered that the driver corresponding to the current order has a procrastination behavior.

[0136] In the above embodiments, by extracting the road condition classification features of the current order based on the vehicle coordinate stream of the current order and inputting the road condition classification features into a pre-trained road condition classification model to identify whether the driver corresponding to the current order has a target behavior, since the road condition classification model fully considers the impact of road conditions on the vehicle driving speed, the accuracy of the identification is improved.

[0137] For the implementation processes of the functions and roles of each unit in the above device, please refer to the implementation processes of the corresponding steps in the above method for details, and will not be elaborated here.

[0138] In an exemplary embodiment, a computer-readable storage medium is further provided. The storage medium stores a computer program, and the computer program is used to execute the above method for identifying target behaviors. The method for identifying target behaviors includes:

[0139] Obtain the vehicle coordinate stream of the current order;

[0140] Extract the road condition classification features of the current order based on the vehicle coordinate stream;

[0141] Input the road condition classification features into a pre-trained road condition classification model to obtain a first identification result;

[0142] If the first identification result is a first preset value, it is identified that the driver corresponding to the current order has a target behavior.

[0143] The above computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0144] For the device embodiment, since it basically corresponds to the method embodiment, please refer to the partial description of the method embodiment for the relevant parts. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0145] Other embodiments of the present application will be readily contemplated by those skilled in the art upon considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include known common general knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only illustrative, and the true scope and spirit of the present application are pointed out by the claims.

[0146] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0147] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for identifying a target behavior, characterized in that, The method includes: Obtaining the vehicle coordinate stream of the current order; Extracting the road condition classification features of the current order based on the vehicle coordinate stream; the road condition classification features include the number of target road segments included in each road condition level; the target road segments are the delayed road segments determined according to the vehicle coordinate stream; Inputting the road condition classification features into a pre-trained road condition classification model to obtain a first recognition result; If the first recognition result is a first preset value, it is recognized that the driver corresponding to the current order has the target behavior.

2. The method according to claim 1, characterized in that, The method further includes: If the first recognition result is a second preset value, extracting the consumption duration classification features of the current order based on the vehicle coordinate stream; Inputting the consumption duration classification features into a pre-trained consumption duration classification model to obtain a second recognition result; If the second recognition result is the first preset value, it is recognized that the driver corresponding to the current order has the target behavior.

3. The method according to claim 1, characterized in that, The extracting the road condition classification features of the current order based on the vehicle coordinate stream includes: Extracting target road segments based on the vehicle coordinate stream; Obtaining the road condition level of each target road segment; Classifying the target road segments according to the road condition level to obtain the number of target road segments included in each road condition level; Taking the number of target road segments included in each road condition level as the road condition classification features of the current order.

4. The method according to claim 2, characterized in that, The extracting the consumption duration classification features of the current order based on the vehicle coordinate stream includes: Extracting target road segments based on the vehicle coordinate stream and calculating the consumption duration of each target road segment; Determining the road segments other than the target road segments in the vehicle coordinate stream as non-target road segments and calculating the consumption duration of each non-target road segment; Classifying the target road segments and the non-target road segments respectively according to the consumption duration to obtain the number of target road segments in different consumption duration ranges and the number of non-target road segments in different consumption duration ranges; Taking the number of target road segments in different consumption duration ranges and the number of non-target road segments in different consumption duration ranges as the consumption duration classification features of the current order.

5. The method according to claim 3 or 4, characterized in that, The extracting target road segments based on the vehicle coordinate stream includes: Calculating the distance between two adjacent coordinate points in sequence according to time; Calculating the time interval between two adjacent coordinate points; Calculating the driving speed between two adjacent coordinate points based on the distance and the time interval; Determining the target road segments based on the consecutive coordinate points whose driving speed is lower than the normal driving speed under the corresponding road condition level.

6. The method according to claim 1 or 2, characterized in that The method further includes: Collecting the vehicle coordinate streams of historical orders; Extracting the road condition classification features of the historical orders based on the vehicle coordinate streams of the historical orders; Training the road condition classification model based on the road condition classification features of the historical orders.

7. The method according to claim 2, characterized in that, The method further includes: Collecting the vehicle coordinate streams of historical orders; Extracting the consumption duration classification features of the historical orders based on the vehicle coordinate streams of the historical orders; Training the consumption duration classification model based on the consumption duration classification features of the historical orders.

8. An identification device for a target behavior, characterized in that, The device includes: An acquisition module, configured to acquire a vehicle coordinate stream of a current order; An extraction module, configured to extract a road condition classification feature of the current order based on the vehicle coordinate stream acquired by the acquisition module; the road condition classification feature includes the number of target road sections included in each road condition level; the target road section is a delay road section determined according to the vehicle coordinate stream; A first input module, configured to input the road condition classification feature extracted by the extraction module into a pre-trained road condition classification model to obtain a first recognition result; A first recognition module, configured to recognize that the driver corresponding to the current order has the target behavior if the first recognition result obtained by the first input module is a first preset value.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program is used to execute the recognition method of the target behavior according to any one of claims 1-7 above.

10. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the recognition method of the target behavior according to any one of claims 1-7 above is implemented.

Images