Method and device for generating waybill, electronic equipment and readable storage medium

Abstract

Embodiments of the present application provide a waybill generation method and device, electronic equipment and readable storage medium, relating to the field of logistics. After obtaining the stay point data in the driving process of the vehicle and the image data corresponding to the stay point data, the image data corresponding to the stay point data is subjected to image recognition to obtain at least one target stay point meeting a target scenario. By clustering each target stay point, at least one stay point cluster is obtained. After obtaining the stay point cluster, the working position of the vehicle can be determined according to the center and radius of each stay point cluster. Then, the waybill for the vehicle transportation is generated according to the working position. In this way, the waybill can be efficiently and truly generated by replacing the manual waybill recording mode, thereby reducing the labor cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics, in particular to a waybill generation method and device, electronic equipment and readable storage medium. BACKGROUND

[0002] In the logistics industry, freight transportation is mostly completed by transportation waybills and their transportation times, so it is extremely important for financial calculation to accurately check each waybill.

[0003] Currently, most of the waybills for freight transportation are recorded offline by manual recording or by drivers manually recording in related program software. Such methods still have many loopholes in waybill recording, and cannot efficiently and cost-effectively calculate and count the real number of business trips and the real number of waybills, thus failing to better perform freight settlement for freight companies. SUMMARY

[0004] Therefore, the embodiments of the present application provide a waybill generation method, device, electronic equipment and readable storage medium, which can efficiently and truly generate waybills by replacing the manual waybill recording method, thereby reducing labor costs.

[0005] In a first aspect, the embodiments of the present application provide a waybill generation method, which comprises:

[0006] obtaining stop point data in a vehicle driving process and image data corresponding to the stop point data;

[0007] performing image recognition on the image data corresponding to the stop point data to obtain at least one target stop point meeting a target scenario;

[0008] clustering each target stop point to obtain at least one stop point cluster;

[0009] determining a work position of the vehicle according to the center and radius of each stop point cluster, and generating a waybill for vehicle transportation according to the work position.

[0010] In an optional embodiment, the image recognition on the image data corresponding to the stop point data to obtain at least one target stop point meeting a target scenario comprises:

[0011] extracting scene features of each scene image in the image data by a preset image recognition model, and analyzing and processing each scene feature to obtain a scene category to which each scene image belongs; each scene image corresponds to a stop point, and the stop point data comprises at least one stop point;

[0012] For each of the scene images, if a scene category to which the scene image belongs is a target scene, a stay point corresponding to the scene image is set as a target stay point.

[0013] In an optional implementation, the clustering of each of the target stay points to obtain at least one stay point cluster includes:

[0014] According to scenes corresponding to each of the target stay points, the target stay points are grouped to obtain a start operation group and an end operation group;

[0015] The target stay points in the start operation group and the target stay points in the end operation group are respectively clustered to obtain a stay point cluster corresponding to the start operation group and a stay point cluster corresponding to the end operation group.

[0016] In an optional implementation, the clustering of the target stay points in the start operation group and the target stay points in the end operation group to obtain the stay point cluster corresponding to the start operation group and the stay point cluster corresponding to the end operation group includes:

[0017] According to dates corresponding to each of the target stay points and a collection device, the target stay points in the start operation group are grouped to obtain at least one first combination, and the target stay points in the end operation group are grouped to obtain at least one second combination;

[0018] According to positions of each of the target stay points, each of the first combinations and each of the second combinations are respectively clustered to obtain a clustering result of each of the first combinations and a clustering result of each of the second combinations;

[0019] The clustering result of each of the first combinations is clustered to obtain the stay point cluster corresponding to the start operation group, and the clustering result of each of the second combinations is clustered to obtain the stay point cluster corresponding to the end operation group.

[0020] In an optional implementation, the determination of the operation position of the vehicle according to the center of each of the stay point clusters includes:

[0021] The start operation position of the vehicle is determined according to the center and the radius of the stay point cluster corresponding to the start operation group;

[0022] The end operation position of the vehicle is determined according to the center and the radius of the stay point cluster corresponding to the end operation group.

[0023] The generation of the waybill of the vehicle transportation according to the operation position includes:

[0024] According to the starting operation position and the ending operation position of the vehicle, a waybill of vehicle transportation is generated.

[0025] In an optional implementation, after the starting operation group corresponding stay point cluster and the ending operation group corresponding stay point cluster are obtained, the method further comprises:

[0026] Detecting whether there is a target stay point cluster with the same center as the starting operation group corresponding stay point cluster in the ending operation group corresponding stay point cluster;

[0027] If the target stay point cluster exists, the target stay point cluster is removed from the ending operation group corresponding stay point cluster.

[0028] In an optional implementation, the stay point data and the image data corresponding to the stay point data in the vehicle driving process are obtained, comprising:

[0029] Obtaining trajectory data of the vehicle driving;

[0030] According to the set speed condition, time condition and road condition, the trajectory data is screened to obtain stay point data satisfying the speed condition, the time condition and the road condition;

[0031] For each stay point in the stay point data, a scene image corresponding to the stay point is obtained to obtain image data corresponding to the stay point data.

[0032] In an optional implementation, after the operation position of the vehicle is determined according to the center and the radius of each stay point cluster, the method further comprises:

[0033] According to the set interference position, it is detected whether there is a target position with the same interference position as the operation position of the vehicle;

[0034] If the target position exists, the target position is removed.

[0035] In a second aspect, an embodiment of the present application provides a waybill generation device, the waybill generation device comprising:

[0036] A data acquisition module is configured to acquire stay point data in a vehicle driving process and image data corresponding to the stay point data;

[0037] An image recognition module is configured to perform image recognition on the image data corresponding to the stay point data to obtain at least one target stay point satisfying a target scene;

[0038] a data clustering module configured to cluster the target stay points to obtain at least one stay point clustering cluster;

[0039] a waybill generation module configured to determine a work position of the vehicle according to the center and the radius of each stay point clustering cluster, and generate a waybill of vehicle transportation according to the work position.

[0040] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the program, the waybill generation method in any of the preceding embodiments is implemented.

[0041] In a fourth aspect, an embodiment of the present application provides a readable storage medium, including a computer program, and when the computer program runs, the electronic device where the readable storage medium is located executes the waybill generation method in any of the preceding embodiments.

[0042] The waybill generation method, device, electronic device, and readable storage medium provided by the embodiments of the present application can obtain stay point data and image data corresponding to the stay point data in the driving process of a vehicle, perform image recognition on the image data corresponding to the stay point data to obtain at least one target stay point meeting a target scenario, cluster each target stay point to obtain at least one stay point clustering cluster, and then determine a work position of the vehicle according to the center and the radius of each stay point clustering cluster, and generate a waybill of vehicle transportation according to the work position. In this way, the waybill can be efficiently and truly generated by replacing the manual waybill recording mode, and the labor cost is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0043] The technical solutions and other beneficial effects of the present application will become apparent after a specific embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0044] Figure 1 A structural schematic diagram of an electronic device provided by an embodiment of the present application.

[0045] Figure 2 A flowchart of a waybill generation method provided by an embodiment of the present application.

[0046] Figure 3 Another flowchart of a waybill generation method provided by an embodiment of the present application.

[0047] Figure 4 Still another flowchart of a waybill generation method provided by an embodiment of the present application.

[0048] Figure 5A block diagram of a waybill generation device provided by an embodiment of the present application.

[0049] Icon: 100-electronic device; 10-waybill generation device; 11-data acquisition module; 12-image recognition module; 13-data clustering module; 14-waybill generation module; 20-memory; 30-processor; 40-communication unit. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0051] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application. In addition, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly specified and limited.

[0052] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected or can communicate with each other; it can be directly connected, or indirectly connected through an intermediate medium; it can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0053] In the present application, unless otherwise explicitly specified and limited, "on" or "under" of a first feature to a second feature can include that the first and second features are in direct contact, or can include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, "on", "above" and "over" of a first feature to a second feature includes that the first feature is directly above and obliquely above the second feature, or only indicates that the first feature is higher in horizontal height than the second feature. "Under", "below" and "underneath" of a first feature to a second feature includes that the first feature is directly below and obliquely below the second feature, or only indicates that the first feature is lower in horizontal height than the second feature.

[0054] The following disclosure provides many different embodiments, or examples, for implementing different structures of the application. For the purpose of simplification, the components and arrangements of the specific examples are described in the following. Of course, they are only examples, and the purpose is not to limit the application. In addition, the application can repeatedly refer to numbers and / or letters in different examples, and such repetition is for the purpose of simplification and clarity, which does not indicate the relationship between the various embodiments and / or arrangements discussed. In addition, the application provides examples of various specific processes and materials, but those skilled in the art can realize the application of other processes and / or the use of other materials.

[0055] As described in the background, the freight transportation of the logistics industry is mostly completed by the transport order and its transport times, and at present, the transport order of freight transportation mostly adopts offline manual recording or driver manual recording in related program software. Such mode still has great flaws in transport order recording, and cannot efficiently and low-costly calculate and count the real number of business and the real number of transport orders, so as to better perform freight settlement for the freight company.

[0056] Taking the clean-up of slag soil as an example, the clean-up of slag soil is an indispensable link in the process of urban construction, and is also the main source of income of the slag soil clean-up company. The slag soil clean-up company and the actual driver complete the freight settlement by the transport order and its transport times.

[0057] In the traditional accounting process of slag soil transport order, in order to ensure the authenticity and correctness of the transport order, the slag soil clean-up company often needs to invest a large amount of manpower and material resources to do accounting work, and such method has many disadvantages. Even after the emergence of the network freight platform, this part of work is transferred to the network freight platform, and there is no great improvement. Specifically as follows:

[0058] The common way of accounting for the transport order of the slag removal company is mainly offline manual record, which uses manual to issue paper vouchers to the driver at the loading and unloading point. This scheme confirms the driver's trip, which has the following disadvantages: first, the paper voucher is not easy to save and is not convenient to count, which is easy to miss and cause labor disputes; second, this scheme only provides a non-objective voucher for loading and unloading, and the evidence chain reported to the logistics supervision platform is incomplete and difficult to prove (the logistics supervision platform requires each transport order to be real, and is supplemented by running track and other real evidence, such as loading and unloading pictures and video images).

[0059] The common scheme of the general network freight platform is to rely on the driver to manually sign in at the loading and unloading site, and each trip requires the driver to manually operate the application (App) and other applications at the loading and unloading site to prove the authenticity and accuracy of the business. If the driver misses the operation, the evidence chain of the transport order will be incomplete. On the other hand, due to the age, cognition and other limitations of the driver, the operation of the App and other applications has a high learning cost, which also leads to the fact that this operation is always uncontrollable. The network freight platform has always been faced with the problem of a large amount of manpower, material and financial resources to ensure that there is no missed order and few missed orders to allow the business to operate normally.

[0060] Another scheme of the general network freight platform is to use radio frequency identification (RFID) card reader and electronic tag to solve the problem of driver operation difficulty and frequency statistics, that is, to place an RFID card reader at the entrance of each loading and unloading site, and to paste a unique electronic tag on the vehicle. When the vehicle enters and exits the loading and unloading entrance, the device automatically counts. This scheme is costly, requires each loading and unloading site to be equipped with an RFID device, and needs to be powered continuously, otherwise it cannot count. The business characteristics of slag itself determine that the loading and unloading sites are random, such as temporary suspension of construction site, completion of construction site transportation business, replacement of other sites due to reaching the carrying capacity of the unloading site, etc. In order to facilitate the treatment and landfill of slag and waste, the unloading site is often set up in the wild, where there is no power supply to support the operation of the RFID device. In particular, this scheme only obtains the triggering information of the loading and unloading site, and does not have any running track evidence information, and still has the problem of incomplete and difficult evidence chain reported to the logistics supervision platform.

[0061] That is, the existing technology still has a large number of omissions in the real transport order statistics, and cannot scientifically, efficiently and low-costly account for and count the real number of trips and the number of real transport orders, and cannot better support the freight settlement of freight companies and the business development of network freight platforms.

[0062] Based on the above research, this embodiment provides a waybill generation method, apparatus, electronic device, and readable storage medium. After acquiring stop point data and corresponding image data during vehicle travel, image recognition is performed on the image data to obtain at least one target stop point that meets the target scenario. By clustering each target stop point, at least one stop point cluster is obtained. After obtaining the stop point clusters, the vehicle's operating position can be determined based on the center and radius of each cluster. Then, a waybill for vehicle transportation is generated based on the operating position. In this way, by replacing manual recording of waybills, waybills can be generated efficiently and accurately, reducing labor costs.

[0063] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of the electronic device provided in this embodiment. Figure 1 As shown, the electronic device 100 includes a waybill generation device 10, a memory 20, a processor 30, and a communication unit 40. The memory 20, processor 30, and communication unit 40 are electrically connected directly or indirectly to each other to achieve signal transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0064] In this embodiment, the waybill generation device 10 includes at least one software functional module that can be stored in the memory 20 in the form of software or firmware. The processor 30 is used to execute the executable module (e.g., the software functional module or computer program included in the waybill generation device 10) stored in the memory 20. When the electronic device 100 is running, the processor 30 communicates with the memory 20 via a bus, and the processor 30 executes the executable module or computer program to implement the waybill generation method described in this embodiment.

[0065] The memory 20 may be, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.

[0066] The processor 30 is configured to perform one or more functions described in the embodiments. In some embodiments, the processor 30 can include one or more processing cores (e.g., a single-core processor (S) or a multi-core processor (S)). For example only, the processor 30 can include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a graphics processing unit (GPU), a physics processing unit (PPU), a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic device (PLD), a controller, a microcontroller unit, a reduced instruction set computer (RISC), or a microprocessor, or any combination thereof.

[0067] For ease of illustration, only one processor is described in the electronic device 100. However, it should be noted that the electronic device 100 in the embodiments can also include multiple processors, and thus the steps performed by one processor described in the embodiments can also be jointly performed by multiple processors or performed individually. For example, if the processor of the electronic device performs steps A and B, it should be understood that steps A and B can also be jointly performed by two different processors or individually performed in one processor. For example, the processor performs step A, the second processor performs step B, or the processor and the second processor jointly perform steps A and B.

[0068] In the embodiments, the method defined by the flow disclosed in any embodiment can be applied in the processor 30 or implemented by the processor 30.

[0069] The communication unit 40 is configured to establish a communication connection between the electronic device 100 and other devices through a network, and to transceive data through the network.

[0070] In some embodiments, the network can be any type of wired or wireless network, or a combination thereof. As examples only, the network can include a wired network, a wireless network, a fiber optic network, a telecommunications network, an intranet, the Internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a Public Switched Telephone Network (PSTN), a Bluetooth network, a ZigBee network, or a Near Field Communication (NFC) network, among others, or any combination thereof.

[0071] In this embodiment, the electronic device can be a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a physical server, or the like, and the specific type of the electronic device is not limited in this embodiment.

[0072] It can be understood that Figure 1 The structure shown is only schematic. The electronic device can also have more or fewer components than Figure 1 shown, or have a different configuration of components than Figure 1 shown. Figure 1 Each component shown can be implemented in hardware, software, or a combination thereof.

[0073] Based on the implementation architecture Figure 1 , this embodiment provides a waybill generation method, which is executed by the electronic device Figure 1 shown, and the steps of the waybill generation method provided in this embodiment are described in detail below. Please refer to Figure 2 , the waybill generation method provided in this embodiment includes steps S101 to S104.

[0074] Step S101: Obtain the stay point data and the image data corresponding to the stay point data in the vehicle driving process.

[0075] The positioning device and the camera device are arranged on each vehicle. The positioning device is configured to acquire the stop point data of the vehicle during the driving. The camera device is configured to acquire the image data of the vehicle during the driving. In this embodiment, the stop point data of the vehicle during the driving can be the stop point data of multiple vehicles.

[0076] Optionally, the positioning device can be a global positioning system (GPS). In this embodiment, the positioning device can acquire the trajectory data of the vehicle during the driving in real time. Correspondingly, the camera device can also capture the scene image of the vehicle during the driving in real time to obtain the image data of the vehicle during the driving. After obtaining the trajectory data and the image data of the vehicle during the driving, the stop points of the vehicle can be screened from the trajectory data based on the stop time and / or the driving speed to obtain the stop point data of the vehicle during the driving. Then, for each stop point in the stop point data, the scene image with the same time point as the time point of the stop point is selected as the scene image corresponding to the stop point, so as to obtain the image data corresponding to the stop point data of the vehicle during the driving.

[0077] Optionally, when the stop points of the vehicle are screened from the trajectory data based on the stop time and / or the driving speed, it can be determined whether the stop time of each trajectory point in the trajectory data is greater than a set stop time threshold. The trajectory point greater than the set stop time threshold is regarded as a stop point. Alternatively, it can be determined whether the driving speed of each trajectory point in the trajectory data is less than a set driving speed threshold. The trajectory point less than the set driving speed threshold is regarded as a stop point. Alternatively, the stop time and the driving speed can be combined to determine whether the stop time of each trajectory point in the trajectory data is greater than a set stop time threshold and whether the driving speed is less than a set driving speed threshold. Then, the trajectory point with the stop time greater than the set stop time threshold and the driving speed less than the set driving speed threshold is regarded as a stop point.

[0078] It can be understood that, in this embodiment, the stop point data of the vehicle includes the positions where the vehicle stops during the transportation and the time of stopping at each position. Optionally, the positions where the vehicle stops can be represented by latitude and longitude.

[0079] In this embodiment, the image data captured by the camera device represents the scene image of the environment where the vehicle is located. Whether the vehicle is working can be determined through the scene image of the environment where the vehicle is located.

[0080] Step S102: performing image recognition on the image data corresponding to the stop point data to obtain at least one target stop point satisfying the target scene.

[0081] Wherein, after obtaining the stay point data and the image data corresponding to the stay point data in the vehicle driving process, image recognition is performed on each scene image in the image data to obtain a scene corresponding to each scene image.

[0082] After obtaining the scene corresponding to each scene image, it is detected whether the scene corresponding to each scene image is a target scene, and then scene images corresponding to the target scene are selected, and the stay point corresponding to the selected scene image is taken as a target stay point, so as to obtain at least one target stay point meeting the target scene.

[0083] In the embodiment, the target scene is a scene of vehicle operation, including a scene of starting operation of the vehicle and a scene of ending operation of the vehicle. The scene of starting operation of the vehicle can be a scene of loading of the vehicle, and the scene of ending operation of the vehicle can be a scene of loading of the vehicle.

[0084] Step S103: Clustering the target stay points to obtain at least one stay point clustering cluster.

[0085] Wherein, after obtaining the target stay points, the target stay points are clustered to obtain the stay point clustering cluster.

[0086] Optionally, when clustering the target stay points, a density-based anti-noise clustering method (Density-Based Spatial Clustering of Applications with Noise, DBSCAN), a k-means algorithm (K-Means), a balanced iterative clustering method based on hierarchical structure (Balanced Iterative Reducing and Clustering using Hierarchies, BIRCH), or the like can be used for clustering. Specifically, the embodiment is not limited, and can be set according to actual needs.

[0087] Optionally, in the embodiment, the DBSCAN model is used to cluster the target stay points.

[0088] Step S104: Determining the operation position of the vehicle according to the center and the radius of each stay point clustering cluster, and generating a waybill of the vehicle transportation according to the operation position.

[0089] In this embodiment, each stay point cluster represents a set of target stay points with similar locations, so after obtaining each stay point cluster, the work center address corresponding to the stay point cluster can be determined according to the center of the stay point cluster, the work area radius corresponding to the stay point cluster can be determined according to the radius of the stay point cluster, then the work range corresponding to the stay point cluster is determined with the work center address as the center and the work area radius as the radius, the work position corresponding to the stay point cluster is determined as the work range obtained, and thus the work position of the vehicle is obtained. After obtaining the work position of the vehicle, the waybill of the vehicle transportation can be generated according to the work position.

[0090] The waybill generation method provided in this embodiment, after obtaining the stay point data and the image data corresponding to the stay point data in the driving process of the vehicle, performs image recognition on the image data corresponding to the stay point data to obtain at least one target stay point meeting the target scene, obtains at least one stay point cluster by clustering each target stay point, and then determines the work position of the vehicle according to the center and the radius of each stay point cluster after obtaining the stay point cluster, and generates the waybill of the vehicle transportation according to the work position. The way of determining the work position by manual or simple equipment and then counting the number of waybills is replaced by the way of automatically generating the waybill by relying on software algorithm to recognize the work position, which reduces the operation cost of the waybill, and saves manpower, material and financial resources.

[0091] In view of the fact that the vehicle may have a short stay due to some external factors such as road congestion during driving, or stay on a road section such as a highway that obviously does not meet the work conditions. In order to improve the accuracy of the data and avoid the influence of these stay data on the subsequent identification of the work position, in this embodiment, please refer to Figure 3 , the step of obtaining the stay point data and the image data corresponding to the stay point data in the driving process of the vehicle can include steps S201 to S203.

[0092] Step S201: Obtain the trajectory data of the vehicle driving.

[0093] Step S202: Filter the trajectory data according to the set speed condition, time condition and road condition to obtain stay point data meeting the speed condition, time condition and road condition.

[0094] Step S203: For each stay point in the stay point data, obtain the scene image corresponding to the stay point to obtain the image data corresponding to the stay point data.

[0095] The trajectory data of the vehicle includes trajectory points, speed, date, time, and device number of the collection device of the vehicle, wherein the trajectory points of the vehicle can be represented by longitude and latitude, the trajectory points, date, and time of the vehicle can be obtained by the positioning device, the speed of the vehicle can be obtained by the speed sensor arranged on the vehicle, and the device number of the collection device represents the device number of the positioning device, and each positioning device has a unique device number. In the embodiment, the trajectory data of the vehicle can be represented in the following manner.

[0096]

[0097] After obtaining the trajectory data of the vehicle, the obtained trajectory data can be subjected to road rectification service to obtain real, undisturbed, and unbiased trajectory points and road grade information corresponding to the trajectory points.

[0098] After the trajectory data is subjected to road rectification service, the trajectory data subjected to the rectification service can be filtered according to the set speed condition, time condition, and road condition to obtain the stay point data satisfying the speed condition, time condition, and road condition.

[0099] In the embodiment, the speed condition can be to detect whether the speed is less than a first set threshold, the time condition can be to detect whether the stay time is greater than a second set threshold, and the road condition can be to detect whether the road grade where the trajectory point is located is a set target grade. Therefore, when the trajectory data is filtered according to the speed condition, time condition, and road condition, whether the speed corresponding to each trajectory point in the trajectory data is less than the first set threshold, whether the stay time is greater than the second set threshold, and whether the road grade where the trajectory point is located is the set target grade can be detected.

[0100] In the embodiment, when the trajectory data is filtered according to the speed condition, time condition, and road condition, for each trajectory point in the trajectory data, only when the trajectory point satisfies all the three conditions, it is determined that the trajectory point belongs to the stay point, and if one of the conditions is not satisfied, it is determined that the trajectory point does not belong to the stay point. That is, in the embodiment, for each trajectory point in the trajectory data, if the speed corresponding to the trajectory point is less than the first set threshold, the stay time is greater than the second set threshold, and the road grade where the trajectory point is located is not the set target grade, it is determined that the trajectory point belongs to the stay point, and if the speed corresponding to the trajectory point is not less than the first set threshold, or the stay time is less than the second set threshold, or the road grade where the trajectory point is located is the set target grade, it is determined that the trajectory point does not belong to the stay point, and the trajectory point is excluded.

[0101] Optionally, in the embodiment, the first set threshold can be 10 m / s, the second set threshold can be 2 minutes, and the set target grade is the grade of the expressway section.

[0102] After the trajectory data is filtered by the set speed condition, time condition and road condition, the stay point data satisfying the speed condition, time condition and road condition is obtained. After the stay point data satisfying the speed condition, time condition and road condition is obtained, for each stay point in the stay point data, a scene image with the same time information is searched according to the time information (i.e. time stamp) corresponding to the stay point. After the scene image with the same time information is searched, the searched scene image is taken as the scene image corresponding to the stay point. In this way, the image data corresponding to the stay point data is obtained.

[0103] After the stay point data in the vehicle driving process and the image data corresponding to the stay point data are obtained, the image data corresponding to the stay point data is subjected to image recognition to obtain a target stay point satisfying a target scene.

[0104] In order to improve the accuracy of image recognition, in the embodiment, the step of subjecting the image data corresponding to the stay point data to image recognition to obtain at least one target stay point satisfying a target scene can include:

[0105] The scene features of each scene image in the image data are extracted by a pre-set image recognition model, and each scene feature is analyzed and processed to obtain the scene category to which each scene image belongs.

[0106] For each scene image, if the scene category to which the scene image belongs is a target scene, the stay point corresponding to the scene image is set as a target stay point.

[0107] The image recognition model can be obtained by training a neural network. Optionally, in the embodiment, the training of the neural network can adopt a supervised training manner, i.e. an optimal model is obtained by training sample data with known true values, and then all inputs are mapped to corresponding outputs by using the model, and the outputs are judged to achieve the purpose of classification, and the obtained model has the ability to classify unknown data. Specifically, a plurality of scene images can be obtained first, and then the scene images are subjected to scene labeling to obtain sample data, and then the sample data is input to the neural network for training, and the parameters of the neural network are adjusted according to the difference between the output value of the neural network and the labeled value, and the optimal model is obtained by iteration, and the optimal model is taken as the image recognition model.

[0108] Optionally, in the embodiment, the neural network can be a convolutional neural network (CNN), or a recurrent neural network (RNN), etc., which can be set according to actual requirements, and the embodiment does not make specific limitation.

[0109] After obtaining the image recognition model, the image recognition model can be used to extract features of each scene image in the image data, to obtain scene features of each scene image, and then analyze and process the scene features of each scene image to obtain a scene category to which each scene image belongs.

[0110] After obtaining the scene category to which each scene image belongs, for each scene image, it is detected whether the scene category to which the scene image belongs is a target scene. If the scene category to which the scene image belongs is the target scene, the stay point corresponding to the scene image is set as a target stay point, indicating that the stay point corresponding to the scene image is a track point at which the work is performed. If the scene category to which the scene image belongs is not the target scene, it indicates that the stay point corresponding to the scene image does not belong to the track point at which the work is performed, i.e., the stay point corresponding to the scene image is not the target stay point.

[0111] It should be noted that, in the embodiment, the target scene includes a scene in which the vehicle starts work and a scene in which the vehicle ends work. The scene in which the vehicle starts work can be a scene in which the vehicle loads goods, and the scene in which the vehicle ends work can be a scene in which the vehicle unloads goods. For each scene image, after the image recognition model extracts the scene features in the scene image, it can analyze whether the extracted scene features are loading scene features or unloading scene features. If the extracted scene features are loading scene features, it is determined that the scene category corresponding to the scene image is a loading scene, i.e., a starting work scene. If the extracted scene features are unloading scene features, it is determined that the scene category corresponding to the scene image is an unloading scene, i.e., an ending work scene. If the extracted scene features are neither loading scene features nor unloading scene features, it is determined that the scene category corresponding to the scene image is another category.

[0112] For example, for the vehicle for transporting the slag, the starting work scene, i.e., the loading scene, has the following characteristics: 1. mostly on-site construction or on the dirt road; 2. containing large construction equipment such as bulldozers, excavators, and cranes; 3. having a construction site gate or a water trough mark; and 4. containing a fence or a green fence. For the ending work scene, i.e., the unloading scene, has the following characteristics: 1. in a less populated area of the dirt road or the gravel road; 2. containing a road roller; and 3. the vehicle box is inclined. Based on this, after the image recognition model extracts the scene features in the scene image, when analyzing whether the extracted scene features are the scene features of the loading scene, it can be analyzed whether the extracted scene features have any one of the characteristic features in the loading scene. If so, it is determined that the extracted scene features are the scene features of the loading scene. If not, it is determined that the extracted scene features are not the scene features of the loading scene.

[0113] Correspondingly, when analyzing whether the extracted scene features are the scene features of the unloading scene, it can be analyzed whether the extracted scene features have any one of the characteristic features in the unloading scene. If so, it is determined that the extracted scene features are the scene features of the unloading scene. If not, it is determined that the extracted scene features are not the scene features of the unloading scene.

[0114] Optionally, in order to improve the accuracy of image recognition, in this embodiment, a plurality of image recognition models can be trained according to different scenes, and then image recognition is performed based on the plurality of image recognition models to obtain the scene category to which the scene image belongs. For example, when identifying the scene category of the scene image, the image recognition model corresponding to the starting work scene can be used to identify whether the scene image has the scene features of the loading scene, and the image recognition model corresponding to the ending work scene can be used to identify whether the scene image has the scene features of the unloading scene, so as to obtain the scene category to which the scene image belongs.

[0115] The waybill generation method provided in this embodiment extracts the scene features of each scene image in the image data through the preset image recognition model, analyzes and processes each scene feature, obtains the scene category to which each scene image belongs, and then determines the target stop point in the stop point data according to the scene category to which each scene image belongs. In this way, the accuracy of the data can be effectively improved.

[0116] After the target stop point is determined based on the scene category to which the scene image belongs, the work position of the vehicle can be determined according to the target stop point.

[0117] In this embodiment, when the work position of the vehicle is determined according to the target stop point, each target stop point can be clustered to obtain at least one stop point cluster, and then the work position of the vehicle is determined according to the center and the radius of each stop point cluster.

[0118] Since the waybill of the vehicle transportation is generated based on the starting position and the ending position of the operation, in order to accurately generate the waybill, in the embodiment, the step of clustering each target stop point to obtain the stop point clustering cluster can include:

[0119] According to the scene corresponding to each target stop point, the target stop points are grouped to obtain the starting operation grouping and the ending operation grouping.

[0120] The target stop points in the starting operation grouping and the target stop points in the ending operation grouping are respectively clustered to obtain the stop point clustering cluster corresponding to the starting operation grouping and the stop point clustering cluster corresponding to the ending operation grouping.

[0121] According to the scene corresponding to each target stop point, the target stop points are grouped, for each target stop point, if the scene corresponding to the target stop point is the scene of the starting operation, the target stop point is divided into the starting operation grouping, and if the scene corresponding to the target stop point is the scene of the ending operation, the target stop point is divided into the ending operation grouping.

[0122] After the target stop points are grouped to obtain the starting operation grouping and the ending operation grouping, the target stop points in the starting operation grouping and the target stop points in the ending operation grouping are respectively clustered to obtain the stop point clustering cluster corresponding to the starting operation grouping and the stop point clustering cluster corresponding to the ending operation grouping.

[0123] In view of the application, when transporting, multiple vehicles can be used at the same time, and since the loading and unloading locations of the transported goods are mostly the same, that is, the starting operation position and the ending operation position of the vehicle are mostly the same, in order to accurately identify the operation position of the vehicle, please refer to Figure 4 , the steps of clustering the target stop points included in the starting operation grouping and the target stop points included in the ending operation grouping to obtain the stop point clustering cluster corresponding to the starting operation grouping and the stop point clustering cluster corresponding to the ending operation grouping can include steps S301 to S303.

[0124] Step S301: According to the date corresponding to each target stop point and the collection device, the target stop points included in the starting operation grouping are grouped to obtain at least one first combination, and the target stop points included in the ending operation grouping are grouped to obtain at least one second combination.

[0125] Step S302: According to the position of each target stop point, each first combination and each second combination are respectively clustered to obtain the clustering result of each first combination and the clustering result of each second combination.

[0126] Step S303: clustering the clustering results of each first combination to obtain the stay point clustering cluster corresponding to the start work grouping, and clustering the clustering results of each second combination to obtain the stay point clustering cluster corresponding to the end work grouping.

[0127] Since each vehicle can generate multiple stay points per day, and each vehicle is equipped with a unique collection device, in the embodiment, the target stay points are grouped according to the dates corresponding to the target stay points and the collection devices, so that the stay point data of each vehicle per day can be obtained.

[0128] Based on this, in the embodiment, the target stay points included in the start work grouping are grouped according to the dates corresponding to the target stay points and the collection devices, so that each first combination obtained represents the stay point data of the start work of each vehicle per day. Correspondingly, the target stay points included in the end work grouping are grouped according to the dates corresponding to the target stay points and the collection devices, so that each second combination obtained represents the stay point data of the end work of each vehicle per day.

[0129] After obtaining the first combination and the second combination, for each first combination, the first combination can be clustered according to the positions of the target stay points in the first combination, so that the clustering result of the first combination is obtained; for each second combination, the second combination can be clustered according to the positions of the target stay points in the second combination, so that the clustering result of the second combination is obtained.

[0130] Each vehicle can have multiple stay points of start work and stay points of end work per day, so in the embodiment, the clustering result of each first combination includes a stay point clustering set of start work of each vehicle per day. Correspondingly, the clustering result of each second combination includes a stay point clustering set of end work of each vehicle per day, and each stay point set includes multiple stay points close in position.

[0131] After obtaining the clustering result of each first combination, clustering the clustering results of each first combination means clustering the stay point clustering sets of start work of multiple vehicles per day again. Correspondingly, after obtaining the clustering result of each second combination, clustering the clustering results of each second combination means clustering the stay point clustering sets of end work of multiple vehicles per day again.

[0132] In the embodiment, when the stay point clustering sets of start work of multiple vehicles per day are clustered again, the center of each stay point clustering set can be taken as a representative point, and then all the representative points taken out are clustered again. Correspondingly, when the stay point clustering sets of end work of multiple vehicles per day are clustered again, the center of each stay point clustering set can also be taken as a representative point, and then all the representative points taken out are clustered again.

[0133] For example, the clustering result of the first combination A includes an a stay point cluster set and a b stay point cluster set, the clustering result of the first combination B includes a c stay point cluster set and a d stay point cluster set, and when the clustering result of each first combination is clustered again, the center of the a stay point cluster set, the center of the b stay point cluster set, the center of the c stay point cluster set and the center of the d stay point cluster set can be taken as representative points, and then the center of the a stay point cluster set, the center of the b stay point cluster set, the center of the c stay point cluster set and the center of the d stay point cluster set are clustered again.

[0134] The embodiment can effectively improve the accuracy by first analyzing the work stay points of a single vehicle in a single day, and then determining the stay point cluster by combining the work stay points of multiple vehicles in multiple days.

[0135] After clustering the clustering result of each first combination, the stay point cluster corresponding to the start work grouping is obtained, and after clustering the clustering result of each second combination, the stay point cluster corresponding to the end work grouping is obtained. Understandably, each stay point cluster corresponding to the start work grouping represents a stay point set of a start work, and each stay point cluster corresponding to the end work grouping represents a stay point set of an end work. A stay point set includes multiple stay points with similar positions.

[0136] After obtaining the stay point cluster corresponding to the start work grouping and the stay point cluster corresponding to the end work grouping, the start work position of the vehicle is determined based on the stay point cluster corresponding to the start work grouping, and the end work position of the vehicle is determined based on the stay point cluster corresponding to the end work grouping.

[0137] In actual applications, the start work scene and the end work scene of the vehicle may have similar places, which may cause the scene category of the image to be identified as both the start work scene and the end work scene during image recognition, so that the target stay point may be divided into both the start work grouping and the end work grouping during grouping of the target stay point, resulting in errors during determination of the work position. Therefore, in order to further improve the accuracy, after obtaining the stay point cluster corresponding to the start work grouping and the stay point cluster corresponding to the end work grouping, the waybill generation method can further include the following steps in the embodiment:

[0138] Detecting whether there is a target stay point cluster with the same center as the center of the stay point cluster corresponding to the start work grouping in the stay point cluster corresponding to the end work grouping.

[0139] If the target stay point cluster exists, the target stay point cluster is removed from the stay point cluster corresponding to the end job grouping.

[0140] In this embodiment, for the detected stay point cluster with the same center, the stay point cluster can be considered as the starting job position. Therefore, when detecting whether the target stay point cluster with the same center as the starting job grouping exists in the stay point cluster corresponding to the end job grouping, for each stay point cluster corresponding to the starting job grouping, the center of the stay point cluster can be compared with the center of each stay point cluster corresponding to the end job grouping. If the target stay point cluster with the same center as the stay point cluster exists in the end job grouping, the target stay point cluster is removed from the stay point cluster corresponding to the end job grouping.

[0141] It can be understood that if the detected stay point cluster with the same center is considered as the end job position, the stay point cluster is removed from the stay point cluster corresponding to the starting job grouping.

[0142] In this embodiment, after obtaining the stay point cluster corresponding to the starting job grouping and the stay point cluster corresponding to the end job grouping, the stay point cluster corresponding to the starting job grouping and the stay point cluster corresponding to the end job grouping are compared, and then the repeatedly divided stay points are removed, so as to improve the accuracy of the job position identification.

[0143] After the comparison and screening processing of the stay point cluster corresponding to the starting job grouping and the stay point cluster corresponding to the end job grouping, the starting job position of the vehicle can be determined according to the processed stay point cluster corresponding to the starting job grouping, and the end job position of the vehicle can be determined according to the processed stay point cluster corresponding to the end job grouping.

[0144] Based on this, in this embodiment, the steps of determining the starting job position of the vehicle according to the processed stay point cluster corresponding to the starting job grouping and determining the end job position of the vehicle according to the processed stay point cluster corresponding to the end job grouping can include:

[0145] The starting job position of the vehicle is determined according to the center and the radius of the stay point cluster corresponding to the starting job grouping.

[0146] The end job position of the vehicle is determined according to the center and the radius of the stay point cluster corresponding to the end job grouping.

[0147] Correspondingly, the step of generating the waybill of the vehicle transportation according to the job position includes:

[0148] According to the starting work position and the ending work position of the vehicle, a waybill of the vehicle transportation is generated.

[0149] According to the center and the radius of the stay point cluster corresponding to the starting work group, when the starting work position of the vehicle is determined, for each stay point cluster corresponding to the starting work group, the work center address corresponding to the stay point cluster is determined according to the center of the stay point cluster, the work area radius corresponding to the stay point cluster is determined according to the radius of the stay point cluster, then the work range corresponding to the stay point cluster is determined with the work center address as the center and the work area radius as the radius, and the determined work range is taken as the starting work position corresponding to the stay point cluster.

[0150] Accordingly, according to the center and the radius of the stay point cluster corresponding to the ending work group, when the ending work position of the vehicle is determined, for each stay point cluster corresponding to the ending work group, the work center address corresponding to the stay point cluster is determined according to the center of the stay point cluster, the work area radius corresponding to the stay point cluster is determined according to the radius of the stay point cluster, then the work range corresponding to the stay point cluster is determined with the work center address as the center and the work area radius as the radius, and the determined work range is taken as the ending work position corresponding to the stay point cluster.

[0151] After the starting work position and the ending work position are determined, the waybill of the vehicle transportation can be generated according to the starting work position and the ending work position.

[0152] In view of the fact that the vehicle will also stay at positions such as gas stations and parking lots during the transportation of goods, and these positions will interfere with the identification of the work position of the vehicle and affect its accuracy, after the work position of the vehicle is determined according to the center and the radius of each stay point cluster, the waybill generation method provided in the embodiment can further include the following steps:

[0153] According to the set interference position, it is detected whether there is a target position same as the interference position in the work position of the vehicle.

[0154] If there is a target position, the target position is removed.

[0155] The set interference position can be a position of an interference site such as a gas station or a parking lot. In this embodiment, the interference position can be obtained in advance, and then, after the working position of the vehicle is determined, the interference position is compared with the determined working position one by one to detect whether the determined working position has a target position identical to the interference position. If the target position identical to the interference position exists, the target position is removed from the determined working position. If the target position identical to the interference position does not exist, the detection result of the determined working position is retained.

[0156] In an optional embodiment, the scene image corresponding to each working position can also be identified based on image recognition to determine whether the scene image has feature information of the interference position. For example, for the interference position such as a gas station or a parking lot, the feature information of the corresponding scene image can be a text identifier such as a gas station or a parking lot. When it is detected that the scene image corresponding to the working position has the feature information of the interference position, it is indicated that the working position is the interference position, and the working position is removed.

[0157] The waybill generation method provided in this embodiment can improve the accuracy of detection of the working position through detection and filtering of the interference position.

[0158] In this embodiment, after the interference position in the working position is screened and filtered, the waybill for vehicle transportation can be generated according to the starting working position and the ending working position in the working position.

[0159] Since the starting working position and the ending working position are obtained based on the stop points of multiple vehicles, for a single vehicle, when the order for transportation of the vehicle is generated, the stop point of the vehicle can be determined first, and then the stop point of the vehicle is matched with the determined starting working position and ending working position to obtain the starting working position and the ending working position of the vehicle. After the starting working position and the ending working position of the vehicle are obtained, the waybill trajectory can be divided according to the starting working position and the ending working position of the vehicle, and the waybill can be generated based on the division result.

[0160] When the waybill trajectory is divided according to the starting working position and the ending working position of the vehicle, every two adjacent starting working position and ending working position of the vehicle are taken as one waybill trajectory for division. For example, for a vehicle, the determined working positions are starting working position A, starting working position B, ending working position C, and starting working position D in sequence, and then, according to the starting working position B and the ending working position C, one waybill trajectory is divided, and then the waybill is generated based on the starting working position B and the ending working position C.

[0161] In actual applications, the driver of the vehicle will stay for a long time due to some factors when the vehicle is transporting goods. For example, when a vehicle is transporting goods, it arrives at a delivery location, the driver rests at the delivery location without unloading, and then departs from the delivery location to a final delivery location for unloading. In this case, when the delivery track is divided, the delivery location where the driver rests is easily regarded as a stop location of the vehicle operation, which leads to an error in the division of the delivery track of the vehicle and reduces the accuracy of the generated delivery order.

[0162] Since the duration of a delivery order of the vehicle is generally within a fixed duration range when the vehicle is transporting goods, that is, the duration of the vehicle from the start operation position to the end operation position is generally within a fixed duration range. Based on this, in order to improve the accuracy of the generated delivery order, in this embodiment, when the delivery order of the vehicle is generated according to the start operation position and the end operation position, the determined stop points of the vehicle can be sorted in time sequence first, and then the stop points of the vehicle are divided based on a set time threshold to obtain the stop segments of the vehicle. For example, the stop points of the vehicle are sorted in time sequence as A1, A2, A3, A4, and A5, wherein the duration between the stop points A3 and A4 is greater than the set time threshold, and then A1 to A3 are regarded as a stop segment and A4 to A5 are regarded as a stop segment based on the stop points A3 and A4.

[0163] After obtaining the stop segments of the vehicle, the stop points included in each stop segment are matched with the determined start operation position and end operation position to obtain the target start operation position and the target end operation position in the stop segment, and then the delivery track is divided according to the target start operation position and the target end operation position, and the delivery order is generated based on the division result. It can be understood that when the delivery track is divided according to the target start operation position and the target end operation position, each two adjacent target start operation positions and target end operation positions are regarded as a delivery track for division.

[0164] Optionally, the set time threshold can be set according to actual needs, and the specific limitation is not made. Optionally, in this embodiment, the set time threshold can be 5 hours, and when the duration between two adjacent stop points of the vehicle is greater than 5 hours, the two stop points can be divided.

[0165] The waybill generation method provided in this embodiment automatically identifies the operation location through scene recognition and in-depth mining of large-scale vehicle data, with high accuracy. At the same time, it generates transportation waybills based on the identified operation location, which is both efficient and accurate. Furthermore, it can be corroborated with images and videos of the entire vehicle transportation process to ensure the authenticity of the business waybills.

[0166] The waybill generation method provided in this embodiment, after acquiring stop point data and corresponding image data during vehicle travel, performs image recognition on the image data corresponding to the stop point data to obtain at least one target stop point that meets the target scenario. By clustering each target stop point, at least one stop point cluster is obtained. After obtaining the stop point clusters, the vehicle's operating location can be determined based on the center and radius of each stop point cluster. Then, a waybill for vehicle transportation is generated based on the operating location. In this way, by replacing the manual recording of waybills, waybills can be generated efficiently and accurately, reducing labor and material costs.

[0167] Based on the same inventive concept, please refer to the following: Figure 5 This embodiment also provides a waybill generation device 10, which is applied to... Figure 1 The electronic device shown. For example... Figure 5 As shown, the waybill generation device 10 may include a data acquisition module 11, an image recognition module 12, a data clustering module 13, and a waybill generation module 14.

[0168] The data acquisition module 11 is used to acquire stop point data during the vehicle's journey and the corresponding image data.

[0169] The image recognition module 12 is used to perform image recognition on the image data corresponding to the stop point data to obtain at least one target stop point that satisfies the target scene.

[0170] The data clustering module 13 is used to cluster each target stop point to obtain at least one stop point cluster.

[0171] The waybill generation module 14 is used to determine the working location of the vehicle based on the center and radius of each stop point cluster, and generate a waybill for vehicle transportation based on the working location.

[0172] In an optional implementation, the image recognition module 12 is used for:

[0173] The scene features of each scene image in the image data are extracted by a preset image recognition model, and the scene features are analyzed and processed to obtain the scene category to which each scene image belongs; each scene image corresponds to a stop point, and the stop point data includes at least one stop point.

[0174] For each scene image, if a scene category to which the scene image belongs is a target scene, a stay point corresponding to the scene image is set as a target stay point.

[0175] In an optional implementation, the image recognition module 12 is configured to:

[0176] The target stay points are grouped according to scenes corresponding to the target stay points, to obtain a start work grouping and an end work grouping.

[0177] The target stay points in the start work grouping and the target stay points in the end work grouping are respectively clustered to obtain stay point clustering clusters corresponding to the start work grouping and stay point clustering clusters corresponding to the end work grouping.

[0178] In an optional implementation, the image recognition module 12 is configured to:

[0179] The target stay points included in the start work grouping are grouped according to dates corresponding to the target stay points and the collection device, to obtain at least one first combination, and the target stay points included in the end work grouping are grouped according to dates corresponding to the target stay points and the collection device, to obtain at least one second combination.

[0180] The first combinations and the second combinations are respectively clustered according to positions of the target stay points, to obtain clustering results of the first combinations and clustering results of the second combinations.

[0181] The clustering results of the first combinations are clustered to obtain stay point clustering clusters corresponding to the start work grouping, and the clustering results of the second combinations are clustered to obtain stay point clustering clusters corresponding to the end work grouping.

[0182] In an optional implementation, the image recognition module 12 is configured to:

[0183] The start work position of the vehicle is determined according to a center and a radius of the stay point clustering cluster corresponding to the start work grouping.

[0184] The end work position of the vehicle is determined according to a center and a radius of the stay point clustering cluster corresponding to the end work grouping.

[0185] The waybill generation module 14 is configured to:

[0186] The waybill of the vehicle transportation is generated according to the start work position and the end work position of the vehicle.

[0187] In an optional implementation, after the stay point clustering clusters corresponding to the start work grouping and the stay point clustering clusters corresponding to the end work grouping are obtained, the waybill generation module 14 is configured to:

[0188] determine whether there is a target stay point cluster identical to the center of the stay point cluster corresponding to the starting job package in the stay point cluster corresponding to the ending job package.

[0189] If the target stay point cluster exists, remove the target stay point cluster from the stay point cluster corresponding to the ending job package.

[0190] In an optional implementation, the data acquisition module 11 is configured to:

[0191] acquire trajectory data of the vehicle.

[0192] filter the trajectory data according to the set speed condition, time condition and road condition, and obtain stay point data satisfying the speed condition, time condition and road condition.

[0193] For each stay point in the stay point data, acquire a scene image corresponding to the stay point, and obtain image data corresponding to the stay point data.

[0194] In an optional implementation, the waybill generation module 14 is configured to:

[0195] According to the set interference position, detect whether there is a target position identical to the interference position in the job position of the vehicle.

[0196] If the target position exists, remove the target position.

[0197] The waybill generation device provided in the embodiment, after acquiring the stay point data and the image data corresponding to the stay point data in the driving process of the vehicle, performs image recognition on the image data corresponding to the stay point data, obtains at least one target stay point satisfying a target scene, performs clustering on each target stay point to obtain at least one stay point cluster, and then determines the job position of the vehicle according to the center and radius of each stay point cluster. Then, according to the job position, a waybill for the vehicle transportation is generated. In this way, the waybill is efficiently and truly generated by replacing the manual recording of the waybill, and the labor cost is reduced.

[0198] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the foregoing method, which will not be described in more detail here.

[0199] On the basis of the foregoing, the embodiment further provides a readable storage medium, wherein the readable storage medium stores a computer program, and the computer program is executed by a processor to implement the waybill generation method of any one of the foregoing embodiments.

[0200] The readable storage medium can be, but is not limited to, a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various storage program code storage media.

[0201] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the readable storage medium described above can refer to the corresponding process in the foregoing method, and will not be described in detail here.

[0202] To sum up, the waybill generation method, device, electronic equipment and readable storage medium provided by the embodiment of the present application can obtain the stay point data in the driving process of the vehicle and the image data corresponding to the stay point data, perform image recognition on the image data corresponding to the stay point data, obtain at least one target stay point meeting the target scene, obtain at least one stay point clustering cluster by clustering the target stay points, and then determine the working position of the vehicle according to the center and radius of each stay point clustering cluster. Then, the waybill of the vehicle transportation can be generated according to the working position. In this way, the waybill can be efficiently and truly generated by replacing the manual waybill recording mode, and the labor cost is reduced.

[0203] The above describes a waybill generation method, device, electronic equipment and readable storage medium provided by the embodiment of the present application in detail. The principle and implementation mode of the present application are described by applying specific examples in this paper. The above embodiment is only used to help understand the technical solution and core idea of the present application. Those skilled in the art should understand that the technical solution recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents. The modification or replacement does not change the essence of the corresponding technical solution from the scope of the technical solution of the embodiments of the present application.

Claims

1. A waybill generation method characterized by, The method comprises: acquiring stop point data in a vehicle driving process and image data corresponding to the stop point data; performing image recognition on the image data corresponding to the stop point data to obtain at least one target stop point meeting a target scene; clustering each target stop point to obtain at least one stop point cluster; determining a work position of the vehicle according to a center and a radius of each stop point cluster, and generating a waybill of vehicle transportation according to the work position. The clustering of each target stop point to obtain at least one stop point cluster comprises: according to a scene corresponding to each target stop point, for each target stop point, if the scene corresponding to the target stop point is a starting work scene, the target stop point is divided into a starting work group; if the scene corresponding to the target stop point is an ending work scene, the target stop point is divided into an ending work group, to obtain a starting work group and an ending work group; clustering target stop points in the starting work group and target stop points in the ending work group respectively to obtain a stop point cluster corresponding to the starting work group and a stop point cluster corresponding to the ending work group; the determination of the work position of the vehicle according to the center of each stop point cluster comprises: determining a starting work position of the vehicle according to the center and the radius of the stop point cluster corresponding to the starting work group; determining an ending work position of the vehicle according to the center and the radius of the stop point cluster corresponding to the ending work group; the generation of the waybill of vehicle transportation according to the work position comprises: generating the waybill of vehicle transportation according to the starting work position and the ending work position of the vehicle.

2. The waybill generation method of claim 1, wherein, The image recognition of the image data corresponding to the stop point data to obtain at least one target stop point meeting a target scene comprises: extracting scene features of each scene image in the image data through a preset image recognition model, and analyzing and processing each scene feature to obtain a scene category to which each scene image belongs; each scene image corresponds to a stop point, and the stop point data comprises at least one stop point; for each scene image, if the scene category to which the scene image belongs is a target scene, a stop point corresponding to the scene image is set as a target stop point.

3. The waybill generation method of claim 1, wherein, The clustering of target stop points included in the starting work group and target stop points included in the ending work group to obtain a stop point cluster corresponding to the starting work group and a stop point cluster corresponding to the ending work group comprises: grouping target stop points included in the starting work group according to dates corresponding to each target stop point and a collection device to obtain at least one first combination, and grouping target stop points included in the ending work group to obtain at least one second combination; clustering each first combination and each second combination according to positions of each target stop point to obtain a clustering result of each first combination and a clustering result of each second combination; The clustering results of each of the first combinations are clustered to obtain a stay point cluster corresponding to the starting job grouping, and the clustering results of each of the second combinations are clustered to obtain a stay point cluster corresponding to the ending job grouping.

4. The waybill generation method of claim 1, wherein, After the stay point cluster corresponding to the starting job grouping and the stay point cluster corresponding to the ending job grouping are obtained, the method further comprises: detecting whether there is a target stay point cluster with the same center as the stay point cluster corresponding to the starting job grouping in the stay point cluster corresponding to the ending job grouping; if the target stay point cluster exists, removing the target stay point cluster from the stay point cluster corresponding to the ending job grouping.

5. The waybill generation method of claim 1, wherein, The obtaining of the stay point data in the vehicle driving process and the image data corresponding to the stay point data comprises: obtaining trajectory data of the vehicle driving; screening the trajectory data according to a set speed condition, a time condition and a road condition to obtain stay point data satisfying the speed condition, the time condition and the road condition; for each stay point in the stay point data, obtaining a scene image corresponding to the stay point to obtain image data corresponding to the stay point data.

6. The waybill generation method of claim 1, wherein, After the work position of the vehicle is determined according to the center and the radius of each of the stay point clusters, the method further comprises: detecting whether there is a target position identical to the interference position in the work position of the vehicle according to a set interference position; if the target position exists, removing the target position.

7. A waybill generation apparatus characterized by comprising: The waybill generation device comprises: a data acquisition module configured to acquire stay point data in a vehicle driving process and image data corresponding to the stay point data; an image recognition module configured to perform image recognition on the image data corresponding to the stay point data to obtain at least one target stay point satisfying a target scene; a data clustering module configured to, for each target stay point, if a scene corresponding to the target stay point is a starting job scene, divide the target stay point into a starting job grouping; if the scene corresponding to the target stay point is an ending job scene, divide the target stay point into an ending job grouping, to obtain a starting job grouping and an ending job grouping; and cluster the target stay points in the starting job grouping and the target stay points in the ending job grouping respectively to obtain a stay point cluster corresponding to the starting job grouping and a stay point cluster corresponding to the ending job grouping; a waybill generation module configured to determine a work position of the vehicle according to the center and the radius of each of the stay point clusters, and generate a waybill for vehicle transportation according to the work position; wherein the waybill generation module is specifically configured to determine a starting work position of the vehicle according to the center and the radius of the stay point cluster corresponding to the starting job grouping, and determine an ending work position of the vehicle according to the center and the radius of the stay point cluster corresponding to the ending job grouping. The waybill generation module is specifically configured to generate a waybill for vehicle transportation according to a starting work position and an ending work position of the vehicle.

8. An electronic device, comprising: The computer program is stored in the memory and executable on the processor, and the processor executes the program to implement the waybill generation method in any one of claims 1 to 6.

9. A readable storage medium, characterized by, The readable storage medium comprises a computer program, and the computer program controls an electronic device where the readable storage medium is located to execute the waybill generation method in any one of claims 1 to 6 when running.

Images