A license plate number completion method and device

Abstract

The present application discloses a license plate number completion method and device. For a vehicle to be completed with a missing license plate number detected, candidate completion vehicles are determined from a historical driving record dataset; each candidate completion vehicle is a vehicle with the same vehicle information as the vehicle to be completed and a known license plate number; based on each candidate completion vehicle, the license plate number of the vehicle to be completed is completed. In this way, based on the technology of big data analysis, the license plate number of a vehicle with abnormal recognition is completed with the license plate number that has been accurately recognized, improving the detection rate of license plate numbers and greatly enhancing the accuracy of the detected license plate numbers at the same time.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of intelligent transportation, and in particular, to a method and device for completing license plate numbers. Background Art

[0002] At present, when the traffic department determines whether there are any violations committed by each vehicle traveling on the road surface, it can only take targeted actions by accurately obtaining the license plate numbers of each vehicle. Among them, license plate recognition technology is a technology that, after processing vehicle images or video sequences captured by a camera through algorithms such as machine vision, image processing, and pattern recognition, can automatically read various license plate information such as license plate numbers, license plate types, and license plate colors.

[0003] However, in actual application scenarios, since the road checkpoints and video structured cameras in urban construction were built in different periods, the types and resolutions of the cameras are different. Even by adjusting the camera parameters or training algorithms, it is very difficult to cover all scenarios, which makes the accuracy of the recognized license plate numbers not high when using license plate recognition technology to recognize license plate numbers.

[0004] Therefore, there is an urgent need for a method that can accurately determine license plate numbers at present. Summary of the Invention

[0005] The present application provides a method and device for completing license plate numbers, which are used to accurately determine the license plate numbers of vehicles with missing license plate numbers.

[0006] In a first aspect, an embodiment of the present application provides a method for completing a license plate number. The method includes: for a vehicle to be completed with a missing license plate number detected, determining each candidate vehicle to be completed from a historical driving record dataset; each candidate vehicle to be completed is a vehicle with the same vehicle information as the vehicle to be completed and a known license plate number; based on each candidate vehicle to be completed, completing the license plate number of the vehicle to be completed.

[0007] Different from the license plate recognition technology in the background art, the above solution determines each candidate vehicle to be completed from a historical driving record dataset by using big data analysis. Among them, since each candidate vehicle to be completed is a vehicle with the same vehicle information as the vehicle to be completed with a missing license plate number and a known license plate number, based on each candidate vehicle to be completed, the license plate number of the vehicle to be completed can be completed. In this method, based on the technology of big data analysis, the license plate number that has been accurately recognized is used to complete the license plate number of the vehicle with abnormal recognition, which improves the detection rate of license plate numbers, and at the same time, the accuracy of the detected license plate numbers is greatly improved.

[0008] In a possible implementation method, the historical driving record data set is composed of the historical driving records of each vehicle in the first set duration closest to the detection time of the vehicle to be filled; the method for filling the license plate number of the vehicle to be filled based on the candidate filling vehicles includes: for any one of the candidate filling vehicles, determining each quasi-filling vehicle from the candidate filling vehicles according to the first historical driving record of the candidate filling vehicle and the detection position of the vehicle to be filled; the quasi-filling vehicles are vehicles that meet the spatio-temporal requirements with the vehicle to be filled; for any one of the quasi-filling vehicles, determining a target vehicle from the quasi-filling vehicles according to the most frequently appearing area of the quasi-filling vehicle and the detection position of the vehicle to be filled, and filling the license plate number of the vehicle to be filled based on the license plate number of the target vehicle; the most frequently appearing area is determined based on the second historical driving record of the quasi-filling vehicle in the second set duration closest to the detection time of the vehicle to be filled, and the second set duration is greater than the first set duration.

[0009] Due to the high density of checkpoint and video surveillance devices in the current public security construction, although the capture effect of some devices on vehicles is not good, the capture effect of most checkpoint devices on vehicles is still very good. Therefore, based on the current image recognition algorithm, the license plate numbers of some vehicles can be correctly recognized. Based on this, in the embodiments of the present application, by using big data analysis technology, including matching each candidate filling vehicle that conforms to the vehicle information of the vehicle to be filled, then judging whether the spatio-temporal relationship between the vehicle to be filled and each candidate filling vehicle is reasonable, and determining each quasi-filling vehicle, and finally judging whether the vehicle to be filled appears in the most frequently appearing area of each quasi-filling vehicle, and finally determining the target vehicle that can be used to update the license plate number of the vehicle to be filled. In this method, based on the technology of big data analysis, the license plate number of a vehicle with abnormal recognition is filled with the license plate number that has been accurately recognized, including first matching vehicle information, then judging the rationality of the spatio-temporal relationship within a relatively short time, and finally judging the most frequently appearing area within a relatively long time, step by step, thus improving the detection rate of the license plate number, and at the same time, the accuracy of the detected license plate number is greatly improved.

[0010] In a possible implementation method, the vehicle to be completed is obtained through the following steps, including: detecting the body information of each vehicle and the license plate number completion identifier obtained in real time, so as to obtain the vehicle information of each vehicle; the body information includes license plate number information, vehicle category information, main brand information, sub - brand information, body color information and license plate color information; if it is determined that the vehicle information of any vehicle among the vehicles includes that the license plate number is missing no more than 2 digits, vehicle category information, main brand information, body color information, license plate color information and does not include the license plate number completion identifier, then the vehicle is used as the vehicle to be completed.

[0011] In the above solution, when the bayonet and video monitoring equipment built on the road capture a vehicle, the vehicle information of the captured vehicle can be obtained through the detection of the body information of the captured vehicle and the license plate number completion identifier; if the vehicle information of the captured vehicle includes that the license plate number is missing no more than 2 digits, vehicle category information, main brand information, body color information and license plate color information, and at the same time does not include the license plate number completion identifier, then it can be determined that the captured vehicle is the vehicle to be completed that needs to complete the license plate number. By making the above constraints on the vehicle to be completed, it can be ensured that when the license plate number of the vehicle to be completed is completed, the accuracy of the completed license plate number can be guaranteed.

[0012] In a possible implementation method, the historical driving record includes the information of the capture position and capture time when the vehicle is captured during driving; for any one of the candidate completion vehicles among the candidate completion vehicles, determining each quasi - completion vehicle from the candidate completion vehicles according to the first historical driving record of the candidate completion vehicle and the detection position of the vehicle to be completed includes: for any one of the candidate completion vehicles among the candidate completion vehicles, according to the i - th distance information between the capture position when the candidate completion vehicle is captured for the i - th time during driving and the detection position of the vehicle to be completed, and according to the i - th time interval between the capture time when the candidate completion vehicle is captured for the i - th time and the detection time of the vehicle to be completed, determining the i - th driving speed of the candidate completion vehicle assuming it is the vehicle to be completed on the road section corresponding to the i - th distance information; determining whether the i - th driving speed conforms to the preset driving speed requirement of the road section corresponding to the i - th distance information; if there is at least one driving speed when captured that conforms to the set speed requirement, then the candidate completion vehicle is used as the quasi - completion vehicle.

[0013] In the above solution, after determining each candidate vehicle to be supplemented that is consistent with the vehicle information of the vehicle to be supplemented, it is possible to first judge the rationality of the spatio-temporal relationship between the two when each candidate vehicle to be supplemented is assumed to be the vehicle to be supplemented. Finally, some candidate vehicles to be supplemented with reasonable spatio-temporal relationships can be used as each quasi-supplemented vehicle. Subsequently, the license plate number of the vehicle to be supplemented can be supplemented based on each quasi-supplemented vehicle. This method calculates the rationality of vehicle travel in a relatively short time, and only uses some candidate vehicles to be supplemented that meet spatio-temporal rationality among each candidate vehicle to be supplemented as each quasi-supplemented vehicle, which can ensure that the license plate number finally supplemented for the vehicle to be supplemented will have a high accuracy.

[0014] In a possible implementation method, for any one of the quasi-supplemented vehicles among the quasi-supplemented vehicles, determining a target vehicle from the quasi-supplemented vehicles according to the most frequently appearing area of the quasi-supplemented vehicle and the detection position of the vehicle to be supplemented includes: for any one of the quasi-supplemented vehicles among the quasi-supplemented vehicles, obtaining the most frequently appearing area of the quasi-supplemented vehicle; the most frequently appearing area is an area where the number of appearances of the quasi-supplemented vehicle determined based on the second historical driving record exceeds a set number of times; if the detection position of the vehicle to be supplemented conforms to the same most frequently appearing area of at least two quasi-supplemented vehicles, the quasi-supplemented vehicle with the largest number of appearances is determined as the target vehicle.

[0015] In the above solution, after determining each candidate vehicle to be supplemented that is consistent with the vehicle information of the vehicle to be supplemented, it is possible to first judge the rationality of the spatio-temporal relationship between the two when each candidate vehicle to be supplemented is assumed to be the vehicle to be supplemented; since there may be multiple candidate vehicles to be supplemented with reasonable spatio-temporal relationships (i.e., quasi-supplemented vehicles), in order to further improve the accuracy of the license plate number finally supplemented for the vehicle to be supplemented, it is also possible to continue to compare whether the detection position of the vehicle to be supplemented is in the most frequently appearing area of each quasi-supplemented vehicle, and if the detection position of the vehicle to be supplemented meets the most frequently appearing area of at least two quasi-supplemented vehicles, a decision can be made according to the number of appearances of the at least two quasi-supplemented vehicles in the most frequently appearing area. For example, the quasi-supplemented vehicle with the largest number of appearances is used as the target vehicle, and its license plate number is used to update the license plate number of the vehicle to be supplemented.

[0016] In a possible implementation method, obtaining the most frequently appearing area of any one of the quasi-completion vehicles for each of the quasi-completion vehicles includes: for any one of the quasi-completion vehicles for each of the quasi-completion vehicles, obtaining the second historical driving record of the quasi-completion vehicle within a second set time period closest to the detection time of the vehicle to be completed; for any capture position recorded in the second historical driving record of the quasi-completion vehicle, setting a first area based on the capture position; for any one of the first areas, counting the number of times the quasi-completion vehicle appears in the first area within the second set time period; and taking the first areas where the number of appearances of the quasi-completion vehicle in each of the first areas meets the set number of times as the most frequently appearing areas of the quasi-completion vehicle.

[0017] In the above solution, it describes how to obtain the most frequently appearing area of the quasi-completion vehicle, including obtaining the second historical driving record of the quasi-completion vehicle within a second set time period closest to the detection time of the vehicle to be completed. Since the second historical driving record includes information about each time the quasi-completion vehicle is captured during the second set time period during driving, including the position of each capture. Thus, for each capture position, a region can be set for this capture position by itself, that is, the first area (such as an area within a radius of two kilometers from the capture position). Then, for each first area, the number of times the quasi-completion vehicle appears in this first area during the second set time period can be counted. Finally, by taking the first areas that meet the set number of times requirement as the most frequently appearing areas of the quasi-completion vehicle. In this method, by setting a size for the capture position by itself, the fault tolerance of the area is increased; and by taking the first areas that meet the number of times requirement as the most frequently appearing areas of the quasi-completion vehicle, it means that the quasi-completion vehicle often goes to these first areas during the second set time period, while the first areas that do not meet the set number of times requirement indicate that these first areas are places where the quasi-completion vehicle rarely appears. In this way, on the one hand, it can provide a valuable calculation basis for the later license plate number completion process, and at the same time, it can also reduce the calculation pressure in the license plate number completion process.

[0018] In a possible implementation method, for the vehicle to be completed, if no vehicle with vehicle information consistent with the vehicle to be completed and a known license plate number is obtained from the historical driving record dataset; or, if no vehicle that meets the spatio-temporal requirements with the vehicle to be completed is determined from each of the candidate completion vehicles; or, if the detection position of the vehicle to be completed does not conform to the most frequently appearing area of any one of the quasi-completion vehicles, wait for a third set time period, and return to the step of determining each candidate completion vehicle from the historical driving record dataset.

[0019] In the above solution, for the vehicle to be completed, if any vehicle whose vehicle information is the same as that of the vehicle to be completed and whose license plate number is known cannot be obtained from the historical driving record dataset, and if any vehicle that meets the spatio-temporal requirements with the vehicle to be completed cannot be determined from the candidate completion vehicles, and if the detection position of the vehicle to be completed does not conform to the most frequently occurring area of each quasi-completion vehicle, these phenomena can all indicate that the time when the vehicle to be completed left is not long ago, and the number of captures of it by road checkpoints and video surveillance devices is not enough. Therefore, there is no record information about the vehicle to be completed in the current historical driving record dataset. Therefore, in order to ensure the accuracy of the license plate number to be completed for the vehicle to be completed, the operation of completing the license plate number of the vehicle to be completed can be performed again after waiting for a period of time.

[0020] In a possible implementation method, a license plate number completion identifier is added to the vehicle to be completed; the license plate number completion identifier is used to add the information of the vehicle to be completed to the historical driving record dataset based on the license plate number completion identifier.

[0021] In the above solution, an aspect of the historical driving record dataset is described, including adding the information of each vehicle to be completed after the license plate number completion operation to the historical driving record dataset. In this way, the historical driving record dataset can be continuously expanded, which helps to better carry out the work of completing the license plate numbers of new vehicles to be completed.

[0022] In a second aspect, an embodiment of the present application provides a license plate number completion device, which includes: a candidate completion vehicle determination unit, configured to determine each candidate completion vehicle from the historical driving record dataset for a vehicle to be completed with a missing license plate number detected; each candidate completion vehicle is a vehicle whose vehicle information is the same as that of the vehicle to be completed and whose license plate number is known; a license plate number completion unit, configured to complete the license plate number of the vehicle to be completed based on each candidate completion vehicle.

[0023] In a third aspect, an embodiment of the present application provides a computing device, including:

[0024] A memory for storing program instructions;

[0025] A processor for calling the program instructions stored in the memory and executing, according to the obtained program, any implementation method in the first aspect.

[0026] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to cause a computer to execute any implementation method in the first aspect. Description of the Drawings

[0027] To more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the attached drawings required for the description of the embodiments. Obviously, the attached drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other attached drawings can also be obtained based on these attached drawings.

[0028] Figure 1 Schematic diagram of a license plate number completion method provided by an embodiment of the present application;

[0029] Figure 2 Schematic diagram of a license plate number completion device provided by an embodiment of the present application;

[0030] Figure 3 Schematic diagram of a computing device provided by an embodiment of the present application. Detailed implementation manners

[0031] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the attached drawings. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present application.

[0032] At present, accurately detecting the license plate numbers of vehicles traveling on the road is of great significance to the traffic department. However, the currently commonly used license plate recognition technologies are difficult to cover all scenarios, resulting in a low recognition accuracy of license plate numbers.

[0033] In view of the above technical problems, an embodiment of the present application provides a license plate number completion method. As Figure 1 shown, it is a schematic diagram of a license plate number completion method provided by an embodiment of the present application. The method includes the following steps:

[0034] Step 101, for the vehicle to be completed with a missing license plate number detected, determine each candidate completion vehicle from the historical driving record dataset.

[0035] Among them, the historical driving record dataset is composed of the historical driving records of each vehicle in the first set duration closest to the detection time of the vehicle to be completed.

[0036] Among them, each candidate completion vehicle is a vehicle with the same vehicle information as the vehicle to be completed and a known license plate number.

[0037] In some embodiments of the present application, the vehicle to be completed is obtained through the following method, including: detecting the body information of each vehicle and the license plate number completion identifier obtained in real time, so as to obtain the vehicle information of each vehicle; the body information includes license plate number information, vehicle category information, main brand information, sub-brand information, body color information, and license plate color information; if it is determined that the vehicle information of any vehicle among the vehicles includes that the missing number of digits of the license plate number is no more than 2, vehicle category information, main brand information, body color information, license plate color information, and does not include the license plate number completion identifier, then the vehicle is used as the vehicle to be completed.

[0038] Based on the collection of image information of vehicles traveling on the road by the checkpoints and video surveillance devices deployed by the traffic department, the license plate numbers of most vehicles can be accurately obtained through image recognition technology. However, there are still some vehicles whose license plate numbers cannot be accurately obtained during the image recognition process due to the deviation between the vehicle and the position of the collection camera during the image collection process. To solve this problem, the embodiments of the present application propose to complete the license plate numbers of vehicles with the missing number of digits of the license plate number no more than 2. In addition, during the process of completing the license plate number, the vehicle information that needs to be obtained at least for the vehicle whose license plate number needs to be completed includes vehicle category information, vehicle main brand information, body color information, license plate color information, and the information of the license plate number that needs to be completed for the vehicle.

[0039] For example, for a specified region, for any vehicle image collected by the checkpoints and video surveillance devices in the specified area, first determine whether the image carries a license plate number completion identifier, and there are the following two situations:

[0040] Situation 1: If it is confirmed that the image carries a license plate number completion identifier (as for why the image carries a license plate number completion identifier and what the meaning of the license plate number completion identifier is, please refer to the following embodiments), then there is no need to perform image recognition technology on it, and only need to store the image and the information of the image (the information refers to the license plate number of the vehicle, vehicle category information, main brand information, body color information, license plate color information, and the information of the capture position and capture time of the vehicle each time it is captured; if there is vehicle sub-brand information, the information can also include vehicle sub-brand information) in the database to enrich the historical driving record data set.

[0041] Situation 2: If it is confirmed that the image does not carry a license plate number completion identifier, then it is necessary to perform image recognition technology on the image. When performing image recognition technology on the image, it can also include the following two situations:

[0042] Case 2.1: If the license plate number of the vehicle in the image can be obtained through image recognition technology, then store the image and its information (the information refers to the license plate number of the vehicle, vehicle category information, main brand information, body color information, license plate color information, and the information of the capture location and capture time when the vehicle was captured this time; if there is vehicle sub-brand information, the information can also include vehicle sub-brand information) in the database, so as to gradually enrich the historical driving record dataset.

[0043] Case 2.2: If the license plate number of the vehicle in the image cannot be obtained through image recognition technology for the time, then determine whether the number of missing digits of the vehicle's license plate number is no more than 2 digits; if it is confirmed that it is no more than 2 digits, then continue to detect parameters such as the category, main brand, sub-brand, body color, and license plate color of the vehicle; if it is confirmed that the category, main brand information, body color information, and vehicle color information of the vehicle can all be obtained (whether the information of the vehicle sub-brand can be detected is not so important), then it can be confirmed that the vehicle is a vehicle to be supplemented, that is, the vehicle is a vehicle that needs to supplement the license plate number.

[0044] It should be noted that when image recognition technology is applied, if it is confirmed that the number of missing digits of the license plate number in the vehicle image is more than 2 digits, or although the number of missing digits does not exceed 2 digits, but at least one of the vehicle category, main brand, body color, license plate color, etc. is missing, then the vehicle image is also entered into the historical driving record dataset, but the vehicle is a vehicle with a missing license plate number in the historical driving record dataset.

[0045] It should be noted that the designated area can be at the national, provincial, municipal, county level, etc., and this application does not make a limitation.

[0046] Some necessary conditions for a vehicle to be a vehicle to be completed with license plate numbers are described above. That is, the vehicle to be completed is a vehicle with no more than 2 missing digits in the license plate number, known vehicle category information, known vehicle main brand information, known body color, and known license plate color. Given the information of the vehicle to be completed, in the embodiments of the present application, a preliminary screening can be performed on each vehicle suspected to be the vehicle to be completed from the historical driving record data set within the first set duration closest to the detection time of the vehicle to be completed, based on the vehicle information of the vehicle to be completed. For example, if the vehicle to be completed is detected at 9:00 am on November 4, 2021, then a preliminary screening can be performed on each vehicle suspected to be the vehicle to be completed based on the historical driving record data set during the period from 0:00 to 9:00 on that day (i.e., the first set duration); if several vehicles can be screened out, and each vehicle in the several vehicles has the same vehicle information as the vehicle to be completed, that is, the same category, main brand, body color, and license plate color as the vehicle to be completed, and the license plate number of each vehicle in the several vehicles is known, then the several vehicles are each candidate completion vehicle. Therefore, in the embodiments of the present application, by matching the vehicle information of the vehicle to be completed with the vehicles with known license plate numbers within a period of time (i.e., the first set duration) before it is detected, when a result can be matched, it means that the vehicle to be completed may be one of the candidate completion vehicles, thus providing an entry point for completing the license plate number of the vehicle to be completed.

[0047] Step 102: Complete the license plate number of the vehicle to be completed based on the candidate completion vehicles.

[0048] Different from the license plate recognition technology in the background art, the above solution determines each candidate completion vehicle from the historical driving record data set by adopting a big data analysis method. Since each candidate completion vehicle is a vehicle with the same vehicle information as the vehicle to be completed with a missing license plate number and a known license plate number, the license plate number of the vehicle to be completed can be completed based on the candidate completion vehicles. In this method, based on the big data analysis technology, the license plate number that has been accurately recognized is used to complete the license plate number of the vehicle with abnormal recognition, which improves the detection rate of the license plate number, and at the same time, the accuracy of the detected license plate number is greatly improved.

[0049] The following will describe some of the above steps in detail with examples.

[0050] In an implementation of the above step 102, the method of completing the license plate number of the vehicle to be completed based on each candidate completion vehicle includes: for any candidate completion vehicle among the candidate completion vehicles, determining each prospective completion vehicle from the candidate completion vehicles according to the first historical driving record of the candidate completion vehicle and the detection position of the vehicle to be completed; the prospective completion vehicles are vehicles that meet the spatio-temporal requirements with the vehicle to be completed; for any prospective completion vehicle among the prospective completion vehicles, determining a target vehicle from the prospective completion vehicles according to the most frequently appearing area of the prospective completion vehicle and the detection position of the vehicle to be completed, and completing the license plate number of the vehicle to be completed based on the license plate number of the target vehicle; the most frequently appearing area is determined based on the second historical driving record of the prospective completion vehicle within a second set time period closest to the detection time of the vehicle to be completed, and the second set time period is greater than the first set time period.

[0051] In some implementations of the present application, the historical driving record includes information on the capture position and capture time when the vehicle is captured during driving; for any candidate completion vehicle among the candidate completion vehicles, determining each prospective completion vehicle from the candidate completion vehicles according to the first historical driving record of the candidate completion vehicle and the detection position of the vehicle to be completed includes: for any candidate completion vehicle among the candidate completion vehicles, determining the i-th driving speed of the candidate completion vehicle assuming it is the vehicle to be completed on the road section corresponding to the i-th distance information according to the i-th distance information between the capture position when the candidate completion vehicle is captured for the i-th time during driving and the detection position of the vehicle to be completed, and according to the i-th time interval between the capture time when the candidate completion vehicle is captured for the i-th time and the detection time of the vehicle to be completed; determining whether the i-th driving speed meets the set speed requirement for the road section corresponding to the i-th distance information; if there is at least one driving speed during capture that meets the set speed requirement, then taking the candidate completion vehicle as a prospective completion vehicle.

[0052] For example, at 9:00 am on November 4, 2021, after detection, a vehicle to be supplemented, designated as vehicle A, can be determined. Then, 9:00 am is the detection time of vehicle A. Based on the vehicle information of vehicle A, it is assumed to search in the historical driving record dataset corresponding to the time period from 0:00 to 9:00 on the same day. If it is determined that the vehicle information of 10 vehicles with known license plate numbers in the historical driving record dataset is respectively consistent with the vehicle information of vehicle A, then these 10 vehicles are determined as candidate supplemented vehicles and are respectively designated as vehicle 1, vehicle 2... vehicle 10. Among them, the time period from 0:00 to 9:00 on November 4, 2021 is the first set duration. Then, for any one of these 10 vehicles (hereinafter, vehicle 1 is used as an example), the following operations can be performed:

[0053] Step 1: Obtain the driving records of vehicle 1 captured by the checkpoints and video surveillance equipment during the road construction from 0:00 to 9:00 on November 4, 2021. Assume that vehicle 1 was captured 5 times in total. For each capture, obtain the capture location and capture time of vehicle 1. Further, based on the detection time of vehicle A, the 5 capture records of vehicle 1 are respectively designated as the 1st capture, the 2nd capture... the 5th capture in the order of the capture time from far to near.

[0054] Step 2: For each capture of vehicle 1, it can be assumed that it is vehicle A that is captured. Thus, under such an assumption, calculate the spatio-temporal rationality that vehicle A and vehicle 1 are the same vehicle, including: If it shows unreasonable in the spatio-temporal relationship, then confirm that vehicle 1 is definitely not vehicle A, and further, the embodiments of the present application will not supplement the license plate number of vehicle A based on the license plate number of vehicle 1; If it shows reasonable in the spatio-temporal relationship, then the embodiments of the present application may supplement the license plate number of vehicle A based on the license plate number of vehicle 1. Among them, there are the following two calculation logics for spatio-temporal rationality:

[0055] Calculation Logic 1:

[0056] For example, assume that the vehicle 1 was captured for the first time at 6:00 am on November 4, 2021, and the capture location at that time was Location 1. Then, by calculating the time difference between the first capture time of vehicle 1 (i.e., 6:00 am) and the detection time of vehicle A (i.e., 9:00 am), and by calculating the shortest distance between the first capture location of vehicle 1 (i.e., Location 1) and the detection location of vehicle A through the road network, based on this time difference and this shortest distance, the speed of vehicle 1 from 6:00 am to 9:00 am when assuming vehicle 1 is vehicle A and traveling on the section from Location 1 to the detection location of vehicle A can be determined, denoted as Speed 1. Next, through the road network data, the fastest passing speed of a vehicle in the objective conditions from 6:00 am to 9:00 am and traveling on the section between Location 1 and the detection location of vehicle A can be determined, denoted as Speed 2. Finally, compare Speed 1 with Speed 2. If it is determined that the degree to which Speed 1 is greater than Speed 2 exceeds the set value, for example, it is determined that Speed 1 is 1.5 times greater than Speed 2, then it can be determined that the assumption that vehicle 1 is vehicle A does not hold. In this way, the possibility that vehicle 1 is vehicle A can be excluded and directly jump to the process of processing another candidate vehicle to be filled, such as jumping to the processing process of vehicle 2. The processing process of vehicle 2 can refer to the processing process of vehicle 1 in the embodiments of this application and will not be elaborated here. Otherwise (otherwise means that the degree to which Speed 1 is greater than Speed 2 does not exceed the set value), it can be processed in the following two ways, including:

[0057] Method 1: According to the same logic, calculate again the rationality of the spatio-temporal relationship between the capture location and capture time data of vehicle 1 when it is captured for the second, third,..., fifth times and the time and location data when vehicle A is detected. The processing process for each capture of vehicle 1 for the second, third,..., fifth times can refer to the processing process of the first capture of vehicle 1 in the embodiments of this application. Among them, as long as the capture data of a certain time does not meet the spatio-temporal rationality requirements, the possibility that vehicle 1 is vehicle A is excluded and directly jump to the process of processing another candidate vehicle to be filled. When and only when the 5 capture records of vehicle 1 all meet the spatio-temporal rationality requirements with vehicle A, vehicle 1 is retained as a quasi-vehicle to be filled.

[0058] Method 2: Consider vehicle 1 as reasonable, and without further calculating the rationality of the spatio-temporal relationship between the data of vehicle 1 when it is captured for the second, third,..., fifth times and the time and location data when vehicle A is detected, directly retain vehicle 1 as a quasi-vehicle to be filled.

[0059] Calculation Logic 2:

[0060] Calculate the spatio-temporal rationality between Vehicle 1 and Vehicle A when Vehicle 1 is captured for the first time, the second time... the fifth time (the calculation logic of spatio-temporal rationality can refer to the details described in Calculation Logic 1 in the previous text and will not be elaborated here); after obtaining 5 comparison results, if at least one of the comparison results meets the spatio-temporal rationality requirements, determine that Vehicle 1 is reasonable, and thus retain Vehicle 1 as a quasi-completion vehicle and directly jump to the process of processing another candidate completion vehicle.

[0061] For the two calculation logics for determining quasi-completion vehicles described above, in their specific use, there is no limitation in this embodiment of the present application on how to use them. For example, in a certain process of determining quasi-completion vehicles, only Calculation Logic 1 can be used, or only Calculation Logic 2 can be used, or the two can be combined. For example, Calculation Logic 1 is used for candidate completion vehicles to be processed with odd numbers, and Calculation Logic 2 is used for candidate completion vehicles to be processed with even numbers.

[0062] Based on the above processing process, after calculating the spatio-temporal rationality of 10 candidate completion vehicles, assume that there are two vehicles, Vehicle 1 and Vehicle 2, that meet the spatio-temporal rationality requirements, that is, Vehicle 1 and Vehicle 2 are each a quasi-completion vehicle.

[0063] In some implementations of the present application, for any one of the quasi-completion vehicles among the various quasi-completion vehicles, determining a target vehicle from the various quasi-completion vehicles according to the most frequently occurring area of the quasi-completion vehicle and the detection position of the vehicle to be completed includes: for any one of the quasi-completion vehicles among the various quasi-completion vehicles, obtaining the most frequently occurring area of the quasi-completion vehicle; the most frequently occurring area is an area where the occurrence times of the quasi-completion vehicle based on the second historical driving record exceed a set number of times; if the detection position of the vehicle to be completed conforms to the same most frequently occurring area of at least two quasi-completion vehicles, determine the quasi-completion vehicle with the most occurrence times as the target vehicle.

[0064] Continuing with the previous example, this embodiment of the present application can determine each quasi-completion vehicle that meets the reasonable driving requirements in terms of spatio-temporal relationship with the vehicle to be completed from the historical driving record dataset corresponding to the first set duration closest to the detection time of the vehicle to be completed; thus, after obtaining each quasi-completion vehicle, a possible implementation is to directly update the license plate number of the vehicle to be completed based on the license plate numbers of each quasi-completion vehicle. Among them, if the number of quasi-completion vehicles is multiple, one can be randomly selected from them to update the license plate number of the vehicle to be completed. At this time, the selected quasi-completion vehicle used to complete the license plate number of the vehicle to be completed is the target vehicle.

[0065] Another possible implementation is to further screen each candidate vehicle to be filled up, so as to select a more reasonable vehicle to update the license plate number of the vehicle to be filled up. Among them, since the vehicles in the historical driving record dataset can only reflect the performance of the vehicles in a relatively short period of time, it is still not comprehensive enough. Therefore, in the embodiments of the present application, it is proposed that some performances of each candidate vehicle to be filled up in daily life, such as the area where the vehicle most frequently appears, can be combined, and then compared with the current detection status of the vehicle to be filled up to confirm the possibility that one or some of the candidate vehicles to be filled up are the vehicle to be filled up.

[0066] In some implementations of the present application, for any candidate vehicle to be filled up among the candidate vehicles to be filled up, obtaining the area where the candidate vehicle most frequently appears includes: for any candidate vehicle to be filled up among the candidate vehicles to be filled up, obtaining the second historical driving record of the candidate vehicle in the second set duration closest to the detection time of the vehicle to be filled up; for any captured position recorded in the second historical driving record of the candidate vehicle, setting a first area based on the captured position; for any first area among the first areas, counting the number of times the candidate vehicle appears in the first area within the second set duration; using the first areas in which the number of times the candidate vehicle appears meets the set number as the area where the candidate vehicle most frequently appears.

[0067] Continuing with the previous example, assume that two candidate vehicles to be filled up that meet the spatio-temporal driving requirements are selected from a total of 10 candidate vehicles to be filled up, namely Vehicle 1 and Vehicle 2. Then, for any one of Vehicle 1 and Vehicle 2 (hereinafter, Vehicle 1 is used as an example for illustration), the area where the vehicle most frequently appears can be obtained through the following method:

[0068] Based on the previous example, if the detection time of vehicle A is 9:00 am on November 4, 2021, then the records of vehicle 1 captured during the time interval from October 4, 2021 to November 3, 2021 can be obtained. Since each captured record contains the capture location, a designated area can be set based on one capture location. For example, a circular area with a radius of 2 kilometers centered on the capture location is used as the designated area. Then, by counting the total number of times vehicle 1 has appeared in the same designated area in the past 30 days (i.e., the number of captures), and taking the designated areas where the number of occurrences is not less than the set value as the most frequently appearing areas of vehicle 1. For example, the number set value can be set to 5 times. Thus, the designated areas with a count greater than or equal to 5 times can be regarded as the most frequently appearing areas of vehicle 1. For the designated areas where the number of occurrences does not meet 5 times, it indicates that vehicle 1 does not frequently appear in these areas in the past month. Therefore, to reduce the computational effort for subsequent matching, the designated areas where the number of occurrences does not meet 5 times can be ignored. Among them, the period from October 4, 2021 to November 3, 2021 is the second set duration.

[0069] According to the above method, after obtaining the most frequently appearing areas of vehicle 1 and vehicle 2 respectively, the detection location of vehicle A can be matched with the most frequently appearing areas of vehicle 1 and the most frequently appearing areas of vehicle 2 respectively. The matching results may include the following situations:

[0070] Situation 1: The detection location of vehicle A is exactly within one of the most frequently appearing areas of vehicle 1. Then it can be determined that vehicle 1 is vehicle A. Thus, the license plate number of vehicle A can be updated using the license plate number of vehicle 1. The update includes directly replacing the license plate number of vehicle A with the license plate number of vehicle 1 and filling in the positions where the license plate number of vehicle A is missing with the license plate numbers at the same positions of vehicle 1. At this time, vehicle 1 is the target vehicle.

[0071] Situation 2: The detection location of vehicle A is exactly within one of the most frequently appearing areas of vehicle 2. Then it can be determined that vehicle 2 is vehicle A. Thus, the license plate number of vehicle A can be updated using the license plate number of vehicle 2. The update includes directly replacing the license plate number of vehicle A with the license plate number of vehicle 2 and filling in the positions where the license plate number of vehicle A is missing with the license plate numbers at the same positions of vehicle 2. At this time, vehicle 2 is the target vehicle.

[0072] Case 3: The detected position of vehicle A is located in one of the most frequently appearing areas of vehicle 1, and at the same time, the detected position of vehicle A is also located in one of the most frequently appearing areas of vehicle 2. Therefore, a decision can be made based on the number of appearances of vehicle 1 and vehicle 2 in the same most frequently appearing area, that is, the vehicle with the largest number of appearances is used as the vehicle for completing the license plate number of vehicle A, that is, the target vehicle. If the number of appearances of vehicle 1 in the same most frequently appearing area is greater than the number of appearances of vehicle 2 in the same most frequently appearing area, the license plate number of vehicle 1 can be used to update the license plate number of vehicle A. The update includes two operations: directly replacing the license plate number of vehicle A with the license plate number of vehicle 1 and filling in the missing positions of the license plate number of vehicle A with the license plate numbers of vehicle 1 at the same positions.

[0073] Case 4: The detected position of vehicle A is neither located in the most frequently appearing area of vehicle 1 nor in the most frequently appearing area of vehicle 2. Then there are two solutions as follows:

[0074] Method 1: Randomly select a vehicle from vehicle 1 and vehicle 2, and update the license plate number of vehicle A with the license plate number of this vehicle;

[0075] Method 2: Do not complete the license plate number of vehicle A this time. After waiting for a set duration, re-determine the license plate number of vehicle A based on the updated historical driving record dataset. The specific process will not be repeated.

[0076] Since the density of checkpoint and video surveillance devices built by the public security currently is very high, although the capture effect of some devices on vehicles is not good, the capture effect of most checkpoint devices on vehicles is still very good. Based on the current image recognition algorithm, the license plate numbers of vehicles can be correctly recognized completely. Based on this, in the embodiments of the present application, by using big data analysis technology, including matching the vehicle information of the target vehicle to find the corresponding first vehicles, then judging whether the spatio-temporal relationship between the target vehicle and each first vehicle is reasonable, and determining each second vehicle. Finally, by judging whether the target vehicle appears in the most frequently appearing area of each second vehicle, the third vehicle that can be used to update the license plate number of the target vehicle is finally determined. In this method, based on the big data analysis technology, the license plate number of the vehicle with abnormal recognition is completed with the license plate number that has been accurately recognized, which improves the detection rate of the license plate number, and at the same time, the accuracy of the detected license plate number is also greatly improved.

[0077] In certain implementations of the present application, for the vehicle to be completed, if any vehicle whose vehicle information is consistent with that of the vehicle to be completed and whose license plate number is known is not obtained from the historical driving record data set; or, any vehicle that meets the time and space requirements of the vehicle to be completed is not determined from the candidate completion vehicles; or, the detected position of the vehicle to be completed does not meet the most common appearance area of any of the quasi-completion vehicles, then wait for a third set time length and return to the step of determining each candidate completion vehicle from the historical driving record data set.

[0078] When using vehicle A as an example to illustrate the solution, it was first assumed that 10 candidate completion vehicles, namely vehicle 1, vehicle 2, ..., vehicle 10, with known license plate numbers, could be identified from the historical driving record data set. However, when matching the historical driving record data set with the first vehicles that have the same vehicle information as vehicle A and a known license plate number based on vehicle A's vehicle information, it is possible that candidate completion vehicles with the same vehicle information as vehicle A and a known license plate number could not be matched. This indicates that vehicle A has not been out long enough, resulting in no recorded information about it in the historical driving record data set. In this case, you can wait for a period of time, such as 10 minutes, and then re-match vehicle A based on the historical driving record data set 10 minutes later. 10 minutes is the third set time.

[0079] Furthermore, when using vehicle A as an example, we assumed that two potential completion vehicles, vehicle 1 and vehicle 2, could be identified from a total of 10 candidate completion vehicles, vehicle 1, vehicle 2, ..., vehicle 10. However, when calculating the spatiotemporal plausibility between each candidate completion vehicle and vehicle A, it is possible that none of the candidate completion vehicles are spatiotemporally plausible with vehicle A. This also indicates that vehicle A has not been out long enough, resulting in no record of it in the historical driving record dataset. In this case, you can wait for a period of time, such as 10 minutes, and then rematch vehicle A based on the historical driving record dataset 10 minutes later.

[0080] In addition, when the solution was illustrated using vehicle A as an example, several situations that may occur when the detection position of vehicle A is consistent with the most frequently appearing areas of vehicle 1 and vehicle 2 were also discussed. Among them, situation 4 illustrates the possible result that the detection position of vehicle A is neither in the most frequently appearing area of vehicle 1 nor in the most frequently appearing area of vehicle 2. One possible way to deal with this result is to wait for a set time, such as 10 minutes, and then re-match vehicle A based on the historical driving record data set 10 minutes later. The principle is that the time that vehicle A has been out is not long enough, resulting in no recorded information about it in the historical driving record data set.

[0081] In certain implementations of the present application, a license plate number completion identifier is added to the vehicle to be completed; the license plate number completion identifier is used to add the information of the vehicle to be completed to the historical driving record dataset based on the license plate number completion identifier.

[0082] Based on the previous example, when the license plate number of vehicle A is determined based on the historical driving record data set, a license plate number completion flag can also be added for vehicle A. Therefore, before adding the information of vehicle A to the historical driving record data set, by confirming that vehicle A carries the license plate number completion flag, vehicle A and the information carried by vehicle A (such as license plate number, vehicle category, vehicle main brand, body color, license plate color) can be directly entered into the historical driving record data set without having to perform the license plate number completion operation process for vehicle A again.

[0083] Based on the same concept, the embodiment of the present application provides a license plate number completion device, such as Figure 2 , which is a schematic diagram of a license plate number completion device provided in an embodiment of the present application, the device includes a candidate completion vehicle determination unit 201 and a license plate number completion unit 202;

[0084] The candidate completion vehicle determination unit 201 is configured to determine candidate completion vehicles from the historical driving record data set for the detected vehicles to be completed with missing license plate numbers; each candidate completion vehicle is a vehicle whose vehicle information is consistent with that of the vehicle to be completed and whose license plate number is known;

[0085] The license plate number completing unit 202 is configured to complete the license plate number of the vehicle to be completed based on the candidate completed vehicles.

[0086] Further, for the device, the historical driving record dataset is composed of the historical driving records of each vehicle in the first set duration closest to the detection time of the vehicle to be supplemented; the license plate number supplementing unit 202 is specifically configured to: for any candidate vehicle to be supplemented among the candidate vehicles to be supplemented, determine each quasi-supplemented vehicle from the candidate vehicles to be supplemented according to the first historical driving record of the candidate vehicle to be supplemented and the detection position of the vehicle to be supplemented; each of the quasi-supplemented vehicles is a vehicle that meets the spatio-temporal requirements with the vehicle to be supplemented; for any quasi-supplemented vehicle among the quasi-supplemented vehicles, determine a target vehicle from the quasi-supplemented vehicles according to the most frequently appearing area of the quasi-supplemented vehicle and the detection position of the vehicle to be supplemented, and supplement the license plate number of the vehicle to be supplemented based on the license plate number of the target vehicle; the most frequently appearing area is determined based on the second historical driving record of the quasi-supplemented vehicle in the second set duration closest to the detection time of the vehicle to be supplemented, and the second set duration is greater than the first set duration.

[0087] Further, for the device, it further includes a vehicle to be supplemented determination unit 203; the vehicle to be supplemented determination unit 203 is configured to: detect the body information and license plate number supplementing identifier of each vehicle obtained in real time, so as to obtain the vehicle information of each vehicle; the body information includes license plate number information, vehicle category information, main brand information, sub-brand information, body color information, and license plate color information; if it is determined that the vehicle information of any vehicle among the vehicles includes that the license plate number is missing no more than 2 digits, vehicle category information, main brand information, body color information, license plate color information, and does not include the license plate number supplementing identifier, then the vehicle is used as the vehicle to be supplemented.

[0088] Further, for the device, it further includes a quasi-supplemented vehicle determination unit 204; the historical driving record includes information on the capture position and capture time when the vehicle is captured during driving; the quasi-supplemented vehicle determination unit 204 is configured to: for any candidate vehicle to be supplemented among the candidate vehicles to be supplemented, according to the i-th distance information between the capture position when the candidate vehicle to be supplemented is captured for the i-th time during driving and the detection position of the vehicle to be supplemented, and according to the i-th time interval between the capture time when the candidate vehicle to be supplemented is captured for the i-th time and the detection time of the vehicle to be supplemented, determine the i-th driving speed of the candidate vehicle to be supplemented assuming it is the vehicle to be supplemented on the road section corresponding to the i-th distance information; determine whether the i-th driving speed meets the set speed requirement for the road section corresponding to the i-th distance information; if there is at least one driving speed at the time of capture that meets the set speed requirement, then the candidate vehicle to be supplemented is used as a quasi-supplemented vehicle.

[0089] Further, for the device, it further includes a target vehicle determination unit 205; the target vehicle determination unit 205 is configured to: for any one of the quasi-completion vehicles among the respective quasi-completion vehicles, obtain the most frequently occurring area of the quasi-completion vehicle; the most frequently occurring area is an area where the occurrence times of the quasi-completion vehicle determined based on the second historical driving record exceed a set number of times; if the detected position of the vehicle to be completed meets the same most frequently occurring area of at least two quasi-completion vehicles, the quasi-completion vehicle with the most occurrence times is determined as the target vehicle.

[0090] Further, for the device, it further includes a most frequently occurring area determination unit 206; the most frequently occurring area determination unit 206 is configured to: for any one of the quasi-completion vehicles among the respective quasi-completion vehicles, obtain the second historical driving record of the quasi-completion vehicle within a second set time period closest to the detection time of the vehicle to be completed; for any one of the captured positions recorded by the quasi-completion vehicle in the second historical driving record, set a first area based on the captured position; for any one of the first areas, count the number of times the quasi-completion vehicle appears in the first area within the second set time period; and use the first areas where the occurrence times of the quasi-completion vehicle in the respective first areas meet the set number of times as the most frequently occurring areas of the quasi-completion vehicle.

[0091] Further, for the device, it further includes a reprocessing unit 207; the reprocessing unit 207 is configured to: for the vehicle to be completed, if no vehicle with vehicle information consistent with the vehicle to be completed and a known license plate number is obtained from the historical driving record data set; or no vehicle meeting the spatio-temporal requirements with the vehicle to be completed is determined from the respective candidate completion vehicles; or the detected position of the vehicle to be completed does not meet the most frequently occurring area of any one of the quasi-completion vehicles, wait for a third set time period, and return to the step of determining the respective candidate completion vehicles from the historical driving record data set.

[0092] Further, for the device, it further includes a license plate number completion flag setting unit 208; the license plate number completion flag setting unit 208 is configured to: add a license plate number completion flag to the vehicle to be completed; the license plate number completion flag is used to add the information of the vehicle to be completed to the historical driving record data set based on the license plate number completion flag.

[0093] The embodiments of the present application also provide a computing device, which may specifically be a desktop computer, a portable computer, a smart phone, a tablet computer, a personal digital assistant (PDA), etc. The computing device may include a central processing unit (CPU), a memory, input / output devices, etc. The input devices may include a keyboard, a mouse, a touch screen, etc. The output devices may include display devices, such as a liquid crystal display (LCD), a cathode ray tube (CRT), etc.

[0094] The memory may include a read-only memory (ROM) and a random access memory (RAM), and provide program instructions and data stored in the memory to the processor. In the embodiments of the present application, the memory may be used to store program instructions of the license plate number completion method;

[0095] The processor is used to call the program instructions stored in the memory and execute the license plate number completion method according to the obtained program.

[0096] As Figure 3 shown, it is a schematic diagram of a computing device provided by the embodiments of the present application. The computing device includes:

[0097] A processor 301, a memory 302, a transceiver 303, and a bus interface 304; wherein, the processor 301, the memory 302, and the transceiver 303 are connected through a bus 305;

[0098] The processor 301 is used to read the program in the memory 302 and execute the above-mentioned license plate number completion method;

[0099] The processor 301 can be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and an NP. It can also be a hardware chip. The above-mentioned hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above-mentioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0100] The memory 302 is used to store one or more executable programs and can store the data used by the processor 301 when performing operations.

[0101] Specifically, the program can include program code, and the program code includes computer operation instructions. The memory 302 can include volatile memory, such as random-access memory (RAM); the memory 302 can also include non-volatile memory, such as flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); the memory 302 can also include a combination of the above types of memory.

[0102] The memory 302 stores the following elements, executable modules, or data structures, or subsets thereof, or extended sets thereof:

[0103] Operation instructions: including various operation instructions for implementing various operations.

[0104] Operating system: including various system programs for implementing various basic services and processing hardware-based tasks.

[0105] The bus 305 can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 3 it is only represented by a thick line in Figure 3 , but it does not mean that there is only one bus or one type of bus.

[0106] The bus interface 304 can be a wired communication access port, a wireless bus interface, or a combination thereof. Among them, the wired bus interface can be, for example, an Ethernet interface. The Ethernet interface can be an optical interface, an electrical interface, or a combination thereof. The wireless bus interface can be a WLAN interface.

[0107] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to make a computer execute the license plate number completion method.

[0108] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a computer program product, or the like. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0109] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0110] These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device realizes the functions in the process Figure 1One or more processes and / or boxes Figure 1 The functions specified in one box or more boxes.

[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 One or more processes and / or boxes Figure 1 The steps of the functions specified in one box or more boxes.

[0112] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0113] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A method for completing a license plate number, characterized in that: include: For the detected vehicles with missing license plate numbers to be supplemented, candidate supplementary vehicles are determined from the historical driving record data set; Each candidate vehicle is a vehicle with the same vehicle information as the vehicle to be completed and a known license plate number; The historical driving record data set is composed of historical driving records of each vehicle within a first set time period closest to the detection time of the vehicle to be supplemented; For any candidate completing vehicle among the candidate completing vehicles, determining two quasi-completing vehicles from the candidate completing vehicles based on the first historical driving record of the candidate completing vehicle and the detected position of the vehicle to be completed; the quasi-completing vehicles are vehicles that meet the time and space requirements of the vehicle to be completed; For any of the quasi-completing vehicles, a target vehicle is determined from the quasi-completing vehicles based on the most frequently appearing area of the quasi-completing vehicle and the detection position of the vehicle to be completed, and the license plate number of the vehicle to be completed is completed based on the license plate number of the target vehicle; the most frequently appearing area is determined based on the second historical driving record of the quasi-completing vehicle in a second set time period closest to the detection time of the vehicle to be completed, and the second set time period is greater than the first set time period.

2. The method according to claim 1, wherein The vehicles to be supplemented are obtained by the following methods, including: Detecting the vehicle body information and license plate number completion mark of each vehicle obtained in real time, thereby obtaining the vehicle information of each vehicle; the vehicle body information includes license plate number information, vehicle type information, main brand information, sub-brand information, vehicle body color information and license plate color information; If it is determined that the vehicle information of any vehicle among the vehicles includes a license plate number that is missing no more than 2 digits, vehicle category information, main brand information, body color information, license plate color information and does not include the license plate number completion mark, then the vehicle will be treated as a vehicle to be completed.

3. The method according to claim 1, wherein Historical driving records include the location and time of the vehicle being photographed while driving; The method of determining, for any candidate completing vehicle among the candidate completing vehicles, two quasi-completing vehicles from the candidate completing vehicles according to the first historical driving record of the candidate completing vehicle and the detected position of the vehicle to be completed, includes: For any candidate completing vehicle among the candidate completing vehicles, determining, based on the i-th distance information between the capture position of the candidate completing vehicle each i times when it is captured and the detection position of the vehicle to be completed, and based on the i-th time interval corresponding to the capture time of the candidate completing vehicle each i times and the detection time of the vehicle to be completed, an i-th driving speed of the candidate completing vehicle, assuming it is the vehicle to be completed on the road section corresponding to the i-th distance information; determining whether the i-th driving speed and the preset driving speed of the road section corresponding to the i-th distance information meet the set speed requirement; If there is at least one vehicle whose driving speed when captured meets the set speed requirement, the candidate completion vehicle is used as a quasi-completion vehicle.

4. The method according to claim 1, wherein The method of determining a target vehicle from among the quasi-completing vehicles based on a most frequently appearing area of the quasi-completing vehicle and a detected position of the vehicle to be completed for each quasi-completing vehicle includes: For any of the quasi-completing vehicles, obtaining a most frequently appearing area of the quasi-completing vehicle; the most frequently appearing area is an area where the quasi-completing vehicle appears more than a set number of times, as determined based on the second historical driving record; If the detected position of the to-be-completed vehicle matches the same most frequently appearing area of at least two quasi-completed vehicles, the quasi-completed vehicle with the most frequent appearances is determined as the target vehicle.

5. The method according to claim 4, wherein The method further comprises: The step of obtaining, for any of the quasi-completing vehicles, a most frequently appearing area of the quasi-completing vehicle includes: For any of the quasi-complementary vehicles, obtaining a second historical driving record of the quasi-complementary vehicle within a second set time period closest to a detection time of the vehicle to be supplemented; For any captured position of the quasi-completing vehicle recorded in the second historical driving record, a first area is set based on the captured position; for any first area in each first area, the number of times the quasi-completing vehicle appears in the first area within the second set time period is counted; and each first area in which the number of appearances of the quasi-completing vehicle in the first area meets the set number is regarded as the most frequently appearing area of the quasi-completing vehicle.

6. The method according to any one of claims 1 to 5, wherein: The method further comprises: For the vehicle to be supplemented, if no vehicle with the same vehicle information as the vehicle to be supplemented and a known license plate number is obtained from the historical driving record data set; or, No vehicle that meets the time and space requirements of the vehicle to be supplemented is determined from the candidate supplemented vehicles; or If the detected position of the to-be-complemented vehicle does not conform to the most frequent appearance area of any of the quasi-complemented vehicles, the process waits for a third set time period and returns to the step of determining each candidate complemented vehicle from the historical driving record data set.

7. The method according to claim 1, wherein The method further comprises: A license plate number completion identifier is added to the vehicle to be completed; the license plate number completion identifier is used to add information of the vehicle to be completed to the historical driving record data set based on the license plate number completion identifier.

8. A computer device, characterized in that: include: memory for storing computer programs; A processor is configured to call a computer program stored in the memory and execute the method according to any one of claims 1 to 7 according to the obtained program.

9. A computer-readable storage medium, characterized in that The storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images