Vehicle information query method and device

Abstract

The present application discloses a vehicle information query method and device. The method includes: obtaining multiple vehicle images, and obtaining multiple vehicle information keywords corresponding to each vehicle image based on the multiple vehicle images; forming a vehicle information dictionary file of multiple vehicle information keywords corresponding to each vehicle serial number based on the multiple vehicle information keywords corresponding to each vehicle image; establishing an inverted index file based on the vehicle information dictionary file; obtaining a vehicle image to be tested, and obtaining multiple corresponding keywords to be tested based on the vehicle image to be tested; querying the keywords to be tested based on the inverted index file to obtain the vehicle information corresponding to the vehicle image to be tested. The vehicle information query method of the present application can achieve fast search, high speed and high efficiency, solves the problem of poor real-time query and low query efficiency of traditional methods, and can meet market demand.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a vehicle information query method and device. Background Art

[0002] With the development of economy, the number of motor vehicles is increasing, which leads to the need for increasing video surveillance or improving camera resolution to solve vehicle management problems. However, as the number of video files stored in the vehicle management platform increases, the management and efficient query and identification of vehicle data are particularly important. However, the traditional vehicle information query method requires the storage and reading of massive videos and images, which makes the traditional query method poor in real-time performance, and thus leads to low query efficiency and cannot meet business needs.

[0003] Application Contents

[0004] The purpose of this application is to solve the technical problem of low query efficiency caused by poor real-time performance of traditional methods. To achieve the above purpose, this application provides a vehicle information query method and device.

[0005] The present application provides a vehicle information query method, comprising:

[0006] Obtaining a plurality of vehicle images, and obtaining a plurality of vehicle information keywords corresponding to each of the vehicle images based on the plurality of vehicle images;

[0007] According to the plurality of vehicle information keywords corresponding to each of the vehicle images, a vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number is formed;

[0008] Creating an inverted index file based on the vehicle information dictionary file;

[0009] Acquire a vehicle image to be tested, and obtain a plurality of corresponding keywords to be tested according to the vehicle image to be tested;

[0010] The keyword to be tested is queried according to the inverted index file to obtain vehicle information corresponding to the vehicle image to be tested.

[0011] In one embodiment, the step of establishing an inverted index file according to the vehicle information dictionary file includes:

[0012] According to the vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, a frequency file of the vehicle information keyword pointing to frequency information is obtained, wherein the frequency information is the frequency of the vehicle information keyword appearing in the corresponding vehicle serial number;

[0013] According to the vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, obtaining a position file of the position information pointed to by the vehicle information keyword, wherein the position information is the position where the vehicle information keyword appears in the vehicle information dictionary file;

[0014] The vehicle information keywords, the frequency file, and the position file form the inverted index file.

[0015] In one embodiment, the step of establishing an inverted index file according to the vehicle information dictionary file further includes:

[0016] Compressing consecutive identical characters of adjacent and identical vehicle information keywords in the inverted index file to obtain a plurality of compressed vehicle information keywords;

[0017] The inverted index file is formed according to the plurality of compressed vehicle information keywords, the frequency file, and the position file.

[0018] In one embodiment, querying the keyword to be tested according to the inverted index file to obtain vehicle information corresponding to the vehicle image to be tested includes:

[0019] According to a binary search algorithm, query the frequency information and position information corresponding to the keyword to be tested in the inverted index file;

[0020] The vehicle information corresponding to the vehicle image to be tested is obtained according to the frequency information and position information corresponding to the keyword to be tested.

[0021] In one embodiment, the vehicle information query method further includes:

[0022] If the keyword to be tested cannot be found according to the inverted index file, the vehicle information dictionary file and the inverted index file are updated according to the keyword to be tested.

[0023] In one embodiment, a plurality of vehicle images are obtained, and based on the plurality of vehicle images, a plurality of vehicle information keywords corresponding to each of the vehicle images are obtained, and the plurality of vehicle images are recognized according to a deep learning recognition algorithm to obtain a plurality of vehicle information keywords corresponding to each of the vehicle images.

[0024] In one embodiment, the present application provides a vehicle information recognition device, comprising:

[0025] A data acquisition module, used to obtain a plurality of vehicle images, and obtain a plurality of vehicle information keywords corresponding to each of the vehicle images according to the plurality of vehicle images;

[0026] A vehicle information dictionary file forming module, used to form a vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number according to the plurality of vehicle information keywords corresponding to each vehicle image;

[0027] An inverted index file forming module, used to create an inverted index file according to the vehicle information dictionary file;

[0028] A keyword acquisition module for obtaining a vehicle image to be tested, and obtaining a plurality of corresponding keywords to be tested according to the vehicle image to be tested;

[0029] The vehicle information acquisition module is used to query the keyword to be tested according to the inverted index file to obtain the vehicle information corresponding to the vehicle image to be tested.

[0030] In one embodiment, the inverted index file forming module includes:

[0031] A frequency file generating module, for obtaining, according to a vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, a frequency file of frequency information pointing to the vehicle information keyword, wherein the frequency information is the frequency at which the vehicle information keyword appears in the corresponding vehicle serial number;

[0032] A position file generation module is used to obtain a position file pointing to position information of the vehicle information keyword based on the vehicle information dictionary file of the multiple vehicle information keywords corresponding to each vehicle serial number, wherein the position information is the position where the vehicle information keyword appears in the vehicle information dictionary file; the vehicle information keyword, the frequency file and the position file form the inverted index file.

[0033] In one embodiment, the inverted index file forming module further includes:

[0034] A character compression module, used for compressing consecutive identical characters of adjacent and identical vehicle information keywords in the inverted index file to obtain a plurality of compressed vehicle information keywords;

[0035] A forming module is used to form the inverted index file according to the multiple compressed vehicle information keywords, the frequency file and the position file.

[0036] In one embodiment, the vehicle information acquisition module includes:

[0037] A frequency position information acquisition module, used to query the frequency information and position information corresponding to the keyword to be tested in the inverted index file according to a binary search algorithm;

[0038] The information acquisition module is used to obtain the vehicle information corresponding to the vehicle image to be tested according to the frequency information and position information corresponding to the keyword to be tested.

[0039] In one embodiment, the vehicle information query device further includes:

[0040] An updating module is used to update the vehicle information dictionary file and the inverted index file according to the keyword to be tested if the keyword to be tested cannot be found according to the inverted index file.

[0041] In one embodiment, in the data acquisition module, the multiple vehicle images are recognized according to a deep learning recognition algorithm to obtain multiple vehicle information keywords corresponding to each of the vehicle images.

[0042] In the above-mentioned vehicle information query method and device, vehicle information such as license plate number, body color, vehicle model name and vehicle type is obtained by identifying multiple vehicle images to form a vehicle information dictionary file. Based on the correspondence between the vehicle serial number and the vehicle information keyword in the vehicle information dictionary file, an inverted index file is formed. According to the inverted index file, the search for the keyword to be tested is realized, and the vehicle information corresponding to the keyword to be tested is found, and then each other vehicle information keyword is obtained, and the vehicle information corresponding to the vehicle image to be tested is obtained. When one of the keywords to be tested cannot realize the vehicle information search, another keyword to be tested can be replaced for search, the search range can be narrowed, and a comprehensive coverage search is realized, and finally the vehicle information corresponding to the vehicle image to be tested is obtained. Therefore, through the vehicle information query method of the present application, an association relationship between vehicle information keywords, inverted index files, vehicle serial numbers and vehicle information dictionary files is established, and the corresponding complete vehicle information can be obtained by searching the vehicle information keywords in the whole process, and the search time of the whole process can be realized in milliseconds. Therefore, the vehicle information query method of the present application can realize fast search, fast speed and high efficiency, solve the problem of poor real-time query and low query efficiency of traditional methods, and can meet market demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the step flow of the vehicle information query method provided by this application.

[0044] Figure 2 It is a schematic diagram of a table of vehicle information keywords, vehicle serial numbers and appearance locations in an embodiment provided by the present application.

[0045] Figure 3 It is a structural schematic diagram of the vehicle information query device provided by this application. DETAILED DESCRIPTION

[0046] The technical solution of the present application is further described in detail below through the accompanying drawings and embodiments.

[0047] See also Figure 1 , the present application provides a vehicle information query method, comprising:

[0048] S10, obtaining a plurality of vehicle images, and obtaining a plurality of vehicle information keywords corresponding to each vehicle image based on the plurality of vehicle images;

[0049] S20, forming a vehicle information dictionary file of a plurality of vehicle information keywords corresponding to each vehicle serial number according to a plurality of vehicle information keywords corresponding to each vehicle image;

[0050] S30, creating an inverted index file according to the vehicle information dictionary file;

[0051] S40, acquiring a vehicle image to be tested, and obtaining a plurality of corresponding keywords to be tested according to the vehicle image to be tested;

[0052] S50, querying the keyword to be tested according to the inverted index file to obtain vehicle information corresponding to the vehicle image to be tested.

[0053] In S10, vehicle images can be acquired based on high-position video and each checkpoint. Multiple vehicle images are constructed to form a vehicle image database. Multiple vehicle images contain various practical application scenarios. Various practical application scenarios can be holographic intersections, highways, etc. based on high-position video roadside parking. Each vehicle image is identified by a deep learning recognition algorithm to obtain vehicle information such as license plate number, body color, model name, vehicle type, etc. in each vehicle image. License plate number, body color, model name, vehicle type, etc. are vehicle information keywords. The deep learning recognition algorithm can be based on deep learning networks such as LSTM, ABCNet, YOLOV4, etc. to realize multi-category recognition of vehicle information.

[0054] In S20, each vehicle image corresponds to multiple vehicle information keywords, and multiple vehicle images correspond to multiple vehicle information keywords. Vehicle information keywords can represent relevant information of the vehicle, such as vehicle license plate number, body color, vehicle model name, vehicle type, or driver information, etc., which are related to the vehicle. A vehicle corresponds one-to-one with a license plate number, a body color, a vehicle model name, and a vehicle type, etc. A vehicle serial number corresponds to a vehicle, and thus corresponds one-to-one with a license plate number, a body color, a vehicle model name, and a vehicle type, etc. Each vehicle serial number corresponds to a license plate number keyword, a body color keyword, a vehicle model name keyword, and a vehicle type keyword. A one-to-one correspondence relationship is formed between each vehicle serial number and the corresponding vehicle information keywords, and thus a vehicle information dictionary file is formed. The structure and categories of the vehicle information dictionary file are as follows:

[0055] Vehicle serial number 1: License plate number: Jing A77777; Body color: black; Vehicle model name: Audi; Vehicle type: compact car; The vehicle information keywords corresponding to vehicle serial number 1 are [Jing A77777][black][Audi][compact car];

[0056] Vehicle serial number 2: License plate number: Ji B88888; Body color: white; Vehicle model name: FAW-Volkswagen; Vehicle type: SUV; The vehicle information keywords corresponding to vehicle serial number 2 are [Ji B88888][white][FAW-Volkswagen][SUV];

[0057] Vehicle serial number 3: License plate number: Lu C99999; Body color: white; Vehicle model name: Mercedes-Benz; Vehicle type: compact car. The vehicle information keywords corresponding to vehicle serial number 3 are [Lu C99999][white][Mercedes-Benz][compact car].

[0058] The structure and categories of the vehicle information dictionary file can also be set according to the actual situation. Vehicle information keywords can also be keywords such as automobile manufacturers that can represent vehicle information.

[0059] In S30, the vehicle information dictionary file includes multiple vehicle information keywords corresponding to each vehicle serial number. Select one vehicle information keyword from the multiple vehicle information keywords corresponding to each vehicle serial number as the query subject and point to the vehicle serial number that owns the vehicle information keyword. It can also be understood as the following association form: the inverted index relationship of vehicle information keywords to vehicle serial numbers, and thus an inverted index file is formed.

[0060] In S40, the image of the vehicle to be tested can be obtained based on the high-position video and each checkpoint. The image of the vehicle to be tested can include various practical application scenarios. According to the image of the vehicle to be tested, a plurality of keywords to be tested in the corresponding image are obtained through a deep learning recognition algorithm for vehicle information query.

[0061] In S50, the keyword to be tested is put into the constructed inverted index file for query, and the vehicle information corresponding to the keyword to be tested can be obtained, thereby obtaining vehicle information such as license plate number, body color, model name and vehicle type.

[0062] The vehicle information query method provided by the present application obtains vehicle information such as license plate number, body color, vehicle model name and vehicle type by identifying multiple vehicle images, and forms a vehicle information dictionary file. Based on the correspondence between the vehicle serial number and the vehicle information keyword in the vehicle information dictionary file, an inverted index file is formed. According to the inverted index file, the search for the keyword to be tested is realized, the vehicle information corresponding to the keyword to be tested is found, and then each other vehicle information keyword is obtained, and the vehicle information corresponding to the vehicle image to be tested is obtained. When one of the keywords to be tested cannot realize the vehicle information search, another keyword to be tested can be replaced for search, the search range can be narrowed, and a comprehensive coverage search is realized, and finally the vehicle information corresponding to the vehicle image to be tested is obtained. Therefore, through the vehicle information query method of the present application, the association relationship between the vehicle information keyword, the inverted index file, the vehicle serial number and the vehicle information dictionary file is established, and the corresponding complete vehicle information can be obtained by searching the vehicle information keyword in the whole process, and the search time of the whole process can be realized in milliseconds. Therefore, the vehicle information query method of the present application can realize fast search, fast speed and high efficiency, solves the problem of poor real-time query and low query efficiency of the traditional method, and can meet market demand.

[0063] In one embodiment, a vehicle information dictionary file and an inverted index file are constructed by an inverted hash index algorithm. By establishing an offline index file, an offline method is formed to query the vehicle, thereby achieving the purpose of real-time and accurate query and search of the vehicle.

[0064] In one embodiment, S20, based on the multiple vehicle information keywords corresponding to each vehicle image, a vehicle information dictionary file of multiple vehicle information keywords corresponding to each vehicle serial number is formed, including:

[0065] S210, compressing consecutive identical characters of identical vehicle information keywords corresponding to two adjacent vehicle serial numbers to obtain a plurality of compressed vehicle information keywords;

[0066] S220, forming a vehicle information dictionary file according to the plurality of compressed vehicle information keywords.

[0067] In this embodiment, two adjacent vehicle serial numbers can be understood as vehicle serial number 1 being adjacent to vehicle serial number 2, and vehicle serial number 2 being adjacent to vehicle serial number 3. The same vehicle information keywords can be understood as being the same license plate number, body color, vehicle model name, or vehicle type. For example: The vehicle model name corresponding to vehicle serial number 1 is FAW-Volkswagen, and the vehicle model name corresponding to vehicle serial number 2 is also FAW-Volkswagen. Then, the vehicle model name of vehicle serial number 2 is compressed for consecutive identical characters to form a compressed vehicle information keyword <consecutive identical character length, different character>, that is, <4, >. The license plate number of vehicle serial number 2 is Jing A23123. If the license plate number of vehicle serial number 3 is Jing A12345, then the license plate number of vehicle serial number 3 is compressed for consecutive identical characters to form <2, 12345>. If the license plate number of vehicle serial number 3 is Jing A23125, then the license plate number of vehicle serial number 3 is compressed for consecutive identical characters to form <6, 5>.

[0068] By performing consecutive identical character compression on the same vehicle information keywords corresponding to two adjacent vehicle serial numbers, the consecutive identical characters appearing in the vehicle information keywords can be omitted, thereby reducing the number of bytes required to store the vehicle information keywords, and thus being able to reduce the storage space of the vehicle information dictionary file.

[0069] In one embodiment, S30, according to the vehicle information dictionary file, establish an inverted index file, including:

[0070] S310, according to the vehicle information dictionary file of multiple vehicle information keywords corresponding to each vehicle serial number, obtain a frequency file of vehicle information keyword pointing frequency information, where the frequency information is the frequency of the vehicle information keyword appearing in the corresponding vehicle serial number;

[0071] S320, according to the vehicle information dictionary of multiple vehicle information keywords corresponding to each vehicle serial number, obtain a position file of vehicle information keyword pointing position information, where the position information is the position where the vehicle information keyword appears in the vehicle information dictionary file; among them, the vehicle information keyword, the frequency file, and the position file form an inverted index file.

[0072] In this embodiment, the vehicle information dictionary file is as follows:

[0073] Vehicle serial number 1: The vehicle information keywords are respectively [Jing A77777][black][Audi][compact car]; Vehicle serial number 2: The vehicle information keywords are respectively [Ji B88888][white][FAW-Volkswagen][SUV]; Vehicle serial number 3: The vehicle information keywords are respectively [Lu C99999][white][Mercedes-Benz][compact car].

[0074] Please refer to Figure 2 As shown in the table, taking the vehicle information keyword "compact car" as an example for illustration: The compact car appears once in each of vehicle serial number 1 and vehicle serial number 3, for a total of 2 times, and the corresponding occurrence frequency is 2, forming 1,3[2], which also forms the frequency file of the vehicle information keyword pointing to the vehicle serial number in step S310; The positions where the compact car appears in vehicle serial number 1 and vehicle serial number 3 are respectively the positions of the 4th vehicle information keyword, forming 4,4, which also forms the position file of the vehicle information keyword pointing to the position information in step S320. The inverted index file stores each vehicle information keyword and also retains pointers to the frequency file and the position file. Through the pointers, the corresponding frequency information and position information of the vehicle information keyword can be found. By searching for the corresponding frequency file and position file of the vehicle information keyword in the inverted index file, the vehicle serial number corresponding to the vehicle information keyword and the position where it appears in the vehicle serial number can be obtained. Furthermore, through the vehicle serial number, other corresponding vehicle information keywords can be obtained from the vehicle information dictionary file, thereby obtaining the vehicle information.

[0075] By using an inverted index to establish an inverted index file and querying the vehicle information keyword in the inverted index file, the corresponding frequency information and position information are obtained. Furthermore, the corresponding vehicle serial number and other relevant vehicle information keywords are obtained, and the query result is returned. By establishing the association relationship among the vehicle information keyword, the inverted index file, the vehicle serial number, and the vehicle information dictionary file, the vehicle information can be quickly queried, and the query can be achieved within milliseconds.

[0076] In one embodiment, S30, according to the vehicle information dictionary file, establishing an inverted index file further includes:

[0077] S330, performing continuous identical character compression on adjacent and identical vehicle information keywords in the inverted index file to obtain multiple compressed vehicle information keywords;

[0078] S340, forming an inverted index file according to the multiple compressed vehicle information keywords, the frequency file, and the position file.

[0079] In this embodiment, the inverted index file includes vehicle information keywords, a frequency file, and a position file. For adjacent and identical characters in the vehicle information keywords, compression can be performed. For example: If two adjacent vehicle information keywords are both "FAW-Volkswagen", then the latter "FAW-Volkswagen" is subjected to continuous identical character compression, forming a compressed vehicle information keyword <length of continuous identical characters, different characters>, that is, <4, >. If two adjacent vehicle information keywords are license plate number Jing A23123 and license plate number Jing A12345 respectively, then the latter license plate number is subjected to continuous identical character compression, forming <2, 12345>.

[0080] By compressing consecutive identical characters of adjacent and identical vehicle information keywords in this embodiment, the size of the inverted index file is reduced, the query work for all vehicle data in the database is reduced, the amount of calculation is reduced, and the real-time performance is improved.

[0081] In one embodiment, S50, querying the keyword to be tested according to the inverted index file to obtain vehicle information corresponding to the image of the vehicle to be tested includes:

[0082] S510, searching the inverted index file for frequency information and position information corresponding to the keyword to be tested according to a binary search algorithm;

[0083] S520, obtaining vehicle information corresponding to the vehicle image to be tested according to the frequency information and position information corresponding to the keyword to be tested.

[0084] In this embodiment, a binary search algorithm is used to quickly locate the keywords to be tested in the inverted index file to realize vehicle library query. When a vehicle passes through a checkpoint or a high-position video acquisition device, the front-end recognition system can identify the license plate number, body color, vehicle model name, vehicle type and other keywords in the vehicle image.

[0085] By binary searching the keyword to be tested in the inverted index file that has been formed, the corresponding frequency information can be queried through the pointer of the keyword to be tested to the frequency file. The frequency information is the frequency of the vehicle information keyword appearing in the corresponding vehicle serial number, and then the vehicle serial number corresponding to the keyword to be tested can be known. According to the vehicle serial number, the corresponding other vehicle information keywords can be obtained. The specific location of the corresponding appearance can be queried through the pointer of the keyword to be tested to the position file. Through the frequency information and position information queried by the keyword to be tested in the inverted index file, the corresponding vehicle information can be quickly queried and the query results are returned.

[0086] In one embodiment, the vehicle information query method further includes:

[0087] S60: If the keyword to be tested cannot be found according to the inverted index file, the vehicle information dictionary file and the inverted index file are updated according to the keyword to be tested.

[0088] In this embodiment, when a new vehicle information keyword is identified in a vehicle image, the vehicle information dictionary file and the inverted index file are updated offline, thereby updating the frequency information and the location information. Therefore, the vehicle information query method provided by the present application can regularly update the vehicle database offline, and quickly and efficiently query vehicle information.

[0089] See also Figure 3The present application provides a vehicle information identification device 100 including a data acquisition module 10, a vehicle information dictionary file forming module 20, an inverted index file forming module 30, a keyword acquisition module 40 and a vehicle information acquisition module 50. The data acquisition module 10 is used to obtain multiple vehicle images, and obtain multiple vehicle information keywords corresponding to each vehicle image based on the multiple vehicle images. The vehicle information dictionary file forming module 20 is used to form a vehicle information dictionary file of multiple vehicle information keywords corresponding to each vehicle serial number based on the multiple vehicle information keywords corresponding to each vehicle image. The inverted index file forming module 30 is used to establish an inverted index file based on the vehicle information dictionary file. The keyword acquisition module 40 is used to obtain a vehicle image to be tested, and obtain the corresponding multiple keywords to be tested based on the vehicle image to be tested. The vehicle information acquisition module 50 is used to query the keywords to be tested based on the inverted index file to obtain the vehicle information corresponding to the vehicle image to be tested.

[0090] In this embodiment, the relevant description of the data acquisition module 10 can refer to the relevant description of S10 in the above embodiment. The relevant description of the vehicle information dictionary file formation module 20 can refer to the relevant description of S20 in the above embodiment. The relevant description of the inverted index file formation module 30 can refer to the relevant description of S30 in the above embodiment. The relevant description of the keyword acquisition module 40 to be tested can refer to the relevant description of S40 in the above embodiment. The relevant description of the vehicle information acquisition module 50 can refer to the relevant description of S50 in the above embodiment.

[0091] In one embodiment, the inverted index file forming module 30 includes a frequency file generating module (not shown in the figure) and a position file generating module (not shown in the figure). The frequency file generating module is used to obtain a frequency file of vehicle information keywords pointing to frequency information based on a vehicle information dictionary file of multiple vehicle information keywords corresponding to each vehicle serial number, and the frequency information is the frequency of the vehicle information keywords appearing in the corresponding vehicle serial number. The position file generating module is used to obtain a position file of vehicle information keywords pointing to position information based on a vehicle information dictionary file of multiple vehicle information keywords corresponding to each vehicle serial number, and the position information is the position where the vehicle information keywords appear in the vehicle information dictionary file; the vehicle information keywords, the frequency file and the position file form an inverted index file.

[0092] In this embodiment, the relevant description of the frequency file generation module can refer to the relevant description of S310 in the above embodiment. The relevant description of the position file generation module can refer to the relevant description of S320 in the above embodiment.

[0093] In one embodiment, the inverted index file forming module 30 further includes a character compression module (not shown in the figure) and a forming module (not shown in the figure). The character compression module is used to compress consecutive identical characters of adjacent and identical vehicle information keywords in the inverted index file to obtain a plurality of compressed vehicle information keywords. The forming module is used to form an inverted index file according to the plurality of compressed vehicle information keywords, the frequency file, and the position file.

[0094] In this embodiment, the description of the character compression module can refer to the description of S330 in the above embodiment. The description of the formation module can refer to the description of S340 in the above embodiment.

[0095] In one embodiment, the vehicle information acquisition module 50 includes a frequency position information acquisition module (not shown in the figure) and an information acquisition module (not shown in the figure). The frequency position information acquisition module is used to query the frequency information and position information corresponding to the keyword to be tested in the inverted index file according to the binary search algorithm. The information acquisition module is used to obtain the vehicle information corresponding to the vehicle image to be tested according to the frequency information and position information corresponding to the keyword to be tested.

[0096] In this embodiment, the relevant description of the frequency position information acquisition module can refer to the relevant description of S510 in the above embodiment. The relevant description of the information acquisition module can refer to the relevant description of S520 in the above embodiment.

[0097] In one embodiment, the vehicle information query device 100 further includes an updating module 60. The updating module 60 is used to update the vehicle information dictionary file and the inverted index file according to the keyword to be tested if the keyword to be tested cannot be found according to the inverted index file.

[0098] In this embodiment, the relevant description of the updating module 60 may refer to the relevant description of S60 in the above embodiment.

[0099] In one embodiment, in the data acquisition module 10, multiple vehicle images are recognized according to a deep learning recognition algorithm to obtain multiple vehicle information keywords corresponding to each vehicle image.

[0100] In this embodiment, the method for using a deep learning recognition algorithm to recognize a vehicle image can refer to the relevant description in the above embodiment.

[0101] In the above-mentioned various embodiments, the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process can be rearranged without departing from the scope of protection of the present disclosure. The attached method claims provide the elements of various steps in an exemplary order and are not intended to be limited to the specific order or hierarchy described.

[0102] Those skilled in the art may also understand that the various illustrative logical blocks, modules and steps listed in the embodiments of the present application may be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly demonstrate the interchangeability of hardware and software, the various illustrative components, modules and steps described above have generally described their functions. Whether such functions are implemented by hardware or software depends on the specific application and the design requirements of the entire system. Those skilled in the art may use various methods to implement the described functions for each specific application, but such implementation should not be understood as exceeding the scope of protection of the embodiments of the present application.

[0103] The various illustrative logic blocks or modules described in the embodiments of the present application can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or the design of any combination of the above functions. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any traditional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.

[0104] The steps of the method or algorithm described in the embodiments of the present application can be directly embedded in the software module executed by the hardware, the processor, or a combination of the two. The software module can be stored in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM or other storage media of any form in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Alternatively, the storage medium can also be integrated into the processor. The processor and the storage medium can be arranged in an ASIC, and the ASIC can be arranged in a user terminal. Optionally, the processor and the storage medium can also be arranged in different components in the user terminal.

[0105] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A vehicle information query method, characterized in that: include: Obtaining a plurality of vehicle images, and obtaining a plurality of vehicle information keywords corresponding to each of the vehicle images based on the plurality of vehicle images; According to the plurality of vehicle information keywords corresponding to each of the vehicle images, a vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number is formed; Creating an inverted index file based on the vehicle information dictionary file; Acquire a vehicle image to be tested, and obtain a plurality of corresponding keywords to be tested according to the vehicle image to be tested; Query the keyword to be tested according to the inverted index file to obtain vehicle information corresponding to the vehicle image to be tested; The step of establishing an inverted index file according to the vehicle information dictionary file includes: According to the vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, a frequency file of the vehicle information keyword pointing to frequency information is obtained, wherein the frequency information is the frequency of the vehicle information keyword appearing in the corresponding vehicle serial number; According to the vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, obtaining a position file of the position information pointed to by the vehicle information keyword, wherein the position information is the position where the vehicle information keyword appears in the vehicle information dictionary file; The vehicle information keywords, the frequency file, and the position file form the inverted index file; The step of establishing an inverted index file according to the vehicle information dictionary file further includes: Compressing consecutive identical characters of adjacent and identical vehicle information keywords in the inverted index file to obtain a plurality of compressed vehicle information keywords; forming the inverted index file according to the plurality of compressed vehicle information keywords, the frequency file and the position file; The querying the keyword to be tested according to the inverted index file to obtain vehicle information corresponding to the vehicle image to be tested includes: According to a binary search algorithm, query the frequency information and position information corresponding to the keyword to be tested in the inverted index file; The vehicle information corresponding to the vehicle image to be tested is obtained according to the frequency information and position information corresponding to the keyword to be tested.

2. The vehicle information query method according to claim 1, characterized in that: The vehicle information query method further includes: If the keyword to be tested cannot be found according to the inverted index file, the vehicle information dictionary file and the inverted index file are updated according to the keyword to be tested.

3. The vehicle information query method according to claim 1, characterized in that: In the step of obtaining a plurality of vehicle images and obtaining a plurality of vehicle information keywords corresponding to each of the vehicle images based on the plurality of vehicle images, the plurality of vehicle images are recognized according to a deep learning recognition algorithm to obtain a plurality of vehicle information keywords corresponding to each of the vehicle images.

4. A vehicle information query device, characterized in that: include: A data acquisition module, used to obtain a plurality of vehicle images, and obtain a plurality of vehicle information keywords corresponding to each of the vehicle images according to the plurality of vehicle images; A vehicle information dictionary file forming module, used to form a vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number according to the plurality of vehicle information keywords corresponding to each vehicle image; An inverted index file forming module, used to create an inverted index file according to the vehicle information dictionary file; A keyword acquisition module for obtaining a vehicle image to be tested, and obtaining a plurality of corresponding keywords to be tested according to the vehicle image to be tested; A vehicle information acquisition module, used to query the keyword to be tested according to the inverted index file to obtain vehicle information corresponding to the vehicle image to be tested; The inverted index file forming module includes: A frequency file generating module, for obtaining, according to a vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, a frequency file of frequency information pointing to the vehicle information keyword, wherein the frequency information is the frequency at which the vehicle information keyword appears in the corresponding vehicle serial number; A position file generating module, for obtaining, according to the vehicle information dictionary file of the plurality of vehicle information keywords corresponding to each vehicle serial number, a position file of the vehicle information keyword pointing to the position information, wherein the position information is the position where the vehicle information keyword appears in the vehicle information dictionary file; the vehicle information keyword, the frequency file and the position file form the inverted index file; The inverted index file forming module also includes: A character compression module, used for compressing consecutive identical characters of adjacent and identical vehicle information keywords in the inverted index file to obtain a plurality of compressed vehicle information keywords; A forming module, configured to form the inverted index file according to the plurality of compressed vehicle information keywords, the frequency file and the position file; The vehicle information acquisition module includes: A frequency position information acquisition module, used to query the frequency information and position information corresponding to the keyword to be tested in the inverted index file according to a binary search algorithm; The information acquisition module is used to obtain the vehicle information corresponding to the vehicle image to be tested according to the frequency information and position information corresponding to the keyword to be tested.

5. The vehicle information query device according to claim 4, characterized in that: The vehicle information query device also includes: An updating module is used to update the vehicle information dictionary file and the inverted index file according to the keyword to be tested if the keyword to be tested cannot be found according to the inverted index file.

6. The vehicle information query device according to claim 4, characterized in that: In the data acquisition module, the multiple vehicle images are recognized according to a deep learning recognition algorithm to obtain multiple vehicle information keywords corresponding to each of the vehicle images.

Images