CN111860512B - Vehicle identification method, device, electronic equipment and computer readable storage medium - Google Patents
Vehicle identification method, device, electronic equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN111860512B CN111860512B CN202010115418.2A CN202010115418A CN111860512B CN 111860512 B CN111860512 B CN 111860512B CN 202010115418 A CN202010115418 A CN 202010115418A CN 111860512 B CN111860512 B CN 111860512B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- license plate
- character
- detected
- characters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 10
- 238000001514 detection method Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 102000006822 Agouti Signaling Protein Human genes 0.000 description 1
- 108010072151 Agouti Signaling Protein Proteins 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Traffic Control Systems (AREA)
Abstract
The application provides a vehicle identification method, a device, electronic equipment and a computer readable storage medium, wherein the method comprises the following steps: obtaining a target image and carrying out license plate recognition on the target image to obtain a license plate recognition result; if no license plate is identified in the license plate identification result, carrying out appearance feature identification on the vehicle to be detected to obtain vehicle identification information of the vehicle to be detected; if the license plate recognition result comprises at least part of characters of the license plate, vehicle recognition is carried out according to at least part of characters of the license plate and pre-stored vehicle information of the known vehicle, and vehicle recognition information of the vehicle to be detected is obtained, wherein the vehicle information comprises vehicle recognition information of the known vehicle, the known license plate and appearance characteristics. According to the scheme, the vehicle license plate recognition is advanced, and when at least part of characters in the license plate are recognized, the recognized part of license plate characters and the pre-stored vehicle recognition information are adopted for vehicle recognition, so that the accuracy of vehicle recognition can be improved.
Description
Technical Field
The present application relates to the field of vehicle identification technologies, and in particular, to a vehicle identification method, device, electronic apparatus, and computer readable storage medium.
Background
At present, the concepts of smart cities and smart traffic have been raised, and a large number of cameras are arranged at intersections in the cities for the fields of automatic driving, video monitoring and the like. The need for tracking hit vehicles, suspicious vehicles, etc. has prompted the study of the vehicle Re-ID topic. Since the characteristics of the vehicles of the same train are very similar, there is caused a problem that the recognition result of the vehicles is inaccurate.
Disclosure of Invention
In view of the above, an object of the present application is to provide a vehicle recognition method, apparatus, electronic device, and computer-readable storage medium, which can solve the problem of difficulty in recognizing a same vehicle in the prior art by narrowing down the range of vehicle appearance feature recognition when detecting a part of characters in a license plate, thereby achieving the effect of improving the accuracy of vehicle recognition.
According to one aspect of the application, an electronic device is provided that may include a storage medium and a processor in communication with the storage medium. The storage medium stores machine-readable instructions executable by the processor. When the electronic device is in operation, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the following operations:
Acquiring a target image and carrying out license plate recognition on the target image to acquire a license plate recognition result, wherein the target image is an image of a vehicle to be detected;
if no license plate is identified in the license plate identification result, carrying out appearance feature identification on the vehicle to be detected to obtain vehicle identification information of the vehicle to be detected;
and if the license plate recognition result comprises at least part of characters of a license plate, carrying out vehicle recognition according to the at least part of characters of the license plate and pre-stored vehicle information of the known vehicle to obtain the vehicle recognition information of the vehicle to be detected, wherein the vehicle information comprises the vehicle recognition information of the known vehicle, the known license plate and the appearance characteristics.
In some embodiments, the step of obtaining the vehicle identification information of the vehicle to be detected includes:
taking at least part of characters of the license plate as first license plate characters;
judging whether the first license plate character contains a complete character of a license plate or not;
if the first license plate character contains part of characters of the license plate, taking a known vehicle containing the first license plate character as a pre-matched vehicle;
For each pre-matched vehicle, calculating a first matching score of the pre-matched vehicle and the vehicle to be detected according to the appearance characteristics of the pre-matched vehicle;
acquiring vehicle identification information of the vehicle to be detected according to the first matching score of each pre-matched vehicle;
and if the first license plate character contains the complete character of the license plate, acquiring the vehicle identification information of the vehicle to be detected according to a matching result of whether the first license plate character is consistent with the known license plate.
In some embodiments, the step of calculating, for each of the pre-matched vehicles, a first match score for the pre-matched vehicle and the vehicle to be detected based on the profile features of the pre-matched vehicle includes:
for each pre-matched vehicle, acquiring a first weight corresponding to the pre-matched vehicle according to the first license plate character;
calculating a first appearance score of the pre-matched vehicle according to the appearance characteristics of the pre-matched vehicle;
and calculating the product of the first weight and the first appearance score to obtain a first matching score of the pre-matching vehicle and the vehicle to be detected.
In some embodiments, the step of obtaining, for each of the pre-matched vehicles, a first weight corresponding to the pre-matched vehicle according to the first license plate character includes:
Judging whether a first character which is the same as characters in a preset character set exists in the first car character, wherein the preset character set comprises characters with various similar shapes;
taking other characters except the first character in the first vehicle character as second characters;
and calculating the sum of the character sub-weights of each character in the first car character to obtain the first weight, wherein the character sub-weight of the second character is larger than the character sub-weight of the first character.
In some embodiments, the step of acquiring the vehicle identification information of the vehicle to be detected according to the first matching score of each of the pre-matched vehicles includes:
screening the first match scores of each pre-matched vehicle for a first match score greater than a confidence threshold;
and acquiring vehicle identification information of the vehicle to be detected according to the first matching score larger than the confidence coefficient threshold value.
In some embodiments, the step of obtaining the vehicle identification information of the vehicle to be detected according to the matching result of the first license plate character and the known license plate comprises:
comparing the first license plate character with each known license plate, and judging whether the known license plate same as the first license plate character exists or not;
If the known license plate which is the same as the first license plate character exists, the known license plate which is the same as the first license plate character is used as a matched license plate;
acquiring vehicle identification information corresponding to the matched license plate as vehicle identification information of a vehicle to be detected;
and if the known license plate which is the same as the first license plate character does not exist, carrying out appearance feature recognition on the vehicle to be detected, and acquiring vehicle recognition information of the vehicle to be detected.
In some embodiments, the step of obtaining the vehicle identification information of the vehicle to be detected according to the matching result of the first license plate character and the known license plate comprises:
matching the known license plates of the known vehicles according to the first license plate characters;
judging whether a known license plate matched with the first license plate character exists or not;
if a known license plate matched with the first license plate character exists, acquiring vehicle identification information of the vehicle to be detected from vehicle identification information corresponding to the known license plate;
and if the known license plate matched with the first license plate character does not exist, identifying the vehicle to be detected according to the appearance characteristics, and obtaining the vehicle identification information of the vehicle to be detected.
In some embodiments, the step of matching the known license plates of the respective known vehicles according to the first license plate character comprises:
judging whether the first character which is the same as the character in the preset character set exists in the first car character or not, wherein other characters except the first character in the first car character are second characters, and the preset character set comprises characters with various similar shapes;
if the first character exists in the first vehicle-mounted characters, acquiring a third character similar to the first character from the preset character set according to a preset corresponding relation between similar characters for each first character;
respectively replacing the first character with a third character similar to the first character aiming at each first character to obtain second car board characters with different combinations;
and matching the first license plate character and the second license plate character with the known license plate respectively.
In some embodiments, the step of obtaining a target image and performing license plate recognition on the target image, and obtaining a license plate recognition result includes:
and obtaining a target image, and carrying out license plate recognition on the target image by adopting a deep neural network to obtain a license plate recognition result.
In some embodiments, a vehicle identification apparatus, the apparatus comprising:
the license plate recognition module is used for acquiring a target image and carrying out license plate recognition on the target image to obtain a license plate recognition result, wherein the target image is an image of a vehicle to be detected;
the appearance recognition module is used for carrying out appearance feature recognition on the vehicle to be detected to obtain vehicle recognition information of the vehicle to be detected when the license plate is not recognized in the license plate recognition result;
and the range determining module is used for carrying out vehicle recognition according to at least part of characters of the license plate and pre-stored vehicle information of the known vehicle when the license plate recognition result comprises at least part of characters of the license plate, and obtaining the vehicle recognition information of the vehicle to be detected, wherein the vehicle information comprises the vehicle recognition information of the known vehicle, the known license plate and the appearance characteristics.
In some embodiments, an electronic device includes: a processor, a storage medium, and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating over the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the method according to any one of the present application.
Based on any one of the above aspects, in the vehicle detection method, the device, the electronic equipment and the readable storage medium provided by the application, firstly, the license plate of the vehicle to be detected is detected, then the vehicle to be detected is identified according to the identification result of the license plate, under the condition that the license plate is not identified, the vehicle to be detected is identified according to the appearance characteristics of the vehicle, when part of characters in the license plate of the vehicle to be detected are identified, the known vehicle is acquired according to the identified characters, the range of the identified vehicle is reduced, and then the matching is performed, so that the accuracy of the identification result of the vehicle to be detected can be improved.
In addition, in some embodiments, when at least part of characters in the license plate of the vehicle to be detected are identified, the vehicle identification information of the vehicle to be detected is obtained according to the situation of the identified at least part of characters, so that the accuracy of the identification result of the vehicle to be detected can be further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 shows a schematic architecture diagram of a service system according to an embodiment of the present application;
fig. 2 is a schematic block diagram of an electronic device according to an embodiment of the present application;
FIG. 3 is a flowchart showing a vehicle identification method according to an embodiment of the present application;
fig. 4 shows a flowchart of a specific method of step S140 in the vehicle identification method according to the embodiment of the present application;
fig. 5 shows a flowchart of a specific method of step S144 in the vehicle identification method according to the embodiment of the present application;
fig. 6 shows a flowchart of a specific method of step S146 in the vehicle identification method according to the embodiment of the present application;
fig. 7 is a schematic structural diagram of a vehicle identification device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for the purpose of illustration and description only and are not intended to limit the scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this disclosure, illustrates operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to or removed from the flow diagrams by those skilled in the art under the direction of the present disclosure.
In addition, the described embodiments are only some, but not all, embodiments of the application. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present application.
In order to enable one skilled in the art to use the present disclosure, the following embodiments are presented in connection with a specific application scenario "cross-border tracking". It will be apparent to those having ordinary skill in the art that the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the application. Although the application is described primarily in the context of cross-border tracking, it should be understood that this is but one exemplary embodiment.
It should be noted that the term "comprising" will be used in embodiments of the application to indicate the presence of the features stated hereafter, but not to exclude the addition of other features.
One aspect of the present application relates to a service system. The system can firstly collect an image to be detected, then identify the vehicle of the image to be detected, and reduce the vehicle range for matching the appearance characteristics by detecting the license plate of the vehicle in the process of identifying the vehicle.
It should be noted that, prior to the application of the present application, the similarity between vehicles is generally calculated directly by using the shape features of the vehicles to identify the vehicles. However, the vehicle identification scheme provided by the application can be used for vehicle identification by combining the appearance characteristics and the license plate. Therefore, through the identified license plate, the vehicle identification scheme can provide more accurate vehicle identification results.
Fig. 1 is a schematic architecture diagram of a service system 10 according to an embodiment of the present application. For example, the service system 10 may be an online transport service platform for a transport service such as a taxi, a ride service, a express, a carpool, a bus service, a driver rental, or a class service, or any combination thereof. Service system 10 may include one or more of server 100, network 200, service request terminal 300, service provider 400, and database 500.
Referring to fig. 2, in some embodiments, the server 100 may include a processor 110. Processor 110 may process information and/or data related to a service request to perform one or more of the functions described herein. For example, the processor 110 may determine the target vehicle based on the service request obtained from the service requester 300. In some embodiments, processor 110 may include one or more processing cores (e.g., a single core processor (S) or a multi-core processor (S)). By way of example only, the Processor 110 may include a central processing unit (Central Processing Unit, CPU), application specific integrated circuit (Application Specific Integrated Circuit, ASIC), special instruction set Processor (Application Specific Instruction-set Processor, ASIP), graphics processing unit (Graphics Processing Unit, GPU), physical processing unit (Physics Processing Unit, PPU), digital signal Processor (Digital Signal Processor, DSP), field programmable gate array (Field Programmable Gate Array, FPGA), programmable logic device (Programmable Logic Device, PLD), controller, microcontroller unit, reduced instruction set computer (Reduced Instruction Set Computing, RISC), microprocessor, or the like, or any combination thereof.
In one embodiment, the server 100 may also include a memory 120.
In some embodiments, the device type corresponding to the service request end 300 and the service providing end 400 may be a mobile device, for example, may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, or an augmented reality device, and may also be a tablet computer, a laptop computer, or a built-in device in a motor vehicle, and so on.
In some embodiments, database 500 may be connected to network 200 to communicate with one or more components in service system 10 (e.g., server 100, service requester 300, service provider 400, etc.). One or more components in service system 10 may access data or instructions stored in database 500 via network 200. In some embodiments, database 500 may be directly connected to one or more components in service system 10, or database 500 may be part of server 100.
The vehicle identification method provided by the embodiment of the present application will be described in detail with reference to the description of the service system 10 shown in fig. 1.
Referring to fig. 3, a flow chart of a vehicle identification method according to an embodiment of the present application is shown, and the method may be executed by a server in the service system 10, where the server stores vehicle information of a plurality of vehicles, where the vehicle information includes vehicle identification information of a known vehicle, a known license plate, and an appearance feature; or, the server is in communication connection with a cloud platform or a storage device, and the like, and the cloud platform or the storage device stores vehicle information of a plurality of vehicles, wherein the vehicle information comprises vehicle identification information of known vehicles, known license plates and appearance characteristics, and in this case, if the server is to execute the method of the embodiment, the vehicle section information can be acquired from the cloud platform or the storage device. The vehicle identification information is information for uniquely identifying the vehicle. The specific implementation process of the vehicle identification method comprises the steps of S110-S140. For ease of understanding, the following describes step S110 to step S140 in detail.
Step S110, obtaining a target image and carrying out license plate recognition on the target image to obtain a license plate recognition result, wherein the target image is an image of a vehicle to be detected.
Step S120, determining whether the license plate recognition result includes at least part of characters of the license plate.
And step S130, if no license plate is identified in the license plate identification result, carrying out appearance feature identification on the vehicle to be detected to obtain vehicle identification information of the vehicle to be detected.
And step S140, if the license plate recognition result comprises at least part of characters of a license plate, recognizing the vehicle to be detected according to at least part of the characters of the license plate and the known vehicle information.
Specifically, if the license plate recognition result includes at least part of characters of a license plate, vehicle recognition is performed according to the at least part of characters of the license plate and pre-stored vehicle information of a known vehicle, and the vehicle recognition information of the vehicle to be detected is obtained.
In this embodiment, the license plate of the vehicle to be detected is detected first, then the vehicle to be detected is identified according to the detection result of the license plate, the appearance feature of the vehicle is identified under the condition that the license plate is not detected, so as to identify the vehicle to be detected, when part of characters in the license plate of the vehicle to be detected are detected, the vehicle is identified according to the identified characters and the known vehicle, so that the range of the known vehicle in the identification process is reduced, and the appearance feature and the license plate are combined to identify the vehicle, so that the accuracy of the identification result of the vehicle to be detected can be improved. The method for license plate recognition in this embodiment may be implemented by using an existing method for license plate recognition.
Referring to fig. 4, in the present embodiment, step S140 includes sub-step S141-sub-step S146.
And step S141, taking at least part of characters of the license plate as first license plate characters.
Step S142, determining whether the first license plate character includes a complete character of the license plate.
Step S143, if a part of the characters of the license plate is included in the first license plate characters, using the known vehicle including the first license plate characters as a pre-matched vehicle.
Step S144, a first match score is calculated for each pre-matched vehicle.
Specifically, for each pre-matching vehicle, calculating a first matching score of the pre-matching vehicle and the vehicle to be detected according to the appearance characteristics of the pre-matching vehicle.
Step S145, acquiring vehicle identification information of the vehicle to be detected according to the first matching score.
Specifically, vehicle identification information of the vehicle to be detected is obtained according to the first matching score of each pre-matched vehicle.
Step S146, if the first license plate character includes a complete character of the license plate, acquiring vehicle identification information of the vehicle to be detected according to the first license plate character.
Specifically, if the first license plate character contains the complete character of the license plate, vehicle identification information of the vehicle to be detected is obtained according to a matching result of whether the first license plate character is consistent with the known license plate.
In this embodiment, different recognition measures are adopted according to the detected character condition on the license plate, when only part of characters of the license plate are included in the detected characters of the license plate, the recognition is performed according to the appearance characteristics of the vehicle, and when the complete license plate characters are detected, the vehicle recognition is performed according to the recognized license plate characters, so that the vehicle recognition result is more accurate.
In this embodiment, the first matching score characterizes the similarity between the pre-matched vehicle and the vehicle to be detected, e.g., the higher the first matching score, the higher the similarity between the pre-matched vehicle and the vehicle to be detected. In specific implementation, the first matching score can be calculated through a deep neural network, and when the first matching score is calculated, the deep neural network can firstly extract the appearance characteristics of the pre-matched vehicle and then calculate the similarity between the appearance characteristics of the vehicle to be detected and the appearance characteristics of the pre-matched vehicle. In the identifying process, a confidence threshold value may be set, and then vehicle identification information of the vehicle to be detected is determined from vehicles corresponding to the first matching score with the similarity (first matching score) larger than the confidence threshold value, for example, the vehicle identification information with the highest first matching score in known vehicles with the similarity with the vehicle to be detected larger than the confidence threshold value is used as the vehicle identification information of the vehicle to be detected, so that the vehicle identification is completed.
In this embodiment, the step of obtaining the vehicle identification information of the vehicle to be detected according to each first matching score may be implemented by first obtaining the similarity between the vehicle to be detected and each pre-matching vehicle as a first matching score by using a deep neural network, then judging whether there is a first matching score greater than a confidence threshold value in the first matching scores between the vehicle to be detected and each pre-matching vehicle, and if there is a first matching score greater than the confidence threshold value in one or more pre-matching vehicles, outputting the vehicle identification information of the pre-matching vehicle corresponding to the largest first matching score among the first matching scores greater than the confidence threshold value as the vehicle identification information of the vehicle to be detected.
In this embodiment, the similarity between each pre-matched vehicle and the vehicle to be detected may be calculated according to the appearance characteristics of the vehicle, and the vehicle is identified by combining the appearance characteristics of the vehicle, so that the identification result may be more accurate.
Referring to fig. 5, in the present embodiment, step S144 includes sub-steps S1441-S1443.
Step S1441, for each pre-matched vehicle, acquiring a first weight corresponding to the pre-matched vehicle according to the first license plate character.
Step S1442, calculating a first appearance score of the pre-matched vehicle according to the appearance characteristics of the pre-matched vehicle.
Step S1443, calculating the product of the first weight and the first appearance score to obtain a first matching score.
Specifically, a product of the first weight and the first appearance score is calculated, and a first matching score of the pre-matching vehicle and the vehicle to be detected is obtained.
The embodiment is specifically used for calculating a first matching score, namely similarity, between the pre-matching vehicle and the vehicle to be detected. In the calculation process, the appearance characteristics of the vehicle are associated with the detected first license plate characters, so that a more accurate calculation result of the first matching score can be obtained.
In this embodiment, when the license plate is identified, the area corresponding to each character can be identified, and when only a part of characters are detected, the position of the characters in the license plate can be determined.
For example, if two characters 5 and 1 of the license plate "Beijing B25172" are not detected, the positions of the characters 5 and 1 can be detected and the positions of other characters in the license plate can be detected at the same time when the license plate is detected, and then, when the pre-matched vehicle is acquired, the acquisition can be performed according to the positions of the detected characters. For example, when the first license plate character is "Beijing" in the "Beijing B25172" license plate, the "Beijing" in the "Beijing B25172" license plate is the first character, the "B" character is the second character, the "third" character 2 "in the" Beijing B25172 "character is the fourth" character in the "Beijing B25172" character is the fifth "character in the" Beijing B25172 "character is the sixth" character in the "Beijing B25172" character is the sixth "character in the" Beijing B25172 "character is the seventh" character in the "Beijing B25172" character, the "sixth" character 2 "character in the" Beijing B25172 "character is the ninth" character, the "undetected" character "5" character is the third "character in the actual license plate, and the" undetected "character" 1 "character is the eighth" character in the actual license plate. At this time, the first character of the corresponding license plate of all the pre-matched vehicles is the character "Beijing", the second character is "B", the fourth character is "2", the fifth character is "5", the sixth character is "1", the seventh character is "7", the ninth character is "2", and the third character and the eighth character are one of all the letters or numbers that can be used to construct the license plate.
Alternatively, in this embodiment, step S1411 includes first determining whether a first character identical to a character in a preset character set including a plurality of characters having similar shapes exists in the first vehicle character.
For example, among the characters employed in the license plate, confusion may occur between the characters "0" and "Q", between the characters "4" and "a", between the characters "8" and "B", between the characters "3" and "8", between the characters "2" and "Z", and thus the characters having similar shapes in the present embodiment may be the character sets of the characters "0" and "Q", the characters "4" and "a", the characters "8" and "B", the characters "3" and "8", that is, the preset character set may contain the characters "0", the characters "Q", the characters "4", the characters "a", the characters "8", the characters "B", the characters "3", the characters "2" and the characters "Z".
Of course, in this embodiment, the preset character set may include only a part of the characters, and the preset character set may include other characters with higher similarity that may form the license plate. It is understood that the types of characters contained in the preset character set may be set according to actual needs. After the first character in the first vehicle card characters is detected, other characters except the first character in the first vehicle card characters can be used as second characters; and finally, calculating the sum of the character sub-weights of each character in the first car character to obtain a first weight, wherein in one possible implementation mode, the character sub-weight of the second character is larger than the character sub-weight of the first character. In another possible embodiment, the character sub-weights of each first character may be the same, and the character sub-weights of each second character may be the same; the character sub-weights of the first character and the second character may also be the same. The sum of the character sub-weights in the complete license plate in the vehicle to be detected may be set to 1.
Taking the example that the weights of the first characters and the second characters are different, for example, when two first characters and two second characters are included in the license plate, then the weight of the first character may be 0.1, and the weight of the second character may be 0.4; alternatively, the weight of the first character may be 0.2 and the weight of the second character may be 0.3.
It can be understood that, in this embodiment, when the scheme is applied to an electronic device, a preset character set may be stored in the electronic device, and a correspondence relationship between each character in the preset character set (which characters are similar characters) may also be stored in the electronic device. In one possible implementation, the preset character set and the data in the preset character set may also be stored on separate readable storage media.
The method and the device are used for specifically calculating the first weight of the license plate of the vehicle to be detected, and the more characters are included in the characters of the first license plate, the greater the possibility that the obtained pre-matched vehicle is the vehicle to be detected, so that the influence of the number of the detected characters on the vehicle recognition result is considered in the calculation process of the first matching score, the more accurate first weight is calculated, and the accuracy of the vehicle recognition result can be improved. The different types of characters are given different weights, so that the influence of the different characters in the vehicle on the final vehicle recognition result can be represented in the calculated first weight, and the accuracy of the vehicle recognition result can be further improved.
In one embodiment, please continue with fig. 6, optionally, step S146 includes sub-step S1461-sub-step S1464.
Step S1461, determining whether there is a known license plate identical to the first license plate character.
Specifically, the first license plate character is compared with each known license plate, and whether the known license plate same as the first license plate character exists or not is judged.
Step S1462, if there is a known license plate identical to the first license plate character, using the known license plate identical to the first license plate character as a matching license plate.
Step S1463, acquiring vehicle identification information of the vehicle to be detected according to the matched license plate.
Specifically, vehicle identification information corresponding to the matched license plate is obtained as vehicle identification information of the vehicle to be detected.
Step S1464, if there is no known license plate identical to the first license plate character, acquiring vehicle identification information of the vehicle to be detected according to the appearance feature.
Specifically, if the known license plate same as the first license plate character does not exist, carrying out appearance feature recognition on the vehicle to be detected, and acquiring vehicle recognition information of the vehicle to be detected.
The embodiment is used for obtaining the vehicle identification information of the vehicle to be detected according to the condition that the detected license plate number is matched with the known vehicle information when the complete license plate number is detected from the vehicle to be detected.
In another embodiment, optionally, step S146 includes, first, matching known license plates of respective known vehicles according to the first license plate character; then, judging whether a known license plate matched with the first license plate character exists or not; if a known license plate matched with the first license plate character exists, acquiring vehicle identification information of the vehicle to be detected from vehicle identification information corresponding to the known license plate; and if the known license plate matched with the first license plate character does not exist, identifying the vehicle to be detected according to the appearance characteristics, and obtaining the vehicle identification information of the vehicle to be detected.
In the license plate detection process, the image acquisition device cannot be ensured to be an image which is right against the license plate and cannot be ensured to be clear when the image acquisition device acquires the license plate image, so that in the embodiment, after the complete license plate is identified according to the image of the vehicle to be detected, whether the known license plate matched with the first license plate character of the vehicle to be detected exists or not is judged, and when the known license plate matched with the first license plate character of the vehicle to be detected exists, the vehicle identification information of the vehicle corresponding to the matched known license plate is used as the vehicle identification information of the vehicle to be detected. And when the known license plate matched with the first license plate character of the vehicle to be detected does not exist, the vehicle to be detected is identified according to the appearance characteristics of the vehicle to be detected. The vehicle identification information of the vehicle to be detected is obtained in different modes according to the matching condition between the first vehicle character and the known license plate, so that the accuracy of the identification result of the vehicle to be detected can be ensured.
Optionally, in this embodiment, the step of matching the known license plates of the known vehicles according to the first license plate character may include: firstly judging whether a first character identical to a character in a preset character set exists in the first vehicle character, wherein other characters except the first character in the first vehicle character are second characters, the preset character set comprises characters with various similar shapes, for example, confusion is likely to occur among characters adopted by a vehicle license plate between characters '0' and 'Q', between characters '4' and 'A', between characters '8' and 'B', between characters '3' and '8', between characters '2' and 'Z', thus, the characters having similar shapes in the present embodiment may be the character "0" and the character "Q", the character "4" and the character "a", the character "8" and the character "B", the character "3" and the character "8", that is, the preset character set may include the character "0", the character "Q", the character "4", the character "a", the character "8", the character "B", the character "3", the character "2" and the character "Z". Of course, in this embodiment, the preset character set may only include a part of the characters, and the preset character set may also include other characters with higher shape similarity used in the license plate.
In this embodiment, the preset character set may be the same or different in the case that the first license plate character includes a partial character of the vehicle and includes a complete character of the vehicle.
After judging whether the first character which is the same as the character in the preset character set exists in the first car character, if the first character exists in the first car character, acquiring a third character which is similar to the first character from the preset character set according to a preset corresponding relation among similar characters for each first character; respectively replacing the first character with a third character similar to the first character aiming at each first character to obtain second car board characters with different combinations; and matching the first license plate character and the second license plate character with the known license plate respectively.
It can be understood that, in this embodiment, when the scheme is applied to an electronic device, a preset character set may be stored in the electronic device, and a correspondence relationship between each character in the preset character set (which characters are similar characters) may also be stored in the electronic device. In one possible implementation, the preset character set and the data in the preset character set may also be stored on separate readable storage media.
In this embodiment, the method is used for acquiring the second license plate character according to the detected first license plate character, that is, for acquiring other possible license plate numbers of the vehicle to be detected, so that further matching is performed according to the second license plate character. For example, when the license plate recognition is performed on the vehicle to be detected, the character "Q" is included in the license plate number of the vehicle to be detected, but since a part of the character "Q" is blocked, the original character "Q" is recognized as the character "0" in the recognized first license plate characters after the license plate recognition is performed on the vehicle to be detected, and at this time, the recognized character "0" can be replaced by the character "Q" so as to obtain the second license plate characters.
For another example, when the license plate recognition is performed on the vehicle to be detected, the character "a" is originally included in the license plate number of the vehicle to be detected, but because the inclination of the image of the license plate part of the vehicle to be detected is large, the original character "a" is recognized as the character "4" in the recognized first license plate character after the license plate recognition is performed on the vehicle to be detected, and at this time, the recognized character "4" (the character "4" in the first license plate character) can be replaced by the character "a", so that the second license plate character is obtained.
For another example, when the license plate recognition is performed on the vehicle to be detected, the character "B" is originally included in the license plate number of the vehicle to be detected, but because the inclination of the image of the license plate part of the vehicle to be detected is large, the original character "B" is recognized as the character "8" in the recognized first license plate characters after the license plate recognition is performed on the vehicle to be detected, and at this time, the recognized character "8" can be replaced by the character "B" or the character "3", so that two second license plate characters are obtained.
When the second vehicle license plate character is specifically obtained, the license plate combination after each first character is replaced by a third character similar to the first character can be obtained, so that the second vehicle license plate character corresponding to each combination is obtained. In license plate recognition, each first character should be replaced with a third character corresponding to the character.
Optionally, in this embodiment, the step of determining whether there is a known license plate matching the first license plate character includes first determining whether there is a known license plate identical to the first license plate character or any one of the second license plate characters; if the known license plate which is the same as the first license plate character or the arbitrary second license plate character exists, judging that the known license plate which is matched with the first license plate character exists, and taking the known license plate as a license plate to be matched; if there is no known license plate that is the same as the first license plate character or the any one of the second license plate characters, it is determined that there is no known license plate that matches the first license plate character.
In this embodiment, the method is used for identifying the vehicle to be detected according to the first license plate character and the second license plate character, and obtaining the vehicle identification information of the vehicle to be detected, so that the possible license plate number of the vehicle to be detected can be obtained under the condition that the first license plate character deviates from the character of the vehicle to be detected, the vehicle to be detected is identified, and the accuracy of vehicle detection is improved.
Still taking the example that the actual license plate number of the vehicle to be detected includes the character "Q", in the license plate recognition, the character "Q" is recognized as the character "0" in the recognized first license plate characters. At this time, if the comparison is made with the known license plate based on the first license plate character only, then there will be no known license plate identical to the first license plate character. At this time, if the second license plate character is obtained, the character "Q" in the first license plate character is replaced with the character "0", so that a second license plate character can be obtained, and at this time, the first license plate character and the second license plate character are respectively compared with each known license plate, and since the character "Q" in the first license plate character is incorrectly recognized as the character "0", and the character in the second license plate character is equivalent to correcting the incorrectly recognized character, the known license plate identical to the second license plate character can be obtained, and thus, the accuracy of vehicle recognition is improved.
In this embodiment, when the second license plate character is acquired, if a plurality of first characters are included in the license plate, the second license plate character may include a combination of possible substitution results for each of the first characters. For example, in a first card character, if character "a" and character "B" are included, then there will be 5 second card characters, respectively, character "a" is replaced with character "4", character "B" is replaced with character "8", character "B" is replaced with character "3", character "a" is replaced with character "4" while character "B" is replaced with character "8", character "a" is replaced with character "4", and character "B" is replaced with character "3".
To aid understanding, the following is illustrated in connection with a complete license plate number.
For example, when the first license plate character is "Beijing B4241732", the character "4" and the character "3" in the first license plate character are the first characters, and then after the character "4" and the character "3" in the license plate are detected, all the first characters in the license plate need to be replaced. And detecting the positions of other characters in the license plate, and then, when the license plate to be matched is acquired, acquiring according to the positions of the detected first characters and the detected second characters. For example, when the first vehicle character is "Beijing B4241732" is the first character, the "B" is the second character, the third character in "Beijing B4241732" is "4", "the fourth character in" Beijing B4241732 "is" 2"," the fifth character in "Beijing B4241732", "the sixth character in" Beijing B4241732 "is" 1"," the seventh character in "7" in "Beijing B4241732", "the eighth character in" 3 "in" Beijing B4241732", and the ninth character in" 2 "in" Beijing B4241732 ". At this time, the first character corresponding to all the pre-matched license plates is the character "Beijing", the second character is the "B", the fourth character is the "2", the sixth character is the "1", the seventh character is the "7", and the ninth character is the "2", so that the obtained license plates to be matched are various combinations after the second character, the fifth character and the eighth character are replaced by the corresponding third characters, that is, the first character, the second character, the fourth character, the sixth character, the seventh character and the ninth character are the same as those of the first license plates in the license plates to be matched.
In one possible implementation manner, the step of obtaining the vehicle identification information of the vehicle to be detected from the vehicle identification information corresponding to the known license plate includes firstly judging the number of the license plates to be matched with the first license plate character; when the number of the license plates to be matched is judged to be larger than two, calculating corresponding license plate scores according to characters contained in the license plates to be matched aiming at each license plate to be matched, wherein the license plate scores are the influence scores of license plate numbers on vehicle recognition results; then, calculating a corresponding second appearance score according to the appearance characteristics of the known vehicle corresponding to the license plate to be matched, wherein the second appearance score is a vehicle identification result score corresponding to the appearance characteristics of the vehicle; finally, calculating a second matching score according to the license plate score and the second appearance score; and acquiring the vehicle identification information of the known vehicle with the highest second matching score as the vehicle identification information of the vehicle to be detected.
In this embodiment, if the obtained plurality of license plates to be matched have the known license plates corresponding to the license plates, it is explained that a large part of the known vehicles corresponding to the known license plates are not vehicles to be detected, and in the known vehicles, there is a high possibility that the appearance characteristics of a part of the vehicles are very different from the appearance characteristics of the vehicles to be detected, so in this embodiment, the similarity between the vehicles to be detected and the known vehicles is evaluated by combining the license plate scores of the vehicles and the second appearance scores corresponding to the appearance characteristics, and the known vehicles with the large difference from the appearance characteristics of the vehicles to be detected are excluded, thereby further reducing the matching range of the appearance characteristics of the vehicles to be detected and improving the accuracy of the vehicle recognition result.
In one possible implementation manner, the second shape score may be calculated first, then the vehicles to be matched are further screened according to the second shape score, the range of the vehicles to be matched corresponding to the license plate to be matched is reduced, and then the vehicles to be detected are further identified according to the vehicles to be matched with the reduced range.
After the detection range of the vehicles is narrowed, judging the number of the vehicles to be matched, and calculating corresponding license plate scores according to characters contained in the license plates to be matched corresponding to the vehicles to be matched aiming at the vehicles to be matched when the number of the vehicles to be matched is greater than two; then, calculating a corresponding second appearance score according to the appearance characteristics of the known vehicle corresponding to the vehicle to be matched, wherein the second appearance score is a vehicle identification result score corresponding to the appearance characteristics of the vehicle; finally, calculating a second matching score according to the license plate score and the second appearance score; and acquiring the vehicle identification information of the known vehicle with the highest second matching score as the vehicle identification information of the vehicle to be detected.
Optionally, in this embodiment, the step of calculating, for each license plate to be matched, a corresponding license plate score according to the characters included in the license plate to be matched includes first obtaining a character weight and an initial score of each character in the license plate to be matched, where the character weight of the first character and the character weight of the third character are smaller than the character weight of the second character. And calculating license plate scores according to the initial scores of each character in the license plate to be matched and the character weights of the characters.
In this embodiment, different weights are given to different characters in the license plate to be matched, so that the influence of the confusable characters (characters with similar shapes) on the license plate scoring of the license plate can be reduced.
In another possible implementation manner, the step of obtaining the vehicle identification information of the vehicle to be detected from the vehicle identification information corresponding to the known license plate includes that the number of the license plates to be matched, which are matched with the first license plate characters, can be judged first; and when the number of the license plates to be matched is judged to be larger than two, calculating the similarity between the corresponding vehicles of each license plate to be matched and the vehicles to be detected according to the appearance characteristics of the vehicles, and finally, taking the vehicle identification information of the known vehicle with the largest similarity as the vehicle identification information of the vehicles to be detected, so that the vehicle identification is carried out by combining the appearance characteristics, and the obtained vehicle identification result is more accurate.
In another possible embodiment, the step of obtaining the vehicle identification information of the vehicle to be detected from the vehicle identification information corresponding to the known license plate may include first determining the number of license plates to be matched that are matched with the first license plate character; when the number of the license plates to be matched is judged to be more than two, acquiring a first license plate image of each license plate to be detected for each license plate to be matched; then, when the number of license plates to be matched is greater than two, obtaining second license plate images of known vehicles corresponding to the license plates to be matched, and detecting the similarity of the first license plate images and the second license plate images; and finally, acquiring the vehicle identification information of the known vehicle corresponding to the second vehicle image with the maximum similarity with the first vehicle image as the vehicle identification information of the vehicle to be detected.
In this embodiment, the similarity between the vehicle to be detected and the vehicle to be matched is directly determined according to the first vehicle license image and the second vehicle license image, so that the problem of inaccurate vehicle identification caused by identification errors in the process of character identification of the license plate can be avoided.
Optionally, in this embodiment, the step of obtaining a target image and performing license plate recognition on the target image to obtain a license plate recognition result includes: and obtaining a target image, and carrying out license plate recognition on the target image by adopting a deep neural network to obtain a license plate recognition result.
In this embodiment, the license plate detection is performed on the target image by using the deep neural network, and the deep neural network can extract the deeper features in the target image, so that the detection result is more accurate.
Optionally, in this embodiment, when the similarity between any two vehicles is detected according to the appearance features of the vehicles, the appearance features of the vehicles may also be extracted by using the deep neural network, and since the deep neural network may extract deeper features, the detection result may be more accurate.
Based on the same inventive concept, the embodiment of the application further provides a vehicle recognition device corresponding to the vehicle recognition method, and since the principle of solving the problem by the device in the embodiment of the application is similar to that of the vehicle recognition method in the embodiment of the application, the implementation of the device can refer to the implementation of the method, and the repetition is omitted.
Referring to fig. 7, a schematic block diagram of a vehicle identification device according to an embodiment of the present application is shown, where the device includes: license plate recognition module 131, appearance recognition module 132, range determination module 133; the vehicle identification means comprises a software functional module which may be stored in the memory in the form of software or firmware or which is solidified in an Operating System (OS) of the electronic device. The license plate recognition module 131 is configured to obtain a target image and perform license plate recognition on the target image to obtain a license plate recognition result, where the target image is an image of a vehicle to be detected.
The license plate recognition module 131 in the present embodiment is configured to perform steps S110 to S120, and for specific description of the license plate recognition module 131, reference may be made to the description of the steps S110 to S120.
And the appearance recognition module 132 is used for carrying out appearance feature recognition on the vehicle to be detected to obtain the vehicle recognition information of the vehicle to be detected when the license plate is not recognized in the license plate recognition result.
The shape recognition module 132 in the present embodiment is used to perform step S130, and a specific description of the shape recognition module 132 may refer to a description of the step S130.
And the range determining module 133 is configured to, when the license plate recognition result includes at least part of characters of a license plate, perform vehicle recognition according to the at least part of characters of the license plate and pre-stored vehicle information of a known vehicle, and obtain vehicle recognition information of the vehicle to be detected, where the vehicle information includes vehicle recognition information of the known vehicle, the known license plate, and an appearance feature.
The range determining module 133 in the present embodiment is configured to perform step S140, and a description of step S140 may be referred to for a specific description of the range determining module 133.
The embodiment of the application firstly detects the license plate of the vehicle to be detected, and then obtains the vehicle identification information of the vehicle to be detected according to the license plate identification result; specifically, in the vehicle identification result, when no license plate exists, the appearance features of the vehicle to be detected are identified. In the vehicle identification result, when a part of license plates exist, a pre-matched vehicle is determined from the known vehicles, and then the vehicle to be detected is identified in the pre-matched vehicle according to the appearance characteristics of the vehicle to be detected. Thereby enabling to improve the accuracy of vehicle identification.
In a possible implementation manner, the range determining module 133 is specifically configured to take at least part of characters of the license plate as first license plate characters; judging whether the first license plate character contains a complete character of a license plate or not; if the first license plate character contains part of characters of the license plate, taking the known vehicle containing the first license plate character as a matched vehicle; for each pre-matched vehicle, calculating a first matching score of the pre-matched vehicle and the vehicle to be detected according to the appearance characteristics of the pre-matched vehicle; acquiring vehicle identification information of the vehicle to be detected according to the first matching score of each pre-matched vehicle; and if the first license plate character contains the complete character of the license plate, acquiring the vehicle identification information of the vehicle to be detected according to a matching result of whether the first license plate character is consistent with the known license plate.
The process flow of each module in the apparatus and the interaction flow between the modules may be described with reference to the related descriptions in the above method embodiments, which are not described in detail herein.
Embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program which, when executed by the processor 110, performs the steps of any of the methods of the embodiments described above.
Specifically, the storage medium can be a general-purpose storage medium, such as a mobile magnetic disk, a hard disk, and the like, and when the computer program on the storage medium is executed, the method of any one of the above embodiments can be executed, so that the problem of low vehicle recognition accuracy with similar appearance features is solved, and the effect of improving the vehicle recognition accuracy is achieved.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described system and apparatus may refer to corresponding procedures in the method embodiments, and are not repeated in the present disclosure. In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, and the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, and for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, indirect coupling or communication connection of devices or modules, electrical, mechanical, or other form.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored on a non-volatile computer readable storage medium executable by the processor 110. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily appreciate variations or alternatives within the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Claims (12)
1. A method of vehicle identification, the method comprising:
acquiring a target image and carrying out license plate recognition on the target image to acquire a license plate recognition result, wherein the target image is an image of a vehicle to be detected;
if no license plate is identified in the license plate identification result, carrying out appearance feature identification on the vehicle to be detected to obtain vehicle identification information of the vehicle to be detected;
and if the license plate recognition result comprises at least part of characters of a license plate, carrying out vehicle recognition according to the at least part of characters of the license plate and pre-stored vehicle information of the known vehicle to obtain the vehicle recognition information of the vehicle to be detected, wherein the vehicle information comprises the vehicle recognition information of the known vehicle, the known license plate and the appearance characteristics.
2. The method according to claim 1, wherein the step of performing vehicle identification based on at least part of characters of the license plate and pre-stored vehicle information of a known vehicle, and obtaining the vehicle identification information of the vehicle to be detected includes:
taking at least part of characters of the license plate as first license plate characters;
judging whether the first license plate character contains a complete character of a license plate or not;
if the first license plate character contains part of characters of the license plate, taking a known vehicle containing the first license plate character as a pre-matched vehicle;
for each pre-matched vehicle, calculating a first matching score of the pre-matched vehicle and the vehicle to be detected according to the appearance characteristics of the pre-matched vehicle;
acquiring vehicle identification information of the vehicle to be detected according to the first matching score of each pre-matched vehicle;
and if the first license plate character contains the complete character of the license plate, acquiring the vehicle identification information of the vehicle to be detected according to a matching result of whether the first license plate character is consistent with the known license plate.
3. The method of claim 2, wherein the step of calculating a first match score for each of the pre-matched vehicles from the profile of the pre-matched vehicle with the vehicle to be detected comprises:
For each pre-matched vehicle, acquiring a first weight corresponding to the pre-matched vehicle according to the first license plate character;
calculating a first appearance score of the pre-matched vehicle according to the appearance characteristics of the pre-matched vehicle;
and calculating the product of the first weight and the first appearance score to obtain a first matching score of the pre-matching vehicle and the vehicle to be detected.
4. The method of claim 3, wherein the step of, for each of the pre-matched vehicles, obtaining a first weight corresponding to the pre-matched vehicle from the first vehicle character comprises:
judging whether a first character which is the same as characters in a preset character set exists in the first car character, wherein the preset character set comprises characters with various similar shapes;
taking other characters except the first character in the first vehicle character as second characters;
and calculating the sum of the character sub-weights of each character in the first car character to obtain the first weight, wherein the character sub-weight of the second character is larger than the character sub-weight of the first character.
5. The method according to claim 2, wherein the step of acquiring the vehicle identification information of the vehicle to be detected from the first matching score of each of the pre-matched vehicles includes:
Screening the first match scores of each pre-matched vehicle for a first match score greater than a confidence threshold;
and acquiring vehicle identification information of the vehicle to be detected according to the first matching score larger than the confidence coefficient threshold value.
6. The method according to claim 2, wherein the step of acquiring the vehicle identification information of the vehicle to be detected based on a result of matching the first license plate character with a known license plate includes:
comparing the first license plate character with each known license plate, and judging whether the known license plate same as the first license plate character exists or not;
if the known license plate which is the same as the first license plate character exists, the known license plate which is the same as the first license plate character is used as a matched license plate;
acquiring vehicle identification information corresponding to the matched license plate as vehicle identification information of a vehicle to be detected;
and if the known license plate which is the same as the first license plate character does not exist, carrying out appearance feature recognition on the vehicle to be detected, and acquiring vehicle recognition information of the vehicle to be detected.
7. The method according to claim 2, wherein the step of acquiring the vehicle identification information of the vehicle to be detected based on a result of matching the first license plate character with a known license plate includes:
Matching the known license plates of the known vehicles according to the first license plate characters;
judging whether a known license plate matched with the first license plate character exists or not;
if a known license plate matched with the first license plate character exists, acquiring vehicle identification information of the vehicle to be detected from vehicle identification information corresponding to the known license plate;
and if the known license plate matched with the first license plate character does not exist, identifying the vehicle to be detected according to the appearance characteristics, and obtaining the vehicle identification information of the vehicle to be detected.
8. The method of claim 7, wherein the step of matching known license plates of respective known vehicles based on the first license plate character comprises:
judging whether the first character which is the same as the character in the preset character set exists in the first car character or not, wherein other characters except the first character in the first car character are second characters, and the preset character set comprises characters with various similar shapes;
if the first character exists in the first vehicle-mounted characters, acquiring a third character similar to the first character from the preset character set according to a preset corresponding relation between similar characters for each first character;
Respectively replacing the first character with a third character similar to the first character aiming at each first character to obtain second car board characters with different combinations;
and matching the first license plate character and the second license plate character with the known license plate respectively.
9. The method according to any one of claims 1 to 8, wherein the steps of acquiring a target image and performing license plate recognition on the target image to obtain a license plate recognition result include:
and obtaining a target image, and carrying out license plate recognition on the target image by adopting a deep neural network to obtain a license plate recognition result.
10. A vehicle identification device, characterized in that the device comprises:
the license plate recognition module is used for acquiring a target image and carrying out license plate recognition on the target image to obtain a license plate recognition result, wherein the target image is an image of a vehicle to be detected;
the appearance recognition module is used for carrying out appearance feature recognition on the vehicle to be detected to obtain vehicle recognition information of the vehicle to be detected when the license plate is not recognized in the license plate recognition result;
and the range determining module is used for carrying out vehicle recognition according to at least part of characters of the license plate and pre-stored vehicle information of the known vehicle when the license plate recognition result comprises at least part of characters of the license plate, and obtaining the vehicle recognition information of the vehicle to be detected, wherein the vehicle information comprises the vehicle recognition information of the known vehicle, the known license plate and the appearance characteristics.
11. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating over the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the method of any one of claims 1 to 9.
12. A computer-readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, performs the steps of the method according to any of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010115418.2A CN111860512B (en) | 2020-02-25 | 2020-02-25 | Vehicle identification method, device, electronic equipment and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010115418.2A CN111860512B (en) | 2020-02-25 | 2020-02-25 | Vehicle identification method, device, electronic equipment and computer readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111860512A CN111860512A (en) | 2020-10-30 |
CN111860512B true CN111860512B (en) | 2023-12-05 |
Family
ID=72985065
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010115418.2A Active CN111860512B (en) | 2020-02-25 | 2020-02-25 | Vehicle identification method, device, electronic equipment and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111860512B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113920596A (en) * | 2021-10-17 | 2022-01-11 | 绿城科技产业服务集团有限公司 | License plate recognition data re-matching method and system for unattended parking lot |
CN118298441A (en) * | 2023-12-26 | 2024-07-05 | 苏州镁伽科技有限公司 | Image matching method, device, electronic equipment and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108875746A (en) * | 2018-05-17 | 2018-11-23 | 北京旷视科技有限公司 | A kind of licence plate recognition method, device, system and storage medium |
WO2019169532A1 (en) * | 2018-03-05 | 2019-09-12 | 深圳前海达闼云端智能科技有限公司 | License plate recognition method and cloud system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106469299B (en) * | 2016-08-31 | 2019-07-19 | 北京邮电大学 | A kind of vehicle search method and device |
-
2020
- 2020-02-25 CN CN202010115418.2A patent/CN111860512B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019169532A1 (en) * | 2018-03-05 | 2019-09-12 | 深圳前海达闼云端智能科技有限公司 | License plate recognition method and cloud system |
CN108875746A (en) * | 2018-05-17 | 2018-11-23 | 北京旷视科技有限公司 | A kind of licence plate recognition method, device, system and storage medium |
Non-Patent Citations (1)
Title |
---|
车辆牌照识别算法的研究与实现;牛进才;余建立;李孟君;;电子测量技术(第06期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111860512A (en) | 2020-10-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108352072B (en) | Object tracking method, object tracking device, and recording medium | |
CN110097068B (en) | Similar vehicle identification method and device | |
CN111145214A (en) | Target tracking method, device, terminal equipment and medium | |
CN111104867A (en) | Recognition model training and vehicle heavy recognition method and device based on component segmentation | |
CN111860512B (en) | Vehicle identification method, device, electronic equipment and computer readable storage medium | |
CN110502982A (en) | The method, apparatus and computer equipment of barrier in a kind of detection highway | |
CN110738219A (en) | Method and device for extracting lines in image, storage medium and electronic device | |
WO2020010715A1 (en) | Electronic device, driving behavior scoring method, and storage medium | |
CN111753642B (en) | Method and device for determining key frame | |
CN111680546A (en) | Attention detection method, attention detection device, electronic equipment and storage medium | |
CN112001269A (en) | Vehicle identification method and device, computer equipment and storage medium | |
CN113408364B (en) | Temporary license plate recognition method, system, device and storage medium | |
CN111126257A (en) | Behavior detection method and device | |
CN116958606B (en) | Image matching method and related device | |
CN114332809A (en) | Image identification method and device, electronic equipment and storage medium | |
CN113011517A (en) | Positioning result detection method and device, electronic equipment and storage medium | |
CN112329886A (en) | Double-license plate recognition method, model training method, device, equipment and storage medium | |
CN111611906A (en) | Obstacle detection method, system and medium for automatic parking | |
CN116385007A (en) | Data detection method, device, computer equipment and computer readable storage medium | |
CN111866589A (en) | Video data verification method and device, electronic equipment and storage medium | |
CN113361413B (en) | Mileage display area detection method, device, equipment and storage medium | |
JP6665522B2 (en) | Information processing apparatus, information processing system, and information processing program | |
CN113658426A (en) | Vehicle accident identification method and device | |
CN113743194A (en) | Face silence living body detection method and device, electronic equipment and storage medium | |
CN113642521A (en) | Traffic light identification quality evaluation method and device and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |