CN111191481B - Vehicle identification method and system - Google Patents
Vehicle identification method and system Download PDFInfo
- Publication number
- CN111191481B CN111191481B CN201811350320.4A CN201811350320A CN111191481B CN 111191481 B CN111191481 B CN 111191481B CN 201811350320 A CN201811350320 A CN 201811350320A CN 111191481 B CN111191481 B CN 111191481B
- Authority
- CN
- China
- Prior art keywords
- license plate
- vehicle
- target picture
- position information
- region
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- 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
- G06V20/63—Scene text, e.g. street names
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The application provides a vehicle identification method and a system, wherein the method comprises the following steps: when the camera detects a vehicle, capturing a target picture from the video stream, identifying the target picture to obtain a license plate identification result, and sending the target picture and the license plate identification result to a server; the server locates the vehicle region in the target picture according to license plate position information in the license plate recognition result; and digging out the target block of the vehicle area from the target picture, and identifying to obtain the information related to the vehicle. Because the camera and the server transmit the target picture and the license plate recognition result, the real-time transmission is good and the reliability is high; the quality loss exists in the encoding process before the video stream is transmitted, and the camera carries out license plate recognition aiming at the data before encoding, so that the accuracy of license plate recognition results is high, the server carries out secondary recognition based on the license plate recognition results, the recognition accuracy can be improved, and due to the high processing capacity of the server chip, all information of the vehicle can be recognized, and the problem that the processing capacity of the camera chip is limited is solved.
Description
Technical Field
The application relates to the technical field of security monitoring, in particular to a vehicle identification method and system.
Background
In the current security monitoring industry, there is an increasing demand for identifying vehicles, and it is common to identify detailed attributes of vehicles, such as the type, color, annual fee, brand, belting, annual inspection mark, and the like of vehicles.
At present, since the chip processing capability integrated inside the front-end camera is limited, all the attributes of the vehicle cannot be identified, the rear-end server analyzes all the attributes of the vehicle in the video stream by sending the video stream collected by the front-end camera to the rear-end server in real time.
However, on the one hand, the video stream transmitted between the front-end camera and the back-end server occupies relatively large network resources, the transmission process is time-consuming and has high delay, and the problems of network congestion and low reliability can occur; the video stream transmitted to the back-end server by the front-end camera is encoded data, and the data has quality loss in the encoding process, so that the accuracy of the analysis result of the back-end server on the video stream is low.
Disclosure of Invention
In view of the above, the present application provides a vehicle recognition method and system to solve the problems that the above recognition method can cause large network resource occupation and low accuracy of analysis results.
According to a first aspect of an embodiment of the present application, there is provided a vehicle identification method, the method including:
when a front-end camera detects a vehicle from a video stream acquired in real time, capturing a target picture from the video stream and identifying the target picture to obtain a license plate identification result, and sending the target picture and the license plate identification result to an analysis server;
the analysis server locates a vehicle region in the target picture according to license plate region position information contained in the license plate recognition result; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
According to a second aspect of an embodiment of the present application, there is provided a vehicle identification system, the system comprising:
the front-end camera is used for capturing a target picture from a video stream acquired in real time and identifying the target picture when the vehicle is detected from the video stream, obtaining a license plate identification result and sending the target picture and the license plate identification result to the analysis server;
the analysis server is used for positioning a vehicle region in the target picture according to license plate region position information contained in the license plate recognition result; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
When the front-end camera detects a vehicle from a video stream acquired in real time, the front-end camera grabs a target picture from the video stream and recognizes the target picture to obtain a license plate recognition result, the target picture and the license plate recognition result are sent to the analysis server, the analysis server positions a vehicle region in the target picture according to license plate region position information contained in the license plate recognition result, and when the vehicle is positioned, a target block corresponding to the vehicle region is scratched from the target picture and recognized to obtain detailed information related to the vehicle.
Based on the description, the video stream occupying resources is not compared because the target picture and the license plate recognition result are transmitted between the front-end camera and the analysis server at the rear end, so that the real-time transmission performance is good and the reliability is high; the front-end camera is used for recognizing the license plate according to the data before encoding, so that the accuracy of the license plate recognition result is high, and the analysis server at the rear end is used for performing secondary recognition based on the license plate recognition result of the front-end camera, so that the recognition accuracy can be improved, meanwhile, due to the fact that the chip processing capacity of the analysis server at the rear end is high, all information related to a vehicle can be recognized, the problem that the chip processing capacity of the front-end camera is limited can be solved, and the user requirements are met.
Drawings
FIG. 1A is a flow chart illustrating an embodiment of a method of vehicle identification according to an exemplary embodiment of the present application;
FIG. 1B is a target picture according to the embodiment of FIG. 1A;
fig. 2 is a block diagram of a vehicle identification system according to an exemplary embodiment of the present application.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the application. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
There are two problems with the implementation in the related art: on the one hand, the video stream transmitted between the front-end camera and the back-end server occupies relatively large network resources, the transmission process is time-consuming and high in delay, and the problems of network congestion and low reliability can occur; the video stream transmitted to the back-end server by the front-end camera is encoded data, and the data has quality loss in the encoding process, so that the accuracy of the analysis result of the back-end server on the video stream is low.
In order to solve the problems, the application provides a vehicle identification method, when a front-end camera detects a vehicle from a video stream acquired in real time, the front-end camera grabs a target picture from the video stream and identifies the target picture to obtain a license plate identification result, the target picture and the license plate identification result are sent to an analysis server, the analysis server positions a vehicle region in the target picture according to license plate region position information contained in the license plate identification result, and when the vehicle is positioned, a target block corresponding to the vehicle region is scratched from the target picture and identified to obtain detailed information related to the vehicle.
Based on the description, the video stream occupying resources is not compared because the target picture and the license plate recognition result are transmitted between the front-end camera and the analysis server at the rear end, so that the real-time transmission performance is good and the reliability is high; the front-end camera is used for recognizing the license plate according to the data before encoding, so that the accuracy of the license plate recognition result is high, and the analysis server at the rear end is used for performing secondary recognition based on the license plate recognition result of the front-end camera, so that the recognition accuracy can be improved, meanwhile, due to the fact that the chip processing capacity of the analysis server at the rear end is high, all information related to a vehicle can be recognized, the problem that the chip processing capacity of the front-end camera is limited can be solved, and the user requirements are met.
The following describes in detail a technical solution for vehicle identification with specific embodiments.
Fig. 1A is a flowchart of an embodiment of a vehicle identification method according to an exemplary embodiment of the present application, and as shown in fig. 1A, the vehicle identification method includes the steps of:
step 101: when a vehicle is detected from a video stream acquired in real time, the front-end camera captures a target picture from the video stream and recognizes the target picture to obtain a license plate recognition result, and the target picture and the license plate recognition result are sent to an analysis server.
In an embodiment, for the process of sending the target picture and the license plate recognition result to the analysis server, the front-end camera may send the target picture and the license plate recognition result to the access server, where the access server sends the target picture to the picture server for storage, and when receiving the storage address of the target picture returned by the picture server, sends the storage address and the license plate recognition result to the database server, where the database server stores the storage address and the license plate recognition result as a vehicle record, and adds the storage address and the license plate recognition result to the vehicle recognition task to the analysis server, where the analysis server may extract the storage address included in the vehicle recognition task when receiving the vehicle recognition task submitted by the database server, and obtain the target picture in the storage space corresponding to the storage address from the picture server.
The license plate recognition result comprises license plate region position information and/or license plate numbers, and the performance loss of the analysis server can be reduced through the access server, the picture server and the database server, so that the analysis server is only used for carrying out recognition processing. The storage address provided by the picture server may be a URL (Uniform Resource Locator ) address. The access server can pull the target picture and the license plate recognition result to the front-end camera in a pulling mode, and the front-end camera can push the target picture and the license plate recognition result to the access server in a pushing mode.
It should be noted that, the front-end camera may be a monitoring camera disposed at a different location, and the access server may obtain monitoring data of each monitoring camera. The chip processing capability in the front-end camera is limited, so that the complexity, accuracy and stability of the license plate recognition algorithm are low, and further, in the license plate recognition result obtained by the front-end camera through the license plate recognition algorithm, the front-end camera may have license plate region position information, license plate number, license plate region position information, license plate number. The front-end camera can also obtain the confidence coefficient of the license plate recognition result (comprising the confidence coefficient corresponding to the license plate region position information and/or the confidence coefficient corresponding to the license plate number) through a license plate recognition algorithm. In addition, when the front-end camera captures the target picture from the video stream, the front-end camera can also obtain the time-space information (capture time and capture place) of the target picture, so that the vehicle record of the database server can also contain the time-space information of the target picture.
It should be further noted that, since there may be multiple vehicles in the target picture, the front-end camera may obtain multiple license plate recognition results with the license plate recognition algorithm, so when the database server receives the storage address of the target picture and the multiple license plate recognition results, each license plate recognition result may be stored as a vehicle record, the same storage address is recorded in each vehicle record, and the database server may obtain a vehicle recognition task for each vehicle record.
Step 102: and the analysis server locates the vehicle region in the target picture according to the license plate region position information contained in the license plate recognition result.
In an embodiment, as can be seen from the description of step 101, the vehicle recognition result further includes a confidence coefficient corresponding to the license plate region position information, so that the analysis server may obtain the reference region position information related to the license plate region position information from the target picture, if the reference region position information is obtained, calculate the confidence coefficient corresponding to the reference region position information, select the maximum confidence coefficient from the confidence coefficient corresponding to the license plate region position information and the confidence coefficient corresponding to the reference region position information, and locate the vehicle region in the target picture by using the license plate region position information or the reference region position information corresponding to the maximum confidence coefficient.
The complexity, accuracy and stability of the license plate recognition algorithm set by the front-end camera are limited, so that the license plate region position information obtained by recognition may not include the whole license plate or may include other regions except the license plate, the chip processing capability of the analysis server is high, the complexity, accuracy and stability of the license plate recognition algorithm (such as a neural network) can be set, the license plate recognition algorithm can acquire the reference region position information related to the license plate region position information from the target picture, the size of the region corresponding to the reference region position information meets the specified size and the region corresponding to the license plate region position information is partially overlapped or completely overlapped, the specified size refers to the maximum size of the license plate in the image, the specified size is set according to practical experience, and the reference region position information is obtained on the basis of the license plate region position information, can include the whole license plate, and the confidence is relatively high compared with the confidence of the license plate region position information. In addition, if the reference area position information related to the license plate area position information cannot be obtained from the target picture through the license plate recognition algorithm, the corresponding area of the license plate area position information in the target picture is not the license plate area, and the fact that the vehicle area cannot be positioned according to the license plate area position information is determined.
It will be appreciated by those skilled in the art that the license plate region location information related to the present embodiment may include the pixel coordinates of the license plate in the image, the pixel width and the pixel height of the license plate.
In an exemplary scenario, as shown in fig. 1B, the area corresponding to the license plate area location information identified by the front-end camera does not include the entire license plate, the confidence coefficient thereof is 900, the area corresponding to the reference area location information obtained by the analysis server according to the license plate area location information identification includes the entire license plate, the confidence coefficient thereof is 980, and since the confidence coefficient 980 of the reference area location information is greater than the confidence coefficient 900 of the license plate area location information, the analysis server subsequently uses the reference area location information to locate the vehicle in the target image.
Step 103: when the analysis server locates the vehicle region, the target block corresponding to the vehicle region is scratched out from the target picture and identified, and detailed information related to the vehicle is obtained.
In one embodiment, the analysis server may identify the detailed information associated with the vehicle via a highly complex, stable identification algorithm.
The detailed information related to the vehicle can comprise characteristic information of the vehicle and all attribute information of the vehicle, and the analysis server can add the detailed information related to the vehicle to a vehicle record stored in the database server so as to be inquired by a client and realize a service function of searching a graph.
In an embodiment, if the license plate recognition result does not include license plate region position information or is not positioned to the vehicle region according to the license plate region position information, the analysis server may position the vehicle region in the target picture according to the license plate number when the license plate recognition result includes the license plate number, and scratch out a target block corresponding to the vehicle region from the target picture and identify the target block when the target block is positioned to the vehicle region, so as to obtain detailed information related to the vehicle.
In an embodiment, as known from the description in step 101, the license plate recognition result further includes a confidence coefficient corresponding to a license plate number, and for the process of positioning the vehicle in the target image according to the license plate number, the reference license plate number in the target image can be first re-recognized, the confidence coefficient corresponding to the reference license plate number is calculated, then each reference license plate number is character-matched with the license plate number to obtain a matched character number of each reference license plate number, a matched character number greater than a preset threshold value is obtained from the matched character numbers, the maximum confidence coefficient is selected from the confidence coefficient corresponding to the license plate number and the confidence coefficient of the reference license plate number corresponding to the maximum matched character number, and finally the vehicle area is positioned in the target image by using the license plate number or the reference license plate number corresponding to the maximum confidence coefficient.
The preset threshold value can be set according to actual requirements, if the number of matching characters larger than the preset threshold value is not found in the final matching result, the matching is failed, and the current flow is ended.
The resolution of the vehicle in the lower position in the target picture is higher as the vehicle gets closer to the camera, so that the resolution of the vehicle in the lower position in the target picture is highest, the maximum confidence process is selected according to the confidence corresponding to the license plate number and the confidence of the reference license plate number corresponding to the maximum matching character number, the analysis server can firstly establish a pixel coordinate system by taking the left lower corner vertex of the target picture as the origin, the horizontal direction as the horizontal axis and the vertical direction as the vertical axis, acquire the ordinate of the reference license plate number in the target picture when the reference license plate number is identified, and select the reference license plate number with the minimum ordinate from the reference license plate numbers corresponding to the maximum matching character numbers if the maximum matching character number is selected, and then select the maximum confidence from the confidence of the reference license plate number with the minimum ordinate and the confidence of the license plate number; if the number of the selected maximum matching characters is 1, the maximum confidence is directly selected from the confidence corresponding to the license plate number and the confidence of the reference license plate number corresponding to the maximum matching character number.
It should be noted that, for the situation that the front-end camera cannot recognize the license plate recognition result, or the license plate recognition result does not include the license plate number according to the license plate region position information, the analysis server may set up a pixel coordinate system with the left lower corner vertex of the target picture as the origin, the horizontal direction as the horizontal axis, and the vertical direction as the vertical axis, then recognize the no-license vehicle in the target picture, and acquire the ordinate of the no-license vehicle in the target picture, then select the no-license vehicle with the minimum ordinate from the recognized no-license vehicles, and recognize the detailed information related to the no-license vehicle.
For the above-mentioned steps 101 to 103, it should be further explained that, because the complexity, accuracy and stability of the license plate recognition algorithm set by the front-end camera are also relatively low, the accuracy and reliability of the recognized license plate region position information is generally higher than those of the recognized license plate number, so that the vehicle region is preferentially positioned according to the license plate region position information, and if the vehicle region is not positioned in the target picture according to the license plate region position information, the vehicle region is further positioned in the target picture according to the license plate number.
In the embodiment of the application, when a vehicle is detected from a video stream acquired in real time, a front-end camera captures a target picture from the video stream and recognizes the target picture to obtain a license plate recognition result, the target picture and the license plate recognition result are sent to an analysis server, the analysis server positions a vehicle region in the target picture according to license plate region position information contained in the license plate recognition result, and when the vehicle is positioned, a target block corresponding to the vehicle region is scratched from the target picture and recognized to obtain detailed information related to the vehicle.
Based on the description, the video stream occupying resources is not compared because the target picture and the license plate recognition result are transmitted between the front-end camera and the analysis server at the rear end, so that the real-time transmission performance is good and the reliability is high; the front-end camera is used for recognizing the license plate according to the data before encoding, so that the accuracy of the license plate recognition result is high, and the analysis server at the rear end is used for performing secondary recognition based on the license plate recognition result of the front-end camera, so that the recognition accuracy can be improved, meanwhile, due to the fact that the chip processing capacity of the analysis server at the rear end is high, all information related to a vehicle can be recognized, the problem that the chip processing capacity of the front-end camera is limited can be solved, and the user requirements are met.
Fig. 2 is a block diagram of a vehicle recognition system according to an exemplary embodiment of the present application, and as shown in fig. 2, the vehicle recognition system includes a front-end camera and an analysis server:
the front-end camera 210 is configured to capture a target picture from a video stream acquired in real time and identify the target picture when a vehicle is detected from the video stream, obtain a license plate identification result, and send the target picture and the license plate identification result to an analysis server;
the analysis server 220 is configured to locate a vehicle region in the target picture according to license plate region location information included in the license plate recognition result; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
In an alternative implementation manner, the license plate recognition result further comprises a confidence corresponding to the license plate region position information,
the analysis server 220 is specifically configured to obtain, from the target picture, reference area location information related to the license plate area location information in the process of locating the vehicle area in the target picture according to license plate area location information included in the license plate recognition result, where an area corresponding to the reference area location information meets a specified size and is partially overlapped or completely overlapped with an area corresponding to the license plate area location information; if the reference area position information is acquired, calculating the confidence coefficient corresponding to the reference area position information, and selecting the maximum confidence coefficient from the confidence coefficient corresponding to the license plate area position information and the confidence coefficient corresponding to the reference area position information; and positioning a vehicle region in the target picture by utilizing license plate region position information or reference region position information corresponding to the maximum confidence coefficient.
In an optional implementation manner, if the license plate recognition result does not include license plate region position information or is not located in the vehicle region according to the license plate region position information, the analysis server 220 is further configured to locate the vehicle region in the target picture according to the license plate number when the license plate recognition result includes the license plate number; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
In an optional implementation manner, the license plate recognition result further includes a confidence coefficient corresponding to a license plate number, and the analysis server 220 is specifically configured to re-recognize a reference license plate number in the target picture and calculate the confidence coefficient corresponding to the reference license plate number in a process of positioning a vehicle region in the target picture according to the license plate number; performing character matching on each reference license plate number and the license plate number to obtain the number of matched characters of each reference license plate number, and obtaining the number of matched characters larger than a preset threshold value from the number of matched characters; selecting the maximum number of matched characters from the acquired number of matched characters; selecting the maximum confidence from the confidence corresponding to the reference license plate number corresponding to the maximum matching character number and the confidence corresponding to the license plate number; and positioning a vehicle area in the target picture by using the license plate number or the reference license plate number corresponding to the maximum confidence.
In an optional implementation manner, if the front-end camera cannot recognize the license plate recognition result, or the license plate recognition result does not include a license plate number according to the license plate region position information, the analysis server 220 is further configured to establish a pixel coordinate system with a lower left corner vertex of the target picture as an origin, a horizontal direction as a horizontal axis, and a vertical direction as a vertical axis; identifying a card-free vehicle in the target picture, and acquiring the ordinate of the card-free vehicle in the target picture; and selecting the card-free vehicle with the smallest ordinate from the identified card-free vehicles, and identifying detailed information related to the card-free vehicle.
The implementation process of the functions and roles of each unit in the above device is specifically shown in the implementation process of the corresponding steps in the above method, and will not be described herein again.
For the device embodiments, reference is made to the description of the method embodiments for the relevant points, since they essentially correspond to the method embodiments. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purposes of the present application. Those of ordinary skill in the art will understand and implement the present application without undue burden.
Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather to enable any modification, equivalent replacement, improvement or the like to be made within the spirit and principles of the application.
Claims (8)
1. A method of vehicle identification, the method comprising:
when a front-end camera detects a vehicle from a video stream acquired in real time, capturing a target picture from the video stream and identifying the target picture to obtain a license plate identification result, and sending the target picture and the license plate identification result to an analysis server;
if the license plate recognition result comprises license plate region position information and confidence corresponding to the license plate region position information; the analysis server acquires reference area position information related to the license plate area position information from the target picture, wherein the area size corresponding to the reference area position information meets the specified size and the area corresponding to the license plate area position information is partially overlapped or completely overlapped;
calculating the confidence coefficient corresponding to the position information of the reference area, and selecting the maximum confidence coefficient from the confidence coefficient corresponding to the position information of the license plate area and the confidence coefficient corresponding to the position information of the reference area; positioning a vehicle region in the target picture by utilizing license plate region position information or reference region position information corresponding to the maximum confidence coefficient;
and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
2. The method of claim 1, wherein if the license plate recognition result does not include license plate region location information or is not located in a vehicle region, the method further comprises:
when the license plate recognition result comprises a license plate number, positioning a vehicle region in the target picture according to the license plate number; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
3. The method of claim 2, wherein the license plate recognition result further includes a confidence level corresponding to a license plate number, and locating a vehicle region in the target picture according to the license plate number includes:
re-identifying the reference license plate number in the target picture, and calculating the confidence coefficient corresponding to the reference license plate number;
performing character matching on each reference license plate number and the license plate number to obtain the number of matched characters of each reference license plate number, and obtaining the number of matched characters larger than a preset threshold value from the number of matched characters;
selecting the maximum number of matched characters from the acquired number of matched characters;
selecting the maximum confidence from the confidence corresponding to the license plate number and the confidence corresponding to the reference license plate number of the maximum matching character number;
and positioning the vehicle area in the target picture by using the license plate number or the reference license plate number corresponding to the maximum confidence.
4. The method of claim 1, wherein if the front-end camera fails to recognize a license plate recognition result or is not located in a vehicle area and the license plate recognition result does not include a license plate number, the method further comprises:
establishing a pixel coordinate system by taking the left lower corner vertex of the target picture as an origin, the horizontal direction as a horizontal axis and the vertical direction as a vertical axis;
identifying a card-free vehicle in the target picture, and acquiring the ordinate of the card-free vehicle in the target picture;
and selecting the card-free vehicle with the smallest ordinate from the identified card-free vehicles, and identifying detailed information related to the card-free vehicle.
5. A vehicle identification system, the system comprising:
the front-end camera is used for capturing a target picture from a video stream acquired in real time and identifying the target picture when the vehicle is detected from the video stream, obtaining a license plate identification result and sending the target picture and the license plate identification result to the analysis server;
the analysis server is used for acquiring reference area position information related to the license plate area position information from the target picture if the license plate recognition result comprises the license plate area position information and the confidence corresponding to the license plate area position information, wherein the area corresponding to the reference area position information meets the specified size and is partially overlapped or completely overlapped with the area corresponding to the license plate area position information; calculating the confidence coefficient corresponding to the position information of the reference area, and selecting the maximum confidence coefficient from the confidence coefficient corresponding to the position information of the license plate area and the confidence coefficient corresponding to the position information of the reference area; positioning a vehicle region in the target picture by utilizing license plate region position information or reference region position information corresponding to the maximum confidence coefficient; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
6. The system of claim 5, wherein if the license plate recognition result does not include license plate region location information or is not located in a vehicle region, the analysis server is further configured to locate a vehicle region in the target picture according to the license plate number when the license plate recognition result includes the license plate number; and when the vehicle region is positioned, a target block corresponding to the vehicle region is scratched out from the target picture and is identified, so that detailed information related to the vehicle is obtained.
7. The system according to claim 6, wherein the license plate recognition result further includes a confidence level corresponding to a license plate number, and the analysis server is specifically configured to re-recognize a reference license plate number in the target picture and calculate the confidence level corresponding to the reference license plate number in positioning a vehicle region in the target picture according to the license plate number; performing character matching on each reference license plate number and the license plate number to obtain the number of matched characters of each reference license plate number, and obtaining the number of matched characters larger than a preset threshold value from the number of matched characters; selecting the maximum number of matched characters from the acquired number of matched characters; selecting the maximum confidence from the confidence corresponding to the reference license plate number corresponding to the maximum matching character number and the confidence corresponding to the license plate number; and positioning a vehicle area in the target picture by using the license plate number or the reference license plate number corresponding to the maximum confidence.
8. The system of claim 5, wherein if the front-end camera cannot recognize the license plate recognition result or is not positioned in a vehicle area and the license plate recognition result does not include a license plate number, the analysis server is further configured to establish a pixel coordinate system with a lower left corner vertex of the target picture as an origin, a horizontal direction as a horizontal axis, and a vertical direction as a vertical axis; identifying a card-free vehicle in the target picture, and acquiring the ordinate of the card-free vehicle in the target picture; and selecting the card-free vehicle with the smallest ordinate from the identified card-free vehicles, and identifying detailed information related to the card-free vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811350320.4A CN111191481B (en) | 2018-11-14 | 2018-11-14 | Vehicle identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811350320.4A CN111191481B (en) | 2018-11-14 | 2018-11-14 | Vehicle identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111191481A CN111191481A (en) | 2020-05-22 |
| CN111191481B true CN111191481B (en) | 2023-08-25 |
Family
ID=70710487
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811350320.4A Active CN111191481B (en) | 2018-11-14 | 2018-11-14 | Vehicle identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111191481B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488108B (en) * | 2020-12-11 | 2024-08-13 | 广州小鹏自动驾驶科技有限公司 | Parking space number identification method and device, electronic equipment and storage medium |
| CN113012439B (en) * | 2021-03-29 | 2022-06-21 | 北京百度网讯科技有限公司 | Vehicle detection method, device, equipment and storage medium |
| CN116206300B (en) * | 2023-05-05 | 2024-02-27 | 浪潮数字粮储科技有限公司 | Multi-manufacturer identification camera integration compatible method, equipment and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IL152637A0 (en) * | 2002-11-04 | 2004-02-19 | Imp Vision Ltd | Automatic, real time and complete identification of vehicles |
| KR20050020350A (en) * | 2003-08-22 | 2005-03-04 | 김영모 | System for detecting and recognizing a car plate area by using stereo vision and method thereof |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
| CN104123852A (en) * | 2014-08-08 | 2014-10-29 | 浪潮电子信息产业股份有限公司 | Vehicle searching method based on cloud computing for parking lot |
| KR101561626B1 (en) * | 2014-08-11 | 2015-10-30 | 주식회사 카눅스 | The Vehicle Black Box Capable of Real-Time Recognizing a License Number Plate for Moving Vehicle |
| CN105678304A (en) * | 2015-12-30 | 2016-06-15 | 浙江宇视科技有限公司 | Vehicle-logo identification method and apparatus |
| CN106297362A (en) * | 2016-08-17 | 2017-01-04 | 重庆元云联科技有限公司 | The method of parking lot reverse car seeking |
| CN106448184A (en) * | 2016-12-15 | 2017-02-22 | 深圳市捷顺科技实业股份有限公司 | Identifying method of Vehicles and exit of vehicles |
| CN106683409A (en) * | 2017-02-08 | 2017-05-17 | 南京杰迈视讯科技有限公司 | Heavy type lorry photographing and identifying management method and the system of the same |
| CN106780886A (en) * | 2016-12-16 | 2017-05-31 | 深圳市捷顺科技实业股份有限公司 | A kind of vehicle identification system and vehicle are marched into the arena, appearance recognition methods |
| AU2017261601A1 (en) * | 2016-06-24 | 2018-01-18 | Accenture Global Solutions Limited | Intelligent automatic license plate recognition for electronic tolling environments |
| CN108520629A (en) * | 2018-04-09 | 2018-09-11 | 天津中兴智联科技有限公司 | A kind of vacation deck identifying system and its judgment method |
| WO2018177071A1 (en) * | 2017-03-31 | 2018-10-04 | 杭州海康威视数字技术股份有限公司 | Method and apparatus for matching registration plate number, and method and apparatus for matching character information |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8218822B2 (en) * | 2007-05-14 | 2012-07-10 | Pips Technology, Inc. | Apparatus and method for recognizing the state of origin of a vehicle license plate |
| US20170140237A1 (en) * | 2015-11-13 | 2017-05-18 | Hunter Engineering Company | Methods For Vehicle Identification And Specification Recall With Localization Optimization For License Plate Recognition |
| KR101873576B1 (en) * | 2016-10-31 | 2018-07-03 | 한국전자통신연구원 | System and method for recognizing information from vehicle license plate |
| US20180268238A1 (en) * | 2017-03-14 | 2018-09-20 | Mohammad Ayub Khan | System and methods for enhancing license plate and vehicle recognition |
-
2018
- 2018-11-14 CN CN201811350320.4A patent/CN111191481B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IL152637A0 (en) * | 2002-11-04 | 2004-02-19 | Imp Vision Ltd | Automatic, real time and complete identification of vehicles |
| KR20050020350A (en) * | 2003-08-22 | 2005-03-04 | 김영모 | System for detecting and recognizing a car plate area by using stereo vision and method thereof |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
| CN104123852A (en) * | 2014-08-08 | 2014-10-29 | 浪潮电子信息产业股份有限公司 | Vehicle searching method based on cloud computing for parking lot |
| KR101561626B1 (en) * | 2014-08-11 | 2015-10-30 | 주식회사 카눅스 | The Vehicle Black Box Capable of Real-Time Recognizing a License Number Plate for Moving Vehicle |
| CN105678304A (en) * | 2015-12-30 | 2016-06-15 | 浙江宇视科技有限公司 | Vehicle-logo identification method and apparatus |
| AU2017261601A1 (en) * | 2016-06-24 | 2018-01-18 | Accenture Global Solutions Limited | Intelligent automatic license plate recognition for electronic tolling environments |
| CN106297362A (en) * | 2016-08-17 | 2017-01-04 | 重庆元云联科技有限公司 | The method of parking lot reverse car seeking |
| CN106448184A (en) * | 2016-12-15 | 2017-02-22 | 深圳市捷顺科技实业股份有限公司 | Identifying method of Vehicles and exit of vehicles |
| CN106780886A (en) * | 2016-12-16 | 2017-05-31 | 深圳市捷顺科技实业股份有限公司 | A kind of vehicle identification system and vehicle are marched into the arena, appearance recognition methods |
| CN106683409A (en) * | 2017-02-08 | 2017-05-17 | 南京杰迈视讯科技有限公司 | Heavy type lorry photographing and identifying management method and the system of the same |
| WO2018177071A1 (en) * | 2017-03-31 | 2018-10-04 | 杭州海康威视数字技术股份有限公司 | Method and apparatus for matching registration plate number, and method and apparatus for matching character information |
| CN108520629A (en) * | 2018-04-09 | 2018-09-11 | 天津中兴智联科技有限公司 | A kind of vacation deck identifying system and its judgment method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111191481A (en) | 2020-05-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110705405B (en) | Target labeling method and device | |
| CN110858394B (en) | Image quality evaluation method and device, electronic equipment and computer readable storage medium | |
| CN111191481B (en) | Vehicle identification method and system | |
| CN110569856B (en) | Sample labeling method and device, and damage category identification method and device | |
| JP6921694B2 (en) | Monitoring system | |
| EP2450667A1 (en) | Vision system and method of analyzing an image | |
| US8948533B2 (en) | Increased quality of image objects based on depth in scene | |
| CN110889314B (en) | Image processing method, device, electronic equipment, server and system | |
| US20170256165A1 (en) | Mobile on-street parking occupancy detection | |
| KR101788225B1 (en) | Method and System for Recognition/Tracking Construction Equipment and Workers Using Construction-Site-Customized Image Processing | |
| KR102297217B1 (en) | Method and apparatus for identifying object and object location equality between images | |
| US10635948B2 (en) | Method for locating one or more candidate digital images being likely candidates for depicting an object | |
| CN111241932A (en) | Automobile exhibition room passenger flow detection and analysis system, method and storage medium | |
| JP2020038632A5 (en) | ||
| CN105760844A (en) | Video stream data processing method, apparatus and system | |
| CN113255651A (en) | Package security check method, device and system, node equipment and storage device | |
| CN117528035B (en) | Object cross-border head tracking method and system based on active notification | |
| US20170122763A1 (en) | Method for ascertaining in a backend, and providing for a vehicle, a data record, describing a landmark, for the vehicle to determine its own position | |
| US20230012137A1 (en) | Pedestrian search method, server, and storage medium | |
| EP3761228A1 (en) | Computer-implemented method | |
| CN113496163B (en) | Obstacle recognition method and device | |
| CN110660000A (en) | Data prediction method, device, equipment and computer readable storage medium | |
| CN107895386A (en) | A kind of multi-platform joint objective autonomous classification method | |
| CN113673449A (en) | Data storage method, device, equipment and storage medium | |
| KR20160116377A (en) | Method and apparatus for identifying place in image contents |
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 |