CN114170810A - Vehicle traveling direction identification method, system and device - Google Patents
Vehicle traveling direction identification method, system and device Download PDFInfo
- Publication number
- CN114170810A CN114170810A CN202111632979.0A CN202111632979A CN114170810A CN 114170810 A CN114170810 A CN 114170810A CN 202111632979 A CN202111632979 A CN 202111632979A CN 114170810 A CN114170810 A CN 114170810A
- Authority
- CN
- China
- Prior art keywords
- license plate
- vehicle
- image data
- target
- classification model
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000013145 classification model Methods 0.000 claims abstract description 67
- 238000012549 training Methods 0.000 claims abstract description 24
- 238000010801 machine learning Methods 0.000 claims abstract description 14
- 238000012544 monitoring process Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 6
- 230000002085 persistent effect Effects 0.000 claims description 3
- 230000001052 transient effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 description 9
- 238000001514 detection method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000000295 complement effect Effects 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
- 238000013459 approach Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/056—Detecting movement of traffic to be counted or controlled with provision for distinguishing direction of travel
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the application discloses a vehicle driving direction identification method, a system and a device thereof, which are used for identifying the driving direction of a vehicle, are applied to a parking lot parking space management system and improve the use experience of a user. The method in the embodiment of the application comprises the following steps: acquiring and storing license plate image data of a vehicle; if the license plate image data meets the use condition of the license plate track algorithm, identifying the driving direction of the vehicle by using the license plate track algorithm; and if the license plate image data does not meet the use conditions of the license plate trajectory algorithm, identifying the vehicle traveling direction of the vehicle by using a target classification model, wherein the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the vehicle traveling direction corresponding to the vehicle head and tail image data, and the target classification model stores the target corresponding relation between the vehicle head and tail image data and the vehicle traveling direction.
Description
Technical Field
The embodiment of the application relates to the field of intelligent video monitoring, in particular to a method, a system and a device for identifying a driving direction.
Background
The intelligent parking lot management system is a general name of modern parking lot vehicle charging and equipment automatic management, and is a high-tech electromechanical integrated product for completely placing a parking lot under the unified management of a computer. According to its design principle, intelligent parking lot management system can divide into three major parts: information acquisition and transmission, information processing and human-computer interface, and information storage and query. According to the design principle and the composition structure, the realizable functions are as follows: temporary vehicle charging, parking lot management, parking space guidance, reverse vehicle searching, special vehicle management, image comparison and the like.
With the maturity, application and popularization of vehicle detection and identification technologies and license plate identification technologies, vehicle detection devices and license plate identification devices have been widely used in places such as parking lot entrances and exits, commercial building entrances and exits, toll station entrances and exits, lane and parking stall monitoring and the like. The vehicle detection device and the license plate recognition device acquire the driving direction information, and judge whether to perform parking space lock locking or unlocking operation or whether to perform timing and charging operation and the like according to the driving direction information. Most of the existing vehicle traveling direction identification methods are obtained by acquiring and analyzing the traveling track of a vehicle.
However, in an actual application scenario, when a vehicle is displaced in a parking space for a short distance or travels at a high speed in a certain displacement, it is difficult for the monitoring device to obtain an image sample that can be used for analyzing the vehicle traveling direction, and it is difficult to analyze the vehicle traveling direction, so that it is difficult for the system to accurately analyze the vehicle traveling situation and the parking space situation in the parking space management process.
Disclosure of Invention
The embodiment of the application provides a vehicle driving direction identification method, a vehicle driving direction identification system and a vehicle driving direction identification device, which are used for identifying the driving direction of a vehicle, are applied to a parking lot parking space management system and improve the use experience of a user.
The method for identifying the vehicle traveling direction comprises the following steps:
acquiring and storing license plate image data of a vehicle;
if the license plate image data meets the use condition of the license plate track algorithm, identifying the driving direction of the vehicle by using the license plate track algorithm;
and if the license plate image data does not meet the use conditions of the license plate trajectory algorithm, identifying the vehicle traveling direction of the vehicle by using a target classification model, wherein the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the vehicle traveling direction corresponding to the vehicle head and tail image data, and the target classification model stores the target corresponding relation between the vehicle head and tail image data and the vehicle traveling direction.
Optionally, the acquiring and storing license plate image data of the vehicle includes:
when the monitoring module monitors a vehicle, the camera module acquires license plate image data of the vehicle and stores the license plate image data to the cache module.
Optionally, if the license plate image data meets the use condition of the license plate trajectory algorithm, identifying the driving direction of the vehicle by using the license plate trajectory algorithm includes:
obtaining license plate information of the license plate by adopting a license plate recognition algorithm on the license plate image data, wherein the license plate information comprises license plate position information and license plate text information;
judging whether the license plate image data is license plate image data of the same vehicle or not according to the license plate text information;
if the vehicle is the same vehicle, judging whether a sequence frame contained in the license plate image data meets the use condition of the license plate track algorithm, wherein the sequence frame is a set of single-frame license plate images acquired from the license plate image data;
and if so, identifying the driving direction of the vehicle by applying the license plate track algorithm to the license plate position information.
Optionally, the identifying the vehicle traveling direction of the vehicle by applying the license plate trajectory algorithm to the license plate position information includes:
acquiring a target sequence frame, wherein the target sequence frame is a set of single-frame license plate images which are acquired from license plate image data of a target vehicle and used for identifying the driving direction of the target vehicle;
acquiring target license plate position information in the target sequence frame, and obtaining a vehicle running track of the target vehicle according to the target license plate position information;
and determining the driving direction of the target vehicle according to the vehicle driving track.
Optionally, before the step of identifying the driving direction of the vehicle by using a target classification model if the license plate image data does not satisfy the use condition of the license plate trajectory algorithm, the method further includes:
acquiring different types of vehicle head and tail image data and vehicle traveling directions corresponding to the vehicle head and tail image data;
and taking the image data of the vehicle head and the vehicle tail of different types and the vehicle traveling direction corresponding to the image data of the vehicle head and the vehicle tail as training samples, and performing machine learning training on an initial classification model by using the training samples to obtain the target classification model, wherein the target corresponding relation between the image data of the vehicle head and the vehicle tail and the vehicle traveling direction is stored in the target classification model.
Optionally, if the license plate image data does not satisfy the use condition of the license plate trajectory algorithm, identifying the vehicle traveling direction by using a target classification model includes:
if the sequence frames contained in the license plate image data do not meet the use conditions of the license plate trajectory algorithm, acquiring a target single-frame license plate image, wherein the target single-frame license plate image is a single-frame license plate image which can be used for the target classification model in the license plate image data of a target vehicle;
and inputting the target single-frame license plate image into the target classification model to obtain the driving direction of the target vehicle output by the target classification model according to the target corresponding relation.
The embodiment of the application provides a traffic direction identification system, its characterized in that includes:
the acquisition unit is used for acquiring and storing license plate image data of the vehicle;
the recognition unit is used for recognizing the driving direction of the vehicle by applying a license plate track algorithm if the license plate image data meets the use condition of the license plate track algorithm;
the identification unit is further configured to identify a driving direction of the vehicle by using a target classification model if the license plate image data does not meet the use condition of the license plate trajectory algorithm, wherein the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the driving direction corresponding to the vehicle head and tail image data, and a target corresponding relation between the vehicle head and tail image data and the driving direction is stored in the target classification model.
Optionally, the obtaining unit is specifically configured to, when the monitoring module monitors a vehicle, obtain license plate image data of the vehicle by using the camera module, and store the license plate image data to the cache module.
Optionally, the recognition unit is specifically configured to obtain license plate information of the license plate by using a license plate recognition algorithm on the license plate image data, where the license plate information includes license plate position information and license plate text information;
judging whether the license plate image data is license plate image data of the same vehicle or not according to the license plate text information;
if the vehicle is the same vehicle, judging whether a sequence frame contained in the license plate image data meets the use condition of the license plate track algorithm, wherein the sequence frame is a set of single-frame license plate images acquired from the license plate image data;
and if so, identifying the driving direction of the vehicle by applying the license plate track algorithm to the license plate position information.
Optionally, the identification unit is specifically configured to acquire a target sequence frame, where the target sequence frame is a set of single-frame license plate images acquired from license plate image data of a target vehicle and used for identifying a vehicle-driving direction of the target vehicle;
acquiring target license plate position information in the target sequence frame, and obtaining a vehicle running track of the target vehicle according to the target license plate position information;
and determining the driving direction of the target vehicle according to the vehicle driving track.
Optionally, the acquiring unit is further configured to acquire different types of vehicle head and tail image data and a vehicle traveling direction corresponding to the vehicle head and tail image data;
and taking the image data of the vehicle head and the vehicle tail of different types and the vehicle traveling direction corresponding to the image data of the vehicle head and the vehicle tail as training samples, and performing machine learning training on an initial classification model by using the training samples to obtain the target classification model, wherein the target corresponding relation between the image data of the vehicle head and the vehicle tail and the vehicle traveling direction is stored in the target classification model.
Optionally, the identification unit is specifically configured to obtain a target single-frame license plate image if the sequence frames included in the license plate image data do not satisfy the use condition of the license plate trajectory algorithm, where the target single-frame license plate image is a single-frame license plate image that can be used in the target classification model and is included in the license plate image data of the target vehicle;
and inputting the target single-frame license plate image into the target classification model to obtain the driving direction of the target vehicle output by the target classification model according to the target corresponding relation.
The embodiment of the application provides a vehicle direction recognition device, its characterized in that includes:
the system comprises a central processing unit, a memory and an input/output interface;
the memory is a transient memory or a persistent memory;
the central processor is configured to communicate with the memory and execute the command operations in the memory to perform the aforementioned travel direction identification method.
The computer-readable storage medium provided by the embodiment of the application comprises instructions, and when the instructions are run on a computer, the instructions enable the computer to execute the method for identifying the driving direction.
According to the technical scheme, the embodiment of the application has the following advantages:
and judging whether the license plate track algorithm or the target classification model is used for identifying the vehicle running direction according to whether the acquired license plate image data meets the use condition of the license plate track algorithm, and improving the accuracy of the vehicle running direction identification result in a complementary combination mode of the two vehicle running directions so that a parking lot parking space management system makes a decision according to the vehicle running direction identification result and improves the use experience of a user.
Drawings
Fig. 1 is a schematic diagram of an implementation manner of a driving direction identification method provided in an embodiment of the present application;
FIG. 2 is a schematic diagram of license plate trajectory changes and corresponding coordinate systems;
fig. 3 is a schematic diagram of another implementation manner of a vehicle traveling direction identification method provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of an embodiment of a vehicle direction identification system according to an embodiment of the present disclosure;
fig. 5 is a schematic diagram of an implementation manner of a driving direction identification device according to an embodiment of the present application.
Detailed Description
The embodiment of the application provides a vehicle driving direction identification method, a vehicle driving direction identification system and a vehicle driving direction identification device, which are used for identifying the driving direction of a vehicle, are applied to a parking lot parking space management system and improve the use experience of a user.
The intelligent parking lot management system is a general name of modern parking lot vehicle charging and equipment automatic management, and is a high-tech electromechanical integrated product for completely placing a parking lot under the unified management of a computer. According to its design principle, intelligent parking lot management system can divide into three major parts: information acquisition and transmission, information processing and human-computer interface, and information storage and query. According to the design principle and the composition structure, the realizable functions are as follows: temporary vehicle charging, parking lot management, parking space guidance, reverse vehicle searching, special vehicle management, image comparison and the like.
With the maturity, application and popularization of vehicle detection and identification technologies and license plate identification technologies, vehicle detection devices and license plate identification devices have been widely used in places such as parking lot entrances and exits, commercial building entrances and exits, toll station entrances and exits, lane and parking stall monitoring and the like. The vehicle detection device and the license plate recognition device acquire the driving direction information and judge whether to lock or unlock the parking spot lock or whether to perform timing and charging operation according to the driving direction information. Most of the existing vehicle traveling direction identification methods are obtained by acquiring and analyzing the traveling track of a vehicle.
However, in an actual application scenario, when a vehicle is displaced in a parking space for a short distance or travels at a high speed in a certain displacement, it is difficult for the monitoring device to obtain an image sample that can be used for analyzing the vehicle traveling direction, and it is difficult to analyze the vehicle traveling direction, so that it is difficult for the system to accurately analyze the vehicle traveling situation and the parking space situation in the parking space management process. For example, in a real scene, the vehicle is faster when the vehicle is driven away from a parking space, a section with few deceleration strips or no gateway, and the like, so that the number plate image data captured by the camera is limited, and a sequence frame which can be used for judging the driving direction cannot be obtained. Or, due to night light or imaging problems of the acquisition device, spatial variation information of a section from the occurrence of a vehicle to the clear imaging of a license plate cannot be normally acquired, so that the position of the vehicle is too close to the camera when the license plate is clear, a sequence frame which can be used for judging the vehicle running direction cannot be acquired, and the vehicle running direction cannot be judged.
Based on this, the embodiment of the application provides a vehicle driving direction identification method combining a license plate track and a classification model, which is used for improving the accuracy of identifying the driving direction of a vehicle and improving the use experience of a user. Referring to fig. 1, an implementation manner of the driving direction identification method provided in the embodiment of the present application includes steps 101 to 103.
101. And acquiring and storing license plate image data of the vehicle.
When the monitoring module monitors that the vehicle enters a target monitoring area, a camera of the camera module is started to acquire license plate image data of the vehicle, and the license plate image data is stored in the cache module.
Specifically, the monitoring module may be implemented by a microwave radar ranging sensor, an infrared ranging sensor, or a laser ranging sensor, and is not limited herein. The camera in the camera module may capture an image in a target monitoring area, the license plate image data of the vehicle may be video image data of the vehicle or image data of the vehicle, and the cache module serves as an intermediate medium having a storage function in this embodiment of the application.
102. And if the license plate image data meets the use condition of the license plate track algorithm, identifying the driving direction of the vehicle by using the license plate track algorithm.
Analyzing and judging the acquired license plate image data of the target vehicle, firstly identifying the license plate information in the acquired license plate image data by using a license plate identification algorithm, confirming whether the vehicles in the same section of license plate image data are the same vehicle or not according to the license plate text information, namely the license plate number, in the license plate information, if the vehicles are the same vehicle, processing the license plate image data to obtain a sequence frame, specifically, performing single-frame interception on the license plate image data according to a certain time interval and splicing the sequence frame according to a time sequence to form a sequence frame, judging whether the acquired sequence frame meets the use condition of using a license plate track algorithm or not, for example, when the use condition of the license plate track algorithm is set that the length of the sequence frame is not less than 10 frames, judging whether the acquired sequence frame reaches 10 frames or not, and if the acquired sequence frame reaches 10 frames, confirming that the sequence frame is the target sequence frame, and (4) obtaining the running track of the target vehicle by applying a license plate track algorithm to the target sequence frame, thereby confirming the running direction. The requirement for the length of the sequence frame in the use condition of the license plate trajectory algorithm can be set according to the actual use condition, and is not limited here.
Specifically, license plate position information of a target vehicle can be confirmed by applying a license plate recognition algorithm to a license plate of each frame in a target sequence frame, a license plate central point can be obtained through the license plate position information, a vehicle traveling direction is judged according to a stable change trend of the license plate central point, under a general condition, a license plate track of a vehicle passing in and out is shown in fig. 2, a rectangular coordinate system is established by taking the upper left corner of a video picture which can be obtained by a camera as an origin, the whole y-axis value of the license plate central point in a vehicle approaching process is increased, the whole y-axis value of the license plate central point in a vehicle leaving process is decreased, and the whole change amplitude of the y-axis value of the license plate central point when the vehicle approaches to be static is smaller.
103. And if the license plate image data does not meet the use condition of the license plate trajectory algorithm, identifying the driving direction of the vehicle by using a target classification model.
Based on the sequence frames obtained in the step 102, if the obtained sequence frames do not satisfy the use condition of applying the license plate trajectory algorithm, for example, when the use condition of the license plate trajectory algorithm is set that the length of the sequence frame is not less than 10 frames, it is determined whether the obtained sequence frames reach 10 frames, if the obtained sequence frames do not reach 10 frames, the license plate trajectory algorithm cannot be used for identifying the vehicle driving direction, but a target classification model can be used for identifying a single-frame license plate image, so as to obtain the vehicle driving direction of the vehicle. The requirement for the length of the sequence frame in the use condition of the license plate trajectory algorithm can be set according to the actual use condition, and is not limited here.
The target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the vehicle traveling direction corresponding to the vehicle head and tail image data, and the target classification model stores the target corresponding relation between the vehicle head and tail image data and the vehicle traveling direction. Specifically, images of the front and the rear of a car, a passenger car and a truck can be obtained first, images of the front and the rear of the car with unobvious features are additionally added to serve as a category, the total number of the categories is 7, the images include regular-size images, inclined large-angle images, local images of the front and the rear of the car serve as image data of the front and the rear of the car, the driving directions corresponding to the image data of the front and the rear of the car serve as training samples, machine learning training is carried out on an initial classification model to obtain a target classification model, generally speaking, the driving directions are determined to be close through the obtained images of the front of the car, the driving directions are determined to be far away through the obtained images of the rear of the car, and the driving directions are determined through the target classification model according to whether the obtained single-frame license plate images of the front of the car or the images of the parking spaces.
In the embodiment, the obtained license plate image data is judged to be used for identifying the vehicle running direction by using the license plate track algorithm or the target classification model according to whether the license plate image data meets the use condition of the license plate track algorithm, and the accuracy of the vehicle running direction identification result is improved in a complementary combination mode of the two vehicle running directions, so that a parking lot parking space management system makes a decision according to the vehicle running direction identification result, and the use experience of a user is improved.
Referring to fig. 3, another implementation manner of the method for identifying a driving direction provided in the embodiment of the present application includes steps 301 to 306.
301. And acquiring a target classification model.
In the embodiment of the application, the target classification model is a neural network model used for obtaining the vehicle traveling direction according to the license plate image data, the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the vehicle traveling direction corresponding to the vehicle head and tail image data, and the target corresponding relation between the vehicle head and tail image data and the vehicle traveling direction is stored in the target classification model. Specifically, images of the front and the rear of a car, a passenger car and a truck can be obtained first, images of the front and the rear of the car with unobvious features are additionally added to serve as a category, the total number of the categories is 7, the images include regular-size images, inclined large-angle images, local images of the front and the rear of the car serve as image data of the front and the rear of the car, the driving directions corresponding to the image data of the front and the rear of the car serve as training samples, machine learning training is carried out on an initial classification model to obtain a target classification model, generally speaking, the driving directions are determined to be close through the obtained images of the front of the car, the driving directions are determined to be far away through the obtained images of the rear of the car, and the driving directions are determined through the target classification model according to whether the obtained single-frame license plate images of the front of the car or the images of the parking spaces.
302. And acquiring and storing license plate image data of the vehicle.
When the monitoring module monitors that the vehicle enters a target monitoring area, a camera of the camera module is started to acquire license plate image data of the vehicle, and the license plate image data is stored in the cache module.
Specifically, the monitoring module may be implemented by a microwave radar ranging sensor, an infrared ranging sensor, or a laser ranging sensor, and is not limited herein. The camera in the camera module may capture an image in a target monitoring area, the license plate image data of the vehicle may be video image data of the vehicle or image data of the vehicle, and the cache module serves as an intermediate medium having a storage function in this embodiment of the application.
303. And obtaining the license plate information of the license plate by adopting a license plate recognition algorithm on the license plate image data.
The acquired license plate image data of the target vehicle is analyzed and judged, license plate information in the acquired license plate image data can be firstly identified by a license plate identification algorithm, the license plate information comprises license plate position information and license plate text information, and whether the vehicles in the same section of license plate image data are the same vehicle or not is confirmed according to the license plate text information, namely the license plate number, in the license plate information.
304. And judging whether the license plate image data meets the use condition of a license plate track algorithm or not according to the license plate information.
According to license plate text information, namely license plate numbers, in license plate information, whether vehicles in the same section of license plate image data are the same vehicle is confirmed, if the vehicles are the same vehicle, the license plate image data are processed to obtain sequence frames, specifically, the license plate image data are subjected to single-frame interception according to a certain time interval and spliced according to a time sequence to form the sequence frames, whether the obtained sequence frames meet the use condition of applying a license plate track algorithm is judged, for example, when the use condition of the license plate track algorithm is set to be that the length of the sequence frames is not less than 10 frames, whether the obtained sequence frames reach 10 frames is judged, if the obtained sequence frames reach 10 frames, the obtained sequence frames are confirmed to be target sequence frames, and the license plate track algorithm is applied to the target sequence frames to obtain the running track of the target vehicle, so that the vehicle running direction is confirmed. The requirement for the length of the sequence frame in the use condition of the license plate trajectory algorithm can be set according to the actual use condition, and is not limited here.
305. And if so, identifying the driving direction of the vehicle by using a license plate track algorithm.
And if the using condition of the license plate track algorithm is met, identifying the driving direction of the vehicle by using the license plate track algorithm.
Specifically, license plate position information of a target vehicle can be confirmed by applying a license plate recognition algorithm to a license plate of each frame in a target sequence frame, a license plate central point can be obtained through the license plate position information, a vehicle traveling direction is judged according to a stable change trend of the license plate central point, under a general condition, a license plate track of a vehicle passing in and out is shown in fig. 2, a rectangular coordinate system is established by taking the upper left corner of a video picture which can be obtained by a camera as an origin, the whole y-axis value of the license plate central point in a vehicle approaching process is increased, the whole y-axis value of the license plate central point in a vehicle leaving process is decreased, and the whole change amplitude of the y-axis value of the license plate central point when the vehicle approaches to be static is smaller.
306. And if not, identifying the driving direction of the vehicle by using the target classification model.
Based on the sequence frames obtained in the step 304, if the obtained sequence frames do not satisfy the use condition of applying the license plate trajectory algorithm, for example, when the use condition of the license plate trajectory algorithm is set that the length of the sequence frames is not less than 10 frames, it is determined whether the obtained sequence frames reach 10 frames, if the obtained sequence frames do not reach 10 frames, the license plate trajectory algorithm cannot be used for identifying the vehicle driving direction, but a target classification model can be used for identifying a single-frame license plate image, so as to obtain the vehicle driving direction of the vehicle. The requirement for the length of the sequence frame in the use condition of the license plate trajectory algorithm can be set according to the actual use condition, and is not limited here.
And inputting the single-frame license plate image into the target classification model, and outputting the driving direction.
In the embodiment, the obtained license plate image data is judged to be used for identifying the vehicle running direction by using the license plate track algorithm or the target classification model according to whether the license plate image data meets the use condition of the license plate track algorithm, and the accuracy of the vehicle running direction identification result is improved in a complementary combination mode of the two vehicle running directions, so that a parking lot parking space management system makes a decision according to the vehicle running direction identification result, and the use experience of a user is improved.
Referring to fig. 4, a driving direction recognition system provided in an embodiment of the present application includes:
an obtaining unit 401, configured to obtain and store license plate image data of a vehicle;
a recognition unit 402, configured to recognize a driving direction of the vehicle by using a license plate trajectory algorithm if the license plate image data meets a use condition of the license plate trajectory algorithm;
the identifying unit 402 is further configured to identify a vehicle traveling direction of the vehicle by using a target classification model if the license plate image data does not satisfy the use condition of the license plate trajectory algorithm, where the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the vehicle traveling direction corresponding to the vehicle head and tail image data, and a target corresponding relationship between the vehicle head and tail image data and the vehicle traveling direction is stored in the target classification model.
The obtaining unit 401 is specifically configured to, when the monitoring module monitors a vehicle, obtain license plate image data of the vehicle by using the camera module, and store the license plate image data in the cache module.
The recognition unit 402 is specifically configured to obtain license plate information of the license plate by using a license plate recognition algorithm on the license plate image data, where the license plate information includes license plate position information and license plate text information;
judging whether the license plate image data is license plate image data of the same vehicle or not according to the license plate text information;
if the vehicle is the same vehicle, judging whether a sequence frame contained in the license plate image data meets the use condition of the license plate track algorithm, wherein the sequence frame is a set of single-frame license plate images acquired from the license plate image data;
and if so, identifying the driving direction of the vehicle by applying the license plate track algorithm to the license plate position information.
The identifying unit 402 is specifically configured to acquire a target sequence frame, where the target sequence frame is a set of single-frame license plate images acquired from license plate image data of a target vehicle and used for identifying a driving direction of the target vehicle;
acquiring target license plate position information in the target sequence frame, and obtaining a vehicle running track of the target vehicle according to the target license plate position information;
and determining the driving direction of the target vehicle according to the vehicle driving track.
The acquiring unit 401 is further configured to acquire different types of vehicle head and tail image data and a vehicle traveling direction corresponding to the vehicle head and tail image data;
and taking the image data of the vehicle head and the vehicle tail of different types and the vehicle traveling direction corresponding to the image data of the vehicle head and the vehicle tail as training samples, and performing machine learning training on an initial classification model by using the training samples to obtain the target classification model, wherein the target corresponding relation between the image data of the vehicle head and the vehicle tail and the vehicle traveling direction is stored in the target classification model.
The recognition unit 402 is specifically configured to obtain a target single-frame license plate image if a sequence frame included in the license plate image data does not satisfy a use condition of the license plate trajectory algorithm, where the target single-frame license plate image is a single-frame license plate image that can be used in the target classification model in the license plate image data of the target vehicle;
and inputting the target single-frame license plate image into the target classification model to obtain the driving direction of the target vehicle output by the target classification model according to the target corresponding relation.
The functions and processes executed by the components in the driving direction identification system of this embodiment are similar to those executed by the components in fig. 1 and 3, and are not described again here.
Fig. 5 is a schematic diagram of a vehicle traveling direction recognition apparatus according to an embodiment of the present disclosure, where the vehicle traveling direction recognition apparatus 500 may include one or more Central Processing Units (CPUs) 501 and a memory 505, and one or more applications or data are stored in the memory 505.
The traffic direction identification device 500 may also include one or more power supplies 502, one or more wired or wireless network interfaces 503, one or more input-output interfaces 504, and/or one or more operating systems, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.
The central processing unit 501 may perform the operations performed by the driving direction identification system in the embodiments shown in fig. 1 and fig. 3, and details thereof are not repeated herein.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.
Claims (10)
1. A method for identifying a traveling direction of a vehicle, comprising:
acquiring and storing license plate image data of a vehicle;
if the license plate image data meets the use condition of the license plate track algorithm, identifying the driving direction of the vehicle by using the license plate track algorithm;
and if the license plate image data does not meet the use conditions of the license plate trajectory algorithm, identifying the vehicle traveling direction of the vehicle by using a target classification model, wherein the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the vehicle traveling direction corresponding to the vehicle head and tail image data, and the target classification model stores the target corresponding relation between the vehicle head and tail image data and the vehicle traveling direction.
2. The method according to claim 1, wherein the acquiring and storing license plate image data of the vehicle comprises:
when the monitoring module monitors a vehicle, the camera module acquires license plate image data of the vehicle and stores the license plate image data to the cache module.
3. The method of claim 1, wherein if the license plate image data meets a condition for using a license plate trajectory algorithm, identifying the vehicle direction using the license plate trajectory algorithm comprises:
obtaining license plate information of the license plate by adopting a license plate recognition algorithm on the license plate image data, wherein the license plate information comprises license plate position information and license plate text information;
judging whether the license plate image data is license plate image data of the same vehicle or not according to the license plate text information;
if the vehicle is the same vehicle, judging whether a sequence frame contained in the license plate image data meets the use condition of the license plate track algorithm, wherein the sequence frame is a set of single-frame license plate images acquired from the license plate image data;
and if so, identifying the driving direction of the vehicle by applying the license plate track algorithm to the license plate position information.
4. The method of claim 3, wherein the identifying the vehicle traveling direction of the vehicle by applying the license plate trajectory algorithm to the license plate position information comprises:
acquiring a target sequence frame, wherein the target sequence frame is a set of single-frame license plate images which are acquired from license plate image data of a target vehicle and used for identifying the driving direction of the target vehicle;
acquiring target license plate position information in the target sequence frame, and obtaining a vehicle running track of the target vehicle according to the target license plate position information;
and determining the driving direction of the target vehicle according to the vehicle driving track.
5. The method of claim 1, wherein before identifying the vehicle heading using the target classification model if the license plate image data does not satisfy the use condition of the license plate tracking algorithm, the method further comprises:
acquiring different types of vehicle head and tail image data and vehicle traveling directions corresponding to the vehicle head and tail image data;
and taking the image data of the vehicle head and the vehicle tail of different types and the vehicle traveling direction corresponding to the image data of the vehicle head and the vehicle tail as training samples, and performing machine learning training on an initial classification model by using the training samples to obtain the target classification model, wherein the target corresponding relation between the image data of the vehicle head and the vehicle tail and the vehicle traveling direction is stored in the target classification model.
6. The method of claim 3, wherein if the license plate image data does not satisfy the use condition of the license plate trajectory algorithm, identifying the vehicle traveling direction using the target classification model comprises:
if the sequence frames contained in the license plate image data do not meet the use conditions of the license plate trajectory algorithm, acquiring a target single-frame license plate image, wherein the target single-frame license plate image is a single-frame license plate image which can be used for the target classification model in the license plate image data of a target vehicle;
and inputting the target single-frame license plate image into the target classification model to obtain the driving direction of the target vehicle output by the target classification model according to the target corresponding relation.
7. A vehicle traveling direction recognition system, comprising:
the acquisition unit is used for acquiring and storing license plate image data of the vehicle;
the recognition unit is used for recognizing the driving direction of the vehicle by applying a license plate track algorithm if the license plate image data meets the use condition of the license plate track algorithm;
the identification unit is further configured to identify a driving direction of the vehicle by using a target classification model if the license plate image data does not meet the use condition of the license plate trajectory algorithm, wherein the target classification model is obtained by performing machine learning training on different types of vehicle head and tail image data and the driving direction corresponding to the vehicle head and tail image data, and a target corresponding relation between the vehicle head and tail image data and the driving direction is stored in the target classification model.
8. The vehicle traveling direction identification method according to claim 7, wherein the obtaining unit is specifically configured to, when the monitoring module monitors a vehicle, obtain license plate image data of the vehicle by using the camera module, and store the license plate image data in the cache module.
9. A traveling direction recognition apparatus, characterized by comprising:
the system comprises a central processing unit, a memory and an input/output interface;
the memory is a transient memory or a persistent memory;
the central processor is configured to communicate with the memory and execute the operations of the instructions in the memory to perform the method of any of claims 1 to 6.
10. A computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111632979.0A CN114170810A (en) | 2021-12-28 | 2021-12-28 | Vehicle traveling direction identification method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111632979.0A CN114170810A (en) | 2021-12-28 | 2021-12-28 | Vehicle traveling direction identification method, system and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114170810A true CN114170810A (en) | 2022-03-11 |
Family
ID=80488382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111632979.0A Pending CN114170810A (en) | 2021-12-28 | 2021-12-28 | Vehicle traveling direction identification method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114170810A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105632175A (en) * | 2016-01-08 | 2016-06-01 | 上海微锐智能科技有限公司 | Vehicle behavior analysis method and system |
| CN108305466A (en) * | 2018-03-13 | 2018-07-20 | 北京智芯原动科技有限公司 | Roadside Parking detection method and device based on vehicle characteristics analysis |
| CN110766009A (en) * | 2019-10-31 | 2020-02-07 | 深圳市捷顺科技实业股份有限公司 | Tail plate identification method and device and computer readable storage medium |
| US20200193721A1 (en) * | 2018-12-17 | 2020-06-18 | Eps Company | Method for providing parking service using image grouping-based vehicle identification |
| CN111339834A (en) * | 2020-02-04 | 2020-06-26 | 浙江大华技术股份有限公司 | Method for recognizing vehicle traveling direction, computer device, and storage medium |
| CN111383460A (en) * | 2020-06-01 | 2020-07-07 | 浙江大华技术股份有限公司 | Vehicle state discrimination method and device and computer storage medium |
| CN111583703A (en) * | 2020-04-29 | 2020-08-25 | 济南博观智能科技有限公司 | Parking lot access control system and method |
| CN112289040A (en) * | 2020-11-25 | 2021-01-29 | 浙江大华技术股份有限公司 | Method and device for identifying vehicle driving direction and storage medium |
-
2021
- 2021-12-28 CN CN202111632979.0A patent/CN114170810A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105632175A (en) * | 2016-01-08 | 2016-06-01 | 上海微锐智能科技有限公司 | Vehicle behavior analysis method and system |
| CN108305466A (en) * | 2018-03-13 | 2018-07-20 | 北京智芯原动科技有限公司 | Roadside Parking detection method and device based on vehicle characteristics analysis |
| US20200193721A1 (en) * | 2018-12-17 | 2020-06-18 | Eps Company | Method for providing parking service using image grouping-based vehicle identification |
| CN110766009A (en) * | 2019-10-31 | 2020-02-07 | 深圳市捷顺科技实业股份有限公司 | Tail plate identification method and device and computer readable storage medium |
| CN111339834A (en) * | 2020-02-04 | 2020-06-26 | 浙江大华技术股份有限公司 | Method for recognizing vehicle traveling direction, computer device, and storage medium |
| CN111583703A (en) * | 2020-04-29 | 2020-08-25 | 济南博观智能科技有限公司 | Parking lot access control system and method |
| CN111383460A (en) * | 2020-06-01 | 2020-07-07 | 浙江大华技术股份有限公司 | Vehicle state discrimination method and device and computer storage medium |
| CN112289040A (en) * | 2020-11-25 | 2021-01-29 | 浙江大华技术股份有限公司 | Method and device for identifying vehicle driving direction and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Satzoda et al. | Multipart vehicle detection using symmetry-derived analysis and active learning | |
| US9336450B2 (en) | Methods and systems for selecting target vehicles for occupancy detection | |
| Sivaraman et al. | Vehicle detection by independent parts for urban driver assistance | |
| CN106295541A (en) | Vehicle type recognition method and system | |
| US9626599B2 (en) | Reconfigurable clear path detection system | |
| CN105825185A (en) | Early warning method and device against collision of vehicles | |
| CN110517500B (en) | Man-vehicle association processing method and device | |
| Jain et al. | Performance analysis of object detection and tracking algorithms for traffic surveillance applications using neural networks | |
| KR102105954B1 (en) | System and method for accident risk detection | |
| CN106257490A (en) | The method and system of detection driving vehicle information | |
| CN104615986A (en) | Method for utilizing multiple detectors to conduct pedestrian detection on video images of scene change | |
| CN111444843B (en) | Multimode driver and vehicle illegal behavior monitoring method and system | |
| Radiuk et al. | Convolutional neural network for parking slots detection | |
| CN111507196A (en) | Vehicle type identification method based on machine vision and deep learning | |
| CN113408364B (en) | Temporary license plate recognition method, system, device and storage medium | |
| Tumas et al. | Acceleration of HOG based pedestrian detection in FIR camera video stream | |
| WO2023071874A1 (en) | Roadside assistance working node determining method and apparatus, electronic device, and storage medium | |
| CN114170810A (en) | Vehicle traveling direction identification method, system and device | |
| Priya et al. | Intelligent parking system | |
| CN114973169A (en) | Vehicle classification counting method and system based on multi-target detection and tracking | |
| CN111161542B (en) | Vehicle identification method and device | |
| CN103366598A (en) | Vehicle in-place parking detection system and detecting method | |
| CN114528253A (en) | Intelligent internet automobile public road dangerous scene extraction method and device, dangerous scene construction method and device and computing equipment | |
| CN111639640A (en) | License plate recognition method, device and equipment based on artificial intelligence | |
| Lakshmanan et al. | Deep Learning Based Vehicle Tracking System Using License Plate Detection And Recognition |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220311 |
|
| RJ01 | Rejection of invention patent application after publication |