CN114078212A

CN114078212A - Accurate vehicle type identification method and device based on ETC portal

Info

Publication number: CN114078212A
Application number: CN202111393051.1A
Authority: CN
Inventors: 王海峰; 席方凯
Original assignee: Beijing Jvsh Technology Co ltd
Current assignee: Beijing Jvsh Technology Co ltd
Priority date: 2021-11-23
Filing date: 2021-11-23
Publication date: 2022-02-22

Abstract

The invention provides a precise vehicle type recognition method and device based on an ETC portal, wherein the vehicle type recognition device comprises: a first image acquisition device for acquiring vehicle front face image information on a lane; a second image acquisition device for acquiring vehicle side image information on the lane; the vehicle identification module is used for obtaining vehicle basic information according to the front image information, and the side axle number identification module is used for obtaining the number of side wheels according to the side image information; determining the vehicle type according to the vehicle basic information and the number of the side wheels; this application is with traditional license plate discernment, and automobile body colour discernment, vehicle brand discernment combine the concatenation of side vehicle, and on the ETC portal was applied to technologies such as axle number detection, can carry out omnidirectional vehicle information acquisition to the vehicle of high-speed operation, can carry out accurate discernment to the motorcycle type of vehicle, for subsequent vehicle fee evasion, vehicle inspection provides effectual evidence.

Description

Accurate vehicle type identification method and device based on ETC portal

Technical Field

The invention relates to the technical field of vehicle monitoring, in particular to a precise vehicle type identification method and device based on an ETC portal, electronic equipment and a storage medium.

Background

In the prior art, an electronic Toll Collection system (etc) has been widely applied to Toll stations of highways, bridges, tunnels and the like, and a vehicle-mounted end of the automatic Toll Collection system is relatively simple. The toll station adopts ETC electronic toll collection technology, so that the traffic capacity of the highway toll station can be improved, and the problem of congestion of the toll station is relieved and even solved; meanwhile, the ETC passing vehicle can be charged and flexibly charged according to an accurate driving path through the ETC identification station, the road network flow of the expressway is balanced, and the operation efficiency of the expressway is improved. The ETC system engineering can also provide basic conditions for traffic information service, urban parking management, major activities, regional traffic management and urban traffic comprehensive management.

The number of axles is an important basis for highway billing, and therefore, when a vehicle arrives at a highway toll station, the number of axles of the vehicle needs to be detected. The mode based on vehicle brand carries out model identification among the prior art, but this kind of mode can only distinguish passenger train freight train, to the freight train, especially 2 axle cars to 6 axle cars, because 2 axle cars to 6 axle cars are all the consistent cross border cars of brand, consequently can't carry out further accurate differentiation to the freight train.

Disclosure of Invention

The embodiment of the invention provides an ETC portal-based accurate vehicle type identification method, an ETC portal-based accurate vehicle type identification device, electronic equipment and a storage medium.

In a first aspect, an embodiment of the present invention provides an ETC portal-based accurate vehicle type recognition apparatus, including:

a first image acquisition device for acquiring vehicle front face image information on a lane;

a second image acquisition device for acquiring vehicle side image information on the lane;

the vehicle identification module is used for obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand;

the side axle number identification module is used for obtaining the number of side wheels according to the side image information; determining the vehicle type according to the vehicle basic information and the number of the side wheels;

the first image acquisition equipment and the second image acquisition equipment are both configured on the ETC portal frame and correspond to the lane.

As a possible implementation manner, the vehicle type recognition apparatus further includes:

stroboscopic light filling lamp, first image acquisition equipment and second image acquisition equipment single lane are installed on the ETC portal.

As a possible implementation, the second image acquisition device protrudes out of the ETC portal and is mounted and fixed in sequence at 90 ° to the lane and facing the adjacent lane.

As one possible implementation, the vehicle identification module includes:

the vehicle detection unit is used for detecting vehicles by adopting a deep learning-based method so as to identify different vehicles in the same image;

and the vehicle body color recognition and vehicle brand recognition unit is used for recognizing license plates and vehicle types based on the vehicle detected by the vehicle detection unit.

As a possible implementation manner, the side shaft number identification module includes:

the vehicle splicing unit is used for splicing a plurality of vehicle side images to obtain a complete image;

and the axle number detection unit is used for detecting the wheels of the vehicle by adopting a deep learning detection algorithm aiming at the complete image and finally outputting the number of the wheels.

As one possible implementation, the vehicle splicing unit includes:

a splicing judgment subunit, wherein if the current frame finds a moving vehicle, the current frame state output state is the splicing start state;

the splicing processing subunit selects an image with the length of len on the right side of the previous frame image for matching, wherein the left coordinate is width-len, the right coordinate is width, and the width of the image is width;

in the set area of the current frame image, left value is width-len 3, right value is width-len, moving from left to right, calculating the image matching fraction, and firstly calculating whether the corresponding pixel points are matched by using the following formula:

|R1-R2|+|G1-G2|+|B1-B2|<thr1

wherein, thr1 takes the value of 20, if matching succeeds, the number sum _ same of successfully matched points is calculated, and the following fraction is calculated:

scores＝(sum_same/sum_pixel)*100

listing from left to right, and selecting a position index with the largest score as a matching position;

and intercepting the current frame image according to left index and right width, and splicing the intercepted image to the right side of the image splicing image of the previous frame.

and the fusion matching module is used for matching the time of acquiring the front image information of the vehicle by the first image acquisition equipment and the time of acquiring the side image information of the vehicle by the second image acquisition equipment, and reporting the identification result in a unified way after the matching is successful.

As one possible implementation manner, the fusion matching module includes:

the time correction unit is used for adjusting the time of the first image acquisition device, the time of the second image acquisition device and the time of the server for running the program to be the same time;

the matching unit, the first image acquisition equipment snapshot time T1 and the side vehicle image splicing start time T2 meet the following conditions, and then matching is carried out:

T2–T1＞thr1

T2–T1＜thr2

wherein thr1 and thr2 are matched time thresholds and take values of 2s and 8s respectively.

And the fusion result reporting unit is used for reporting the result after fusing the result according to a preset rule.

In a second aspect, an embodiment of the present invention provides a precision vehicle type identification method based on an ETC portal, where the vehicle type identification method includes:

acquiring front image information of a vehicle on a lane through first image acquisition equipment, and acquiring side image information of the vehicle through second image acquisition equipment;

obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand;

obtaining the number of side wheels according to the side image information;

determining the vehicle type according to the vehicle basic information and the number of the side wheels;

the first image acquisition equipment and the second image acquisition equipment are both arranged on the ETC portal frame and correspond to the lane.

As a possible implementation manner, the vehicle type identification method further includes matching time for acquiring the front image information of the vehicle by the first image acquisition device and time for acquiring the side image information of the vehicle by the second image acquisition device, and reporting the identification result in a unified manner after matching is successful.

In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, where the memory stores a computer program thereon, and the processor implements the method according to any one of the second aspect when executing the program.

In a fourth aspect, an embodiment of the invention provides a computer-readable storage medium on which a computer program is stored, which, when executed by a processor, implements the method according to any one of the second aspects.

The embodiment of the invention provides an ETC portal-based accurate vehicle type recognition device, which comprises: a first image acquisition device for acquiring vehicle front face image information on a lane; a second image acquisition device for acquiring vehicle side image information on the lane; the vehicle identification module is used for obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand; the side axle number identification module is used for obtaining the number of side wheels according to the side image information; determining the vehicle type according to the vehicle basic information and the number of the side wheels; the first image acquisition equipment and the second image acquisition equipment are both configured on the ETC portal frame and correspond to the lane. This application is with traditional license plate discernment, automobile body colour discernment, vehicle brand discernment combines side vehicle concatenation, and on the high-speed portal was applied to techniques such as axle number detection, can carry out omnidirectional vehicle information acquisition to the vehicle of high-speed operation, mainly include the vehicle front view, the side view, the license plate number, the automobile body colour, brand information, axle number information can carry out accurate discernment to the motorcycle type of vehicle, escapes for subsequent vehicle, and vehicle inspection provides effectual evidence.

It should be understood that the statements herein reciting aspects are not intended to limit the critical or essential features of any embodiment of the invention, nor are they intended to limit the scope of the invention. Other features of the present invention will become apparent from the following description.

Drawings

In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and that other drawings can be obtained by those skilled in the art without inventive exercise.

Fig. 1 shows an installation diagram of an ETC portal-based precision vehicle type recognition apparatus according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram illustrating an ETC portal-based accurate vehicle type recognition device according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an ETC portal-based precision vehicle type recognition device according to another embodiment of the present invention;

fig. 4 is a flowchart illustrating a precise vehicle type identification method based on an ETC portal according to an embodiment of the present invention;

fig. 5 shows a block diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in one or more embodiments of the present disclosure, the technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in one or more embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from one or more of the embodiments described herein without making any inventive step shall fall within the scope of protection of this document.

In the related art, with the reform of the vehicle charging standard in the high-speed field, the axle type charging is adopted in the past, the axle number charging is adopted at present, the axle number is the number of vehicle wheels counted through the side face, and the current charging types are divided into 2-axle passenger cars, 2-axle trucks, 3-axle cars, 4-axle cars, 5-axle cars and 6-axle cars according to the axle number.

At present, a sectional charging scheme is adopted for high-speed charging, wherein the vehicle type identification aiming at an ETC portal frame is mostly based on the brand identification of a front vehicle and the vehicle type is reversely pushed, but the scheme cannot accurately distinguish 2-to-6-axle vehicles.

It should be noted that, the description of the embodiment of the present invention is only for clearly illustrating the technical solutions of the embodiment of the present invention, and does not limit the technical solutions provided by the embodiment of the present invention.

Fig. 1 shows an installation schematic diagram of a precise vehicle type recognition device based on an ETC portal according to an embodiment of the invention. Fig. 2 is a schematic structural diagram illustrating an ETC portal-based accurate vehicle type recognition device according to an embodiment of the present invention;

referring to fig. 1-2, the vehicle type recognition apparatus includes:

a first image pickup device 10 for picking up image information of a front face of a vehicle on a lane;

a second image acquiring device 20 for acquiring vehicle side image information on the lane;

the vehicle identification module 101 is used for obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand;

the side axle number identification module 102 is used for obtaining the number of side wheels according to the side image information; determining the vehicle type according to the vehicle basic information and the number of the side wheels;

the first image acquiring device 10 and the second image acquiring device 20 are both disposed on the ETC portal 30 and correspond to a lane.

The accurate motorcycle type recognition device based on ETC portal that this embodiment provided includes: a first image acquisition device for acquiring vehicle front face image information on a lane; a second image acquisition device for acquiring vehicle side image information on the lane; the vehicle identification module is used for obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand; the side axle number identification module is used for obtaining the number of side wheels according to the side image information; determining the vehicle type according to the vehicle basic information and the number of the side wheels; the first image acquisition equipment and the second image acquisition equipment are both arranged on the ETC portal frame and correspond to the lane; the vehicle identification device has the advantages that the traditional license plate identification, the vehicle color identification, the vehicle brand identification and side vehicle splicing, the axle number detection and other technologies are applied to the high-speed portal, the vehicle running at high speed can be subjected to all-around vehicle information collection, the vehicle identification device mainly comprises a vehicle front view, a side view, a license plate number, the vehicle color, the brand information and the axle number information, the vehicle type of the vehicle can be accurately identified, and effective evidence is provided for subsequent vehicle fee evasion and vehicle inspection.

In some embodiments, the vehicle type recognition apparatus further includes:

stroboscopic light filling lamp, first image acquisition equipment 10 and second image acquisition equipment 20 single lane are installed on ETC portal 30. The single lane installation has the characteristics of high vehicle pixel quality, high identification accuracy rate and the like. First image acquisition equipment 10, second image acquisition equipment 20 and stroboscopic light filling lamp need install in the lane directly over the intermediate position, and every lane corresponds one set, and snapshot equipment stretches out the portal and perpendicular 90 faces adjacent lane and installs in proper order fixedly.

Fig. 3 is a schematic structural diagram of an ETC portal-based precision vehicle type recognition device according to another embodiment of the present invention; as shown in fig. 3, the vehicle identification module 101 includes:

a vehicle detection unit 1011 for performing vehicle detection by using a deep learning-based method to identify different vehicles in the same image; for example, a deep learning-based method can be adopted for vehicle detection, and the method can adopt fast-rcnn, ssd, yolo, and can also adopt haar + adaboost, hog + svm and other algorithms.

A vehicle body color recognition and vehicle brand recognition unit 1012 that performs license plate and vehicle type recognition based on the vehicle detected by the vehicle detection unit; for example, a traditional algorithm or a deep learning algorithm can be adopted for license plate recognition, and a vehicle type recognition adopts a surface model, a resnet model, a mobilenet model and the like; meanwhile, the method also comprises the step of judging the number plate shielding.

Specifically, the side axis number identification module 102 includes:

the vehicle splicing unit 1021 is used for splicing a plurality of vehicle side images to obtain a complete image;

the axle number detecting unit 1022 detects the wheels of the vehicle by using a deep learning detection algorithm with respect to the complete image, and finally outputs the number of the wheels. For the spliced map, for example, a deep learning detection algorithm may be adopted to detect the wheels of the vehicle, and finally output the number of the wheels, and the detection method may adopt a method such as fast-rcnn, ssd, yolov3, or the like.

Specifically, the vehicle splicing unit 1021 includes:

a splicing judgment subunit 10211, configured to output a current frame state as a splicing start state if the current frame finds a moving vehicle;

specifically, the method of discriminating a moving vehicle is as follows:

calculating the sum of the number of moving pixel points of each column of the moving image, then traversing from left to right, counting the sum of the number of moving pixels of each three columns, and if the following conditions are met, indicating that the vehicle is found:

sum>H*R

sum is the sum of the current three rows of motion pixels, H is the height of the motion image, and R is a proportional threshold which can be 1.2; at the same time, the time to start the matching is output.

The splicing processing subunit 10212 selects an image with the length len at the right side of the previous frame image for matching, wherein the left coordinate is width-len, the right coordinate is width, and the width of the image is width;

specifically, the matching calculation method includes moving the selected image area from left to right in a set area (left value is width-len 3, right value is width-len) of the current frame image, calculating an image matching score, and firstly calculating whether corresponding pixel points are matched or not by using the following formula:

|R1-R2|+|G1-G2|+|B1-B2|<thr1

wherein thr1 can be 20, if the above conditions are satisfied, the matching of this point is successful, the number sum _ same of the successfully matched points is calculated, and the following fraction is calculated:

scores＝(sum_same/sum_pixel)*100

and (5) listing from left to right, and selecting the position index with the maximum score as a matching position.

The image splicing method is that the current frame image is cut according to left as index and right as width, the cut images are spliced to the right side of the image splicing image of the previous frame, and if no moving vehicle is found, the splicing is finished.

It should be noted that, the motion image calculation method may also adopt a mixed gaussian method to calculate whether each pixel point is a motion pixel; the image matching method can also adopt a complex matching method based on the characteristic points, such as surf, sift and the like.

As a preferred embodiment, the vehicle type recognition apparatus further includes:

and the fusion matching module 103 is used for matching the time of acquiring the front image information of the vehicle by the first image acquisition device 10 with the time of acquiring the side image information of the vehicle by the second image acquisition device 20, and reporting the identification result in a unified manner after the matching is successful.

Specifically, the fusion matching module 103 includes:

a time adjustment unit 1031 that adjusts the server time of the first image acquisition device, the server time of the second image acquisition device, and the running program to the same time;

the matching unit 1032, the first image capturing device snapshot time T1, and the side vehicle image splicing start time T2 meet the following conditions, and perform matching:

T2–T1＞thr1

T2–T1＜thr2

A fusion result reporting unit 1033 for reporting the result after fusing the result according to a preset rule.

Taking the first image acquisition device 10 and the second image acquisition device 20 as top-mounted cameras and side-mounted cameras as examples; the fusion matching module 103 performs the following operations:

one) time correction: and adjusting the time of the top camera, the side camera and the server running the program to the same time.

Two) matching: because the top mounted camera takes a candid photograph earlier, splices after the side concatenation camera, so with top mounted camera candid photograph time T1, and the side vehicle begins concatenation time T2, satisfies following condition, then matches:

T2–T1＞thr1

T2–T1＜thr2

the time thresholds for which thr1 and thr2 are matched may be 2 and 8, respectively, in seconds. The top-mounted camera can snap through setting an image snapping line, for example, a head target of a tracked car or truck is judged to pass through the snapping line, a snapping signal is given, the snapping signal is transmitted to the top-mounted camera, a front picture of the car is snapped, and a license plate is recognized.

And thirdly), reporting a fusion result: fusing the results according to the following rules, and then reporting:

1) judging according to the vehicle brand identified by the top-mounted snapshot camera, and if the vehicle is a passenger car, outputting the vehicle types of passenger 1, passenger 2, passenger 3 and passenger 4 according to rules corresponding to the brands;

2) if the truck is identified by the vehicle brand, then the delivery is goods 1, goods 2, goods 3, goods 4, goods 5 and goods 6 based on the number of wheels detected in the side mosaic.

Based on the same inventive concept, the embodiment of the invention also provides a precise vehicle type identification method based on the ETC portal, which can be used for realizing the precise vehicle type identification device based on the ETC portal described in the embodiment, and is described in the embodiment below. Because the principle of this accurate motorcycle type recognition device solution problem based on ETC portal is similar with the accurate motorcycle type recognition device based on ETC portal, consequently the implementation of the accurate motorcycle type recognition method based on ETC portal can refer to the implementation of an accurate motorcycle type recognition device based on ETC portal, and the repetition is no longer repeated. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. While the system described in the embodiments below is preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated.

Fig. 3 shows a flowchart of a precise vehicle type identification method based on the ETC portal according to an embodiment of the present invention. As shown in fig. 3, the risk assessment method includes:

the vehicle type identification method comprises the following steps:

s20, acquiring front image information of the vehicle on the lane through a first image acquisition device, and acquiring side image information of the vehicle through a second image acquisition device;

s40, obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand;

s60, obtaining the number of side wheels according to the side image information;

s80, determining the vehicle type according to the vehicle basic information and the number of the side wheels;

In the embodiment, the front image information of the vehicle on the lane is acquired through the first image acquisition equipment, and the side image information of the vehicle is acquired through the second image acquisition equipment; obtaining vehicle basic information according to the front image information, wherein the vehicle basic information comprises a license plate number, a vehicle body color and a vehicle brand; obtaining the number of side wheels according to the side image information; determining the vehicle type according to the vehicle basic information and the number of the side wheels; the vehicle identification device has the advantages that the traditional license plate identification, the vehicle color identification, the vehicle brand identification, the side vehicle splicing, the axle number detection and other technologies are combined, the vehicle identification device is applied to a high-speed portal for the first time, all-round vehicle information collection can be carried out on a vehicle running at a high speed, the vehicle identification device mainly comprises a vehicle front view, a side view, a license plate number, vehicle color, brand information and axle number information, and effective evidence is provided for subsequent vehicle fee evasion and vehicle inspection.

In some embodiments, the vehicle type identification method further includes matching the time when the first image acquisition device acquires the front image information of the vehicle and the time when the second image acquisition device acquires the side image information of the vehicle, and reporting the identification result in a unified manner after matching is successful.

Taking the first image acquisition device and the second image acquisition device as top-mounted cameras and side-mounted cameras as examples; matching according to the time when the first image acquisition equipment acquires the front image information of the vehicle and the time when the second image acquisition equipment acquires the side image information of the vehicle, wherein after the matching is successful, the unified reporting of the identification result comprises the following steps:

T2–T1＞thr1

T2–T1＜thr2

the time thresholds for which thr1 and thr1 are matched may be 2 and 8, respectively, in seconds.

An embodiment of the present invention also provides a computer electronic device, and fig. 5 shows a schematic structural diagram of an electronic device to which an embodiment of the present invention can be applied, and as shown in fig. 5, the computer electronic device includes a Central Processing Unit (CPU)501 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)502 or a program loaded from a storage section 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for system operation are also stored. The CPU 501, ROM 502, and RAM 503 are connected to each other via a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.

The following components are connected to the I/O interface 505: an input portion 506 including a keyboard, a mouse, and the like; an output portion 507 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 508 including a hard disk and the like; and a communication section 509 including a network interface card such as a LAN card, a modem, or the like. The communication section 509 performs communication processing via a network such as the internet. The driver 510 is also connected to the I/O interface 505 as necessary. A removable medium 511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 510 as necessary, so that a computer program read out therefrom is mounted into the storage section 508 as necessary.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

As another aspect, the present invention further provides a computer-readable storage medium, which may be the computer-readable storage medium included in the ETC portal based precision vehicle type identification apparatus in the above embodiments; or it may be a computer-readable storage medium that exists separately and is not built into the electronic device. The computer readable storage medium stores one or more programs for use by one or more processors in performing a method for ETC portal-based precision vehicle model identification as described herein.

The foregoing description is only exemplary of the preferred embodiments of the invention and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the spirit of the invention. For example, the above features and (but not limited to) features having similar functions disclosed in the present invention are mutually replaced to form the technical solution.

Claims

1. The utility model provides an accurate motorcycle type recognition device based on ETC portal, its characterized in that, motorcycle type recognition device includes:

the side axle number identification module is used for obtaining the number of side wheels according to the side image information;

2. The vehicle type recognition apparatus according to claim 1, characterized in that the vehicle type recognition apparatus further comprises:

3. The vehicle type recognition apparatus according to claim 1, wherein the second image capturing device protrudes from the ETC portal frame and is sequentially installed and fixed to face an adjacent lane at 90 ° to the lane.

4. The vehicle type recognition device according to claim 3, wherein the vehicle recognition module includes:

5. The vehicle type recognition device according to any one of claims 1 to 4, wherein the side axis number recognition module includes:

6. The vehicle type recognition device according to claim 5, wherein the vehicle splicing unit includes:

|R1-R2|+|G1-G2|+|B1-B2|<thr1

scores＝(sum_same/sum_pixel)*100

7. The vehicle type recognition apparatus according to any one of claims 1 to 4, characterized in that the vehicle type recognition apparatus further comprises:

8. The vehicle type recognition device according to claim 7, wherein the fusion matching module includes:

T2–T1＞thr1

T2–T1＜thr2

wherein thr1 and thr2 are matched time thresholds and take values of 2s and 8s respectively;

9. The accurate vehicle type identification method based on the ETC portal is characterized by comprising the following steps:

obtaining the number of side wheels according to the side image information;

10. The vehicle type identification method according to claim 9, further comprising matching the time when the first image acquisition device acquires the front image information of the vehicle and the time when the second image acquisition device acquires the side image information of the vehicle, and reporting the identification results after matching is successful.