CN117496607B

CN117496607B - ETC (electronic toll collection) -based intelligent parking lot management method and system

Info

Publication number: CN117496607B
Application number: CN202311472213.XA
Authority: CN
Inventors: 颜国顺; 胡俊华
Original assignee: Wuhan Wireless Flying Science And Technology Co ltd
Current assignee: Wuhan Wireless Flying Science And Technology Co ltd
Priority date: 2023-11-07
Filing date: 2023-11-07
Publication date: 2024-06-07
Anticipated expiration: 2043-11-07
Also published as: CN117496607A

Abstract

The invention discloses an intelligent parking lot management method and system based on ETC charging, wherein the invention combines an image recognition method and an ETC electronic license plate recognition method to jointly recognize license plate numbers of the same vehicle, then judges whether the vehicle is a fake license plate vehicle or not by judging whether the vehicle is the same, and finishes parking management of the vehicle by using different parking lot charging payment methods based on a judgment result; when the electronic license plate is consistent with the license plate identified by the image, fee deduction is carried out by utilizing ETC equipment, and if the electronic license plate is inconsistent with the license plate identified by the image, the real license plate identified by the image is adopted to calculate the parking fee, and a parking payment code is generated, so that an imposter can manually pay the parking fee; therefore, the invention solves the problem that the traditional technology brings economic loss to the real vehicle owner caused by the impossibility of the license plate, improves the reliability of ETC parking charging, and is suitable for large-scale application and popularization in the ETC parking charging field.

Description

ETC (electronic toll collection) -based intelligent parking lot management method and system

Technical Field

The invention belongs to the technical field of parking charging management, and particularly relates to an intelligent parking lot management method and system based on ETC charging.

Background

ETC equipment is widely applied at a high-speed entrance and exit by the convenient characteristic of no-parking charging; the ETC equipment comprises an OBU and an RSU, wherein the OBU is provided with identification information (such as a license plate) of a vehicle, the OBU is generally arranged on a windshield in front of the vehicle, the RSU is arranged beside a toll station, and a loop sensor is arranged under the ground of a lane and is combined with the RSU to realize vehicle fee deduction.

At present, the ETC device has a certain application in a parking lot, although the function of no-parking charging is realized, the ETC parking charging has the problem of vehicle impossibility (namely fake license) and is characterized in that if an OBU of a vehicle A is arranged on the vehicle C and a corresponding user of the vehicle A opens the ETC charging function, the ETC device acquires information of the vehicle A when the vehicle C passes through the ETC device of the parking lot, and thus, when the vehicle C leaves the parking lot, a parking system directly deducts parking cost from an account bound by the ETC device of the vehicle A, thereby bringing economic loss to the user of the vehicle A and being unfavorable for the management of the parking lot; based on this, how to provide an intelligent parking lot management method for ETC charging capable of preventing license plate from being falsified has become a problem to be solved.

Disclosure of Invention

The invention aims to provide an ETC (electronic toll collection) -based intelligent parking management method and system, which are used for solving the problem that the ETC parking charging in the prior art has license plate impossibility and causes economic loss to real vehicle owners.

In order to achieve the above purpose, the present invention adopts the following technical scheme:

In a first aspect, an intelligent parking lot management method based on ETC charging is provided, including:

When a target vehicle enters the recognition range of an ETC parking system of a parking lot, acquiring electronic license plate information and license plate images of the target vehicle;

Performing feature extraction processing on the license plate image to obtain a color feature vector of each pixel point in the license plate image;

According to the color feature vectors of the pixel points, license plate region detection processing is carried out on the license plate image so as to extract at least one candidate license plate region from the license plate image;

for each candidate license plate region in the at least one candidate license plate region, acquiring a color characteristic value of each pixel point in each candidate license plate region, and performing edge detection processing on each candidate license plate region based on the color characteristic value of each pixel point in each candidate license plate region so as to obtain an edge detection image corresponding to each candidate license plate region after the edge detection processing;

Determining a real license plate region of the target vehicle from edge detection images of each candidate license plate region, and carrying out image recognition on the real license plate region to obtain license plate recognition information of the target vehicle;

If the electronic license plate information is inconsistent with the license plate identification information, a first entering license plate library is obtained, wherein each first entering license plate in the first entering license plate library is license plate identification information of each entering vehicle, of which the electronic license plate information is inconsistent with the license plate identification information, when entering a parking lot;

judging whether a first entering license plate matched with license plate identification information of a target vehicle exists in a first entering license plate library;

if yes, generating a parking pair Fei Ma, and visually displaying at an exit gate of the parking lot;

If the electronic license plate information is consistent with the license plate identification information, a second entrance license plate library is obtained, wherein each second entrance license plate in the second entrance license plate library is the electronic license plate information of each entrance vehicle with the consistent electronic license plate information and license plate identification information when entering a parking lot;

Judging whether a second entering license plate matched with the electronic license plate information of the target vehicle exists in the second entering license plate library;

If so, calculating the parking cost based on the electronic license plate information of the target vehicle and a second entrance license plate matched with the electronic license plate information, and deducting the parking cost from an account bound by ETC equipment of the target vehicle.

Based on the above disclosure, when a target vehicle passes through the recognition range of the ETC parking system of the parking lot, electronic license plate information (namely license plate information recorded by OBU equipment on the vehicle) and license plate images (namely license plate images shot on site) of the target vehicle are respectively obtained, then at least one candidate license plate region is extracted from the license plate images, and the real license plate region of the target vehicle is determined from the at least one candidate license plate region by an edge detection method; then, carrying out image recognition on the real license plate region, thereby obtaining the on-site license plate recognition information of the target vehicle; therefore, the invention can carry out parking charging management according to the electronic license plate information read by the system and the license plate information identified on site.

Specifically, whether the electronic license plate information of the target vehicle is consistent with license plate identification information or not is judged; if the license plate information is consistent with the license plate identification information, indicating that the problem of license plate impossibility exists, at the moment, acquiring a second entering license plate library (the license plate library stores the electronic license plate information of each entering vehicle with the consistent electronic license plate information when entering a parking lot), and then judging whether the second entering license plate library has a second entering license plate matched with the electronic license plate information of the target vehicle; if so, the vehicle is indicated to be an off-road vehicle, and at the moment, the parking cost is calculated according to the matched recording time of the second entrance license plate and the reading time of the electronic license plate information, and the cost is deducted from the user bound by ETC equipment of the target vehicle.

Similarly, if the electronic license plate information and the license plate identification information of the target vehicle are inconsistent, the problem that the target vehicle has an imposter license plate is described, and at this time, the license plate identified on site is required to be taken as a real license plate, and a first entering license plate library is acquired (the license plate library stores license plate identification information of each entering vehicle, of which the electronic license plate information and the license plate identification information are inconsistent when entering a parking lot, so that the license plate library is equivalent to the real license plate information describing that the imposter vehicle enters the parking lot); then, judging whether a first entering license plate matched with license plate identification information of the target vehicle exists in a first entering license plate library; if so, indicating that the vehicle is an off-road vehicle, generating a parking payment code at the moment, so that a user of the target vehicle can leave the vehicle after manually scanning the code to pay parking fees.

Through the design, the license plate number of the same vehicle is jointly identified by combining the image identification method and the ETC electronic license plate identification method, whether the vehicle is a fake license plate vehicle or not is obtained by judging whether the license plate number is the same, and based on a judgment result, parking management of the vehicle is completed by using different parking lot charging payment methods; when the electronic license plate is consistent with the license plate identified by the image, fee deduction is carried out by utilizing ETC equipment, and if the electronic license plate is inconsistent with the license plate identified by the image, the real license plate identified by the image is adopted to calculate the parking fee, and a parking payment code is generated, so that an imposter can manually pay the parking fee; therefore, the invention solves the problem that the traditional technology brings economic loss to the real vehicle owner caused by the impossibility of the license plate, improves the reliability of ETC parking charging, and is suitable for large-scale application and popularization in the ETC parking charging field.

In one possible design, the feature extraction processing is performed on the license plate image to obtain a color feature vector of each pixel point in the license plate image, including:

For any pixel point in the license plate image, acquiring a red component, a blue component and a green component of the any pixel point, and calculating a first color component and a second color component of the any pixel point by adopting the following formula (1) and formula (2);

（1）

（2）

In the above-mentioned formula (1), Representing a first color component of said arbitrary pixel point,/>Sequentially representing a red component, a blue component and a green component of any pixel point;

In the above-mentioned formula (2), Representing a second color component of the arbitrary pixel point,/>Representing the saturation of any pixel point,/>Representing the brightness of any pixel point, wherein the saturation and the brightness of the any pixel point are calculated based on the red component, the blue component and the green component of the any pixel point,/>Representing saturation component weights,/>Representing luminance component weights;

Wherein, （3）

（4）

In the formula (3),Represents the saturation threshold, in the above equation (4)/(All represent luminance thresholds;

based on the first color component and the second color component of any pixel point, constructing a color feature vector of any pixel point by adopting the following formula (5);

（5）

In the above-mentioned formula (5), Color feature vector representing any pixel point,/>Representing a transpose operation.

In one possible design, according to the color feature vector of each pixel, license plate region detection processing is performed on the license plate image, so as to extract at least one candidate license plate region from the license plate image, including:

Calculating a first probability value and a second probability value of each pixel point based on the color feature vector of each pixel point, wherein the first probability value of any pixel point is the probability that the color corresponding to the any pixel point belongs to the license plate color, and the second probability value is the probability that the color corresponding to the any pixel point belongs to the noise color;

According to the first probability value and the second probability value of each pixel point in the license plate image, carrying out binarization processing on the license plate image to obtain a binarized image;

And carrying out morphological operation on the binarized image so as to extract at least one connected domain in the binarized image after the morphological operation, and taking the extracted at least one connected domain as the at least one candidate license plate region.

In one possible design, calculating a first probability value for each pixel based on the color feature vector for each pixel includes:

for any pixel point, calculating a first probability value of the any pixel point according to the following formula (6);

（6）

in the above-mentioned formula (6), A first probability value representing the arbitrary pixel point,/>Color feature vector representing any pixel point,/>Mean vector representing the gaussian distribution of the colours of the license plate image,/>Representing a first covariance of the color Gaussian distribution,/>Representing a second covariance of the color Gaussian distribution,/>Representing a transpose operation;

Correspondingly, according to the first probability value and the second probability value of each pixel point in the license plate image, performing binarization processing on the license plate image to obtain a binarized image, and then, including:

Calculating the ratio between a first probability value and a second probability value of each pixel point in the license plate image;

And carrying out pixel resetting processing on each pixel point according to the ratio between the first probability value and the second probability value of each pixel point so as to obtain the binarized image after pixel resetting, wherein if the ratio between the first probability value and the second probability value of any pixel point is greater than or equal to a preset threshold value, the pixel value of any pixel point is set to 1, and if the ratio between the first probability value and the second probability value of any pixel point is less than the preset threshold value, the pixel value of any pixel point is set to 0.

In one possible design, obtaining a color feature value for each pixel in each candidate license plate region includes:

For an ith pixel point in any candidate license plate region, acquiring an RBG value of the ith pixel point in the license plate image, and calculating a color characteristic value of the ith pixel point based on the RGB value of the ith pixel point;

adding 1 to i, and re-acquiring RBG values of the ith pixel point in the license plate image until i is equal to m, so as to obtain color characteristic values of all pixel points in any candidate license plate region, wherein the initial value of i is 1, and m is the total number of pixel points in any candidate license plate region;

Correspondingly, based on the color characteristic value of each pixel point in each candidate license plate region, performing edge detection processing on each candidate license plate region comprises the following steps:

For any candidate license plate region, constructing edge feature vectors of all pixel points in the any candidate license plate region based on color feature values of all pixel points in the any candidate license plate region;

And carrying out edge detection processing on the any one candidate license plate region based on the edge feature vector of each pixel point in the any one candidate license plate region so as to obtain an edge detection image corresponding to the any one candidate license plate region after the edge detection processing.

In one possible design, the color feature values of the ith pixel point include a red feature value, a blue feature value, and a green feature value;

the method for constructing the edge feature vector of each pixel point in any candidate license plate region based on the color feature values of each pixel point in any candidate license plate region comprises the following steps:

For an ith pixel point in any candidate license plate area, acquiring a four-neighbor pixel point of the ith pixel point;

Calculating the difference value between the red characteristic value of the ith pixel point and the red characteristic value of each four-adjacent-domain pixel point, the difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of each four-adjacent-domain pixel point, and the difference value between the green characteristic value of the ith pixel point and the green characteristic value of each four-adjacent-domain pixel point;

Based on the difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of each four-adjacent-domain pixel point, a first edge characteristic value of the ith pixel point is constructed;

Screening out the pixel point of the four adjacent domains corresponding to the maximum difference value from the blue characteristic value of the ith pixel point and the difference value between the blue characteristic values of the pixel points of each four adjacent domains to serve as a target pixel point;

Calculating the blue characteristic value, the red characteristic value and the green characteristic value of the ith pixel point, and the difference value between the blue characteristic value, the red characteristic value and the green characteristic value of the target pixel point, and screening out the minimum difference value from the calculated difference values;

Calculating a second edge characteristic value of the ith pixel point based on a target difference value and the minimum difference value, wherein the target difference value is a difference value between a blue characteristic value of the ith pixel point and a blue characteristic value of the target pixel point;

Calculating a third edge characteristic value of the ith pixel point according to the difference value between the blue characteristic value, the red characteristic value and the green characteristic value of the ith pixel point and the blue characteristic value, the red characteristic value and the green characteristic value of the target pixel point;

and constructing an edge feature vector of the ith pixel point by using the first edge feature value, the second edge feature value and the third edge feature value.

In one possible design, based on the edge feature vector of each pixel point in the any one candidate license plate region, edge detection processing is performed on the any one candidate license plate region, so as to obtain an edge detection image corresponding to the any one candidate license plate region after the edge detection processing, including:

According to the edge feature vector of each pixel point in any one candidate license plate region, carrying out edge detection processing on each pixel point in the any one candidate license plate region according to the following formula (7) so as to obtain a binarization value of each pixel point in the any one candidate license plate region after the edge detection processing;

（7）

in the above-mentioned formula (7), Representing the binarization value of the ith pixel point in any candidate license plate area,Representing the edge feature vector of the ith pixel point in the any candidate license plate region,，/>Sequentially representing a first edge characteristic value, a second edge characteristic value and a third edge characteristic value of the ith pixel point, wherein/>Representing edge detection weight row vector,/>Representing a binarization threshold,/>Represents a transpose operation, and/>；

And determining an edge detection image corresponding to any one candidate license plate region based on the binarization value of each pixel point in the any one candidate license plate region.

In one possible design, the constructing the first edge feature value of the ith pixel point based on the difference between the blue feature value of the ith pixel point and the blue feature value of each of the four neighboring pixel points includes:

Screening out the largest difference value and the smallest difference value from the difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of each pixel point in the four adjacent domains to be used as a first difference value and a second difference value respectively;

Calculating the difference between the absolute value of the first difference and the absolute value of the second difference to obtain the first edge characteristic value;

Correspondingly, calculating the second edge feature value of the ith pixel point based on the target difference value and the minimum difference value includes:

Judging whether the target difference value is equal to the minimum difference value;

if yes, the target difference value is used as the second edge characteristic value, otherwise, the second edge characteristic value is taken as 0.

In one possible design, determining whether a first entering license plate matched with license plate identification information of the target vehicle exists in the first entering license plate library includes:

if not, taking the target vehicle as a new entering vehicle, and acquiring the shooting time of the license plate image;

Taking the shot image of the license plate image as the entering time of the target vehicle, and storing license plate identification information and the entering time of the target vehicle into the first entering license plate library.

In a second aspect, an intelligent parking lot management system based on ETC charging is provided, including:

The license plate information acquisition unit is used for acquiring electronic license plate information and license plate images of the target vehicle when the target vehicle enters the recognition range of the ETC parking system of the parking lot;

The feature extraction unit is used for carrying out feature extraction processing on the license plate image so as to obtain a color feature vector of each pixel point in the license plate image;

The license plate extraction unit is used for carrying out license plate region detection processing on the license plate image according to the color feature vectors of the pixel points so as to extract at least one candidate license plate region from the license plate image;

The license plate extraction unit is further used for acquiring color characteristic values of each pixel point in each candidate license plate region for each candidate license plate region in the at least one candidate license plate region, and performing edge detection processing on each candidate license plate region based on the color characteristic values of each pixel point in each candidate license plate region so as to obtain an edge detection image corresponding to each candidate license plate region after the edge detection processing;

The license plate recognition unit is used for determining a real license plate region of the target vehicle from the edge detection images of each candidate license plate region and carrying out image recognition on the real license plate region so as to obtain license plate recognition information of the target vehicle;

The parking management unit is used for acquiring a first entering license plate library when the electronic license plate information is inconsistent with the license plate identification information, wherein each first entering license plate in the first entering license plate library is license plate identification information of each entering vehicle of which the electronic license plate information is inconsistent with the license plate identification information when entering a parking lot;

The parking management unit is used for judging whether a first entering license plate matched with license plate identification information of the target vehicle exists in the first entering license plate library;

If yes, the parking management unit is used for generating a parking pair Fei Ma and visually displaying at an exit gate of the parking lot;

the parking management unit is used for acquiring a second entering license plate library when the electronic license plate information is consistent with the license plate identification information, wherein each second entering license plate in the second entering license plate library is the electronic license plate information of each entering vehicle with the consistent electronic license plate information and license plate identification information when entering a parking lot;

The parking management unit is used for judging whether a second entering license plate matched with the electronic license plate information of the target vehicle exists in the second entering license plate library;

If so, the parking management unit is further configured to calculate a parking fee based on the electronic license plate information of the target vehicle and a second entering license plate matched with the electronic license plate information, and deduct the parking fee from an account bound by ETC equipment of the target vehicle.

In a third aspect, an intelligent parking lot management device based on ETC charging is provided, taking the device as an electronic device as an example, and the intelligent parking lot management device includes a memory, a processor and a transceiver, which are sequentially connected in communication, where the memory is used to store a computer program, the transceiver is used to send and receive a message, and the processor is used to read the computer program, and execute the intelligent parking lot management method based on ETC charging as in the first aspect or any one of the first aspect.

In a fourth aspect, there is provided a storage medium having instructions stored thereon which, when executed on a computer, perform the ETC billing-based intelligent parking lot management method as set forth in the first aspect or any one of the possible designs of the first aspect.

In a fifth aspect, there is provided a computer program product comprising instructions which, when run on a computer, cause the computer to perform the ETC billing based intelligent parking management method as in the first aspect or any one of the possible designs of the first aspect.

The beneficial effects are that:

(1) The invention combines the image recognition method and the ETC electronic license plate recognition method to jointly recognize the license plate number of the same vehicle, then judges whether the vehicle is a fake license plate vehicle or not by judging whether the vehicle is the fake license plate vehicle, and finishes parking management of the vehicle by using different parking lot charging payment methods based on the judging result; when the electronic license plate is consistent with the license plate identified by the image, fee deduction is carried out by utilizing ETC equipment, and if the electronic license plate is inconsistent with the license plate identified by the image, the real license plate identified by the image is adopted to calculate the parking fee, and a parking payment code is generated, so that an imposter can manually pay the parking fee; therefore, the invention solves the problem that the traditional technology brings economic loss to the real vehicle owner caused by the impossibility of the license plate, improves the reliability of ETC parking charging, and is suitable for large-scale application and popularization in the ETC parking charging field.

Drawings

Fig. 1 is a schematic step flow diagram of an intelligent parking lot management method based on ETC charging according to an embodiment of the present invention;

Fig. 2 is a schematic structural diagram of an intelligent parking lot management system based on ETC charging according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention will be briefly described below with reference to the accompanying drawings and the description of the embodiments or the prior art, and it is obvious that the following description of the structure of the drawings is only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art. It should be noted that the description of these examples is for aiding in understanding the present invention, but is not intended to limit the present invention.

It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments of the present invention.

It should be understood that for the term "and/or" that may appear herein, it is merely one association relationship that describes an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a alone, B alone, and both a and B; for the term "/and" that may appear herein, which is descriptive of another associative object relationship, it means that there may be two relationships, e.g., a/and B, it may be expressed that: a alone, a alone and B alone; in addition, for the character "/" that may appear herein, it is generally indicated that the context associated object is an "or" relationship.

Examples:

Referring to fig. 1, the intelligent parking lot management method based on ETC charging provided in the present embodiment combines an image recognition technology and an ETC license plate recognition technology to perform license plate recognition on the same vehicle, so as to obtain electronic license plate information and license plate recognition information of the vehicle; then, comparing whether the two are the same or not, and adopting different parking payment methods to carry out parking management of the vehicle; if the electronic license plate information is inconsistent with the license plate identification information, indicating that the vehicle has a fake license plate problem, at the moment, calculating parking cost by adopting a real license plate identified by an image, and generating a parking payment code so as to enable a fake user to manually pay the parking cost; if the two are the same, the problem that the license plate of the vehicle is not fraudulent use is solved, and at the moment, ETC equipment can be directly used for deducting fees; therefore, the method avoids the problem that the traditional technology brings economic loss to a real vehicle owner due to the fact that a license plate is used, improves the reliability of ETC parking charging, and is suitable for large-scale application and popularization in the ETC parking charging field; the method may be executed on the control side of the parking lot system, alternatively, the control side may be a personal computer (personal computer, PC), and it is understood that the foregoing execution subject is not limited to the embodiment of the present application, and accordingly, the operation steps of the method may be executed as shown in the following steps S1 to S5.

S1, when a target vehicle enters an identification range of an ETC parking system of a parking lot, acquiring electronic license plate information and license plate images of the target vehicle; in the present embodiment, ETC recognition devices (i.e., RSU and loop sensor) and cameras are provided at the entrance gate and the exit gate, respectively, of the parking lot, for example; thus, when the target vehicle enters the RSU identification range of the exit gate or the entrance gate, the RSU reads the electronic license plate information stored in the OBU equipment on the target vehicle; meanwhile, the controller also controls a camera at the outlet or inlet gate to shoot license plate images of the target vehicle; therefore, the controller can carry out parking charging management of the target vehicle according to the read electronic license plate information of the target vehicle and the photographed license plate image.

In specific application, the license plate image acquired by way of example is an RGB image, and the license plate information in the license plate image is identified by the image identification technology in the embodiment; then, comparing the license plate information identified by the image with the electronic license plate information so as to determine whether the target vehicle is a fake vehicle; finally, according to whether the target vehicle is a fake vehicle, different parking charging payment methods can be selected to carry out parking management of the target vehicle.

Optionally, when image recognition of the license plate image is performed, the real license plate area of the target vehicle in the license plate image needs to be extracted first; the process of extracting the real license plate region may be, but not limited to, the following steps S2 to S5.

S2, carrying out feature extraction processing on the license plate image to obtain a color feature vector of each pixel point in the license plate image; when the method is specifically applied, each pixel point in the license plate image can be subjected to color space conversion, and then color feature vectors of each pixel point are obtained through color components obtained after conversion; wherein, the color feature vector of each pixel point is obtained in the same process, and any pixel point in the license plate image is taken as an example for specific explanation; alternatively, the foregoing process of obtaining the color feature vector of any pixel point may be, but not limited to, as shown in the following steps S21 and S22.

S21, for any pixel point in the license plate image, obtaining a red component, a blue component and a green component of the any pixel point, and calculating a first color component and a second color component of the any pixel point by adopting the following formula (1) and formula (2).

（1）

（2）

In the above-mentioned formula (1),Representing a first color component of said arbitrary pixel point,/>Sequentially representing a red component, a blue component and a green component of any pixel point; in the above formula (2)/(Representing a second color component of the arbitrary pixel point,/>Representing the saturation of any pixel point,/>Representing the brightness of any pixel point, wherein the saturation and the brightness of the any pixel point are calculated based on the red component, the blue component and the green component of the any pixel point,Representing saturation component weights,/>Representing luminance component weights; wherein, when calculating the first color component, if the blue component is larger than the green component, the calculated color component is calculated by subtracting the blue component from the green component by 360 degrees, namely, the calculation formula is: /(I)。

Alternatively, the saturation and brightness of any of the above pixels can be calculated by, for example and without limitation, using the following formulas (8) and (9).

（8）

（9）

The saturation component weight and the luminance component weight of any pixel point can be obtained by, for example, but not limited to, the following formula (3) and formula (4) in sequence.

（3）

（4）

In the formula (3),Represents the saturation threshold, in the above equation (4)/(All represent luminance thresholds; in the present embodiment, the saturation threshold and the luminance threshold may be specifically set according to actual use, and are not specifically limited herein; still further still, the method further comprises,Which may be 0.25 and 0.6 in sequence, i.e. to suppress high luminance by introducing a luminance threshold and to compensate for low luminance.

Whereby the first color component and the second color component of any one of the pixel points can be calculated based on the foregoing formulas (1) - (4), and formulas (8) and (9); then, the color feature vector of any pixel point can be constructed according to the first color component and the second color component; the construction process of the color feature vector is as follows in step S22.

S22, constructing a color feature vector of any pixel point based on the first color component and the second color component of the any pixel point by adopting the following formula (5).

（5）

In the above-mentioned formula (5),Color feature vector representing any pixel point,/>Representing a transpose operation.

Through the steps S21 and S22, the color feature vector of any pixel point can be constructed; and then, the color feature vector of each other pixel point in the license plate image can be obtained by the same method.

After color feature vectors of all pixel points in the vehicle image are obtained, candidate license plate areas can be extracted based on the color feature vectors; alternatively, the process of extracting the candidate vehicle region may be, but is not limited to, as shown in step S3 described below.

S3, carrying out license plate region detection processing on the license plate image according to the color feature vectors of the pixel points so as to extract at least one candidate license plate region from the license plate image; in specific implementation, the probability that the color of each pixel point is the license plate color and the interference color can be calculated according to the color feature vector of each pixel point; then, based on the calculated two probability values, binarizing the pixel points; finally, at least one candidate license plate region can be extracted from the binarized image by using morphological open operation and morphological close operation; the process of extracting the candidate license plate region may be, but is not limited to, steps S31 to S33 described below.

S31, calculating a first probability value and a second probability value of each pixel point based on the color feature vector of each pixel point, wherein the first probability value of any pixel point is the probability that the color corresponding to the any pixel point belongs to the license plate color, and the second probability value is the probability that the color corresponding to the any pixel point belongs to the noise color; when the method is applied specifically, the probability that the colors corresponding to the pixels belong to license plate colors and interference colors is calculated, so that the classification of the pixels in the vehicle image is carried out based on the two probability values; similarly, taking any pixel point in the license plate image as an example, the calculation process of the first probability value and the second probability value is specifically described.

Alternatively, for any pixel, the first probability value of the any pixel may be calculated according to the following formula (6), for example and without limitation.

（6）

In the above-mentioned formula (6),A first probability value representing the arbitrary pixel point,/>Color feature vector representing any pixel point,/>Mean vector representing the gaussian distribution of the colours of the license plate image,/>Representing a first covariance of the color Gaussian distribution,/>Representing a second covariance of the color Gaussian distribution,/>Representing a transpose operation.

In particular implementations, the mean vector and the first covariance of the gaussian distribution of colors of the license plate image may be calculated, for example, but not limited to, in the following manner.

The first step: and obtaining a plurality of sample license plate images with different brightness.

And a second step of: calculating sample color feature vectors of each pixel point in each sample license plate image; in this embodiment, the method for obtaining the sample color feature vector of each pixel point in any sample license plate image can be seen from the foregoing step S21 and step S22, and the principle thereof is not repeated.

And a third step of: based on the color feature vectors of the respective samples, and according to the following formula (10) and formula (11) in sequence, a mean vector and a first covariance of the color Gaussian distribution are calculated.

（10）

（11）

In the above-mentioned formula (10),Representing the q-th sample color feature vector,/>Representing the total number of sample color feature vectors.

In the above-mentioned formula (11),Representing a transpose operation.

Meanwhile, the calculation can be performed, for example, but not limited to, according to the following formula (12)。

（12）/>

The above formula (12),Representing a second mean vector,/>。

And subtracting the first probability value from 1 to obtain the second probability value of any pixel point.

From the above formulas (6), (10) - (12), the first probability value and the second probability value of any pixel point in the license plate image can be calculated; then, the first probability value and the second probability value of each other pixel point in the license plate image can be calculated according to the same principle, and the calculation process is not repeated.

After the first probability value and the second probability value of each pixel point in the license plate image are obtained, binarization processing of the image can be performed based on the calculated probability values; the binarization processing procedure is as follows in step S32.

S32, performing binarization processing on the license plate image according to the first probability value and the second probability value of each pixel point in the license plate image to obtain a binarized image; when the method is applied specifically, the ratio between the first probability value and the second probability value of each pixel point in the license plate image can be calculated first; then, according to the ratio between the first probability value and the second probability value of each pixel point, carrying out pixel resetting processing on each pixel point so as to obtain the binarized image after pixel resetting; if the ratio between the first probability value and the second probability value of any pixel point is greater than or equal to a preset threshold (the preset threshold is a binarization threshold, which can be specifically set according to practical use), the pixel value of any pixel point is set to 1, and if the ratio between the first probability value and the second probability value of any pixel point is less than the preset threshold, the pixel value of any pixel point is set to 0; therefore, through the design, the binarization processing of the license plate image can be realized.

After the binarization processing of the license plate image is completed, a morphological algorithm can be utilized to extract the connected domain in the binarization image, so that the extracted connected domain is used as a candidate license plate region; the extraction process of the candidate license plate region is shown in the following step S33.

S33, carrying out morphological operation on the binarized image to extract at least one connected domain in the binarized image after the morphological operation, and taking the extracted at least one connected domain as the at least one candidate license plate region; in this embodiment, the method may include, but is not limited to, performing an open operation on the binary image and then performing a close operation, so as to extract a plurality of connected domains from the binary image; the structural element used in the example opening operation is a first rectangular structural element, and the size of the structural element is the same as that of a standard license plate; meanwhile, the structural element used in the example closing operation is a second rectangular structural element, the size of the structural element is larger than that of the first rectangular structural element, and the length-width ratio is equal to that of the license plate; in addition, morphological operations are a common technique for extracting connected domains, and the principle thereof is not described in detail.

Therefore, through the steps S31 to S33, at least one candidate license plate region can be extracted from the license plate image, and then, the real license plate region of the target vehicle needs to be extracted from the at least one candidate license plate region, so that license plate identification information of the target vehicle can be obtained based on the real license plate region.

Optionally, in this embodiment, edge detection needs to be performed on each candidate license plate region, and then a real license plate region of the target vehicle is extracted based on the edge detection image; the edge detection process is as follows in step S4.

S4, for each candidate license plate region in the at least one candidate license plate region, acquiring a color characteristic value of each pixel point in each candidate license plate region, and carrying out edge detection processing on each candidate license plate region based on the color characteristic value of each pixel point in each candidate license plate region so as to obtain an edge detection image corresponding to each candidate license plate region after the edge detection processing.

In this embodiment, for example, but not limited to, the color feature value of each pixel point in each candidate license plate region may be calculated according to the RGB value of each pixel point in each candidate license plate region in the license plate image; alternatively, the following description will specifically describe the calculation process of the color feature value by taking any candidate license plate region as an example, which may be, but not limited to, those shown in the following steps S41 and S42.

S41, for an ith pixel point in any candidate license plate region, acquiring an RBG value of the ith pixel point in the license plate image, and calculating a color characteristic value of the ith pixel point based on the RGB value of the ith pixel point; when the method is specifically applied, the R component, the B component and the G component of the ith pixel point are divided by 255 respectively, so that the color characteristic value of the ith pixel point can be obtained, namely the ith pixel point comprises a red characteristic value, a blue characteristic value and a green characteristic value; then, the color characteristic values of the rest pixel points in any one candidate license plate region can be calculated by the same method; wherein the looping process is as shown in step S42 below.

S42, adding 1 to i, and re-acquiring RBG values of the ith pixel point in the license plate image until i is equal to m, so as to obtain color characteristic values of all pixel points in any candidate license plate region, wherein the initial value of i is 1, and m is the total number of pixel points in any candidate license plate region.

After calculating the color feature value of each pixel point in each candidate license plate region based on the step S41 and the step S42, the edge detection processing of the corresponding candidate license plate region can be performed based on the calculated color feature value; in this case, any candidate license plate region is taken as an example, and the edge detection process is specifically described as shown in step S43 and step S44 below.

S43, constructing edge feature vectors of all pixel points in any candidate license plate region based on color feature values of all pixel points in the any candidate license plate region for the any candidate license plate region; in the implementation, for the ith pixel point in any candidate license plate area, a neighborhood pixel point of the ith pixel point can be obtained, and then, according to the color characteristic value of each neighborhood pixel point and the color characteristic value of the ith pixel point, the edge characteristic vector of the ith pixel point is calculated; the calculation process may be, but is not limited to, those shown in steps S43a to S43 h.

S43a, for an ith pixel point in any candidate license plate area, acquiring a pixel point of a four-adjacent-domain pixel point of the ith pixel point; in this embodiment, the color feature value of the pixel point in the four neighboring domains of the ith pixel point is obtained, so that the edge feature vector is constructed by combining the color feature values of the ith pixel point.

The calculation process is as follows in steps S43b to S43 h.

S43b, calculating a difference value between the red characteristic value of the ith pixel point and the red characteristic value of each four-adjacent-domain pixel point, a difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of each four-adjacent-domain pixel point, and a difference value between the green characteristic value of the ith pixel point and the green characteristic value of each four-adjacent-domain pixel point; in this embodiment, assuming that the red feature value of the ith pixel point is R1, and the red feature values of the corresponding pixel points in four adjacent domains are R2, R3, R4 and R5 in sequence, then the difference between R1 and R2, the difference between R1 and R3, the difference between R1 and R4 and the difference between R1 and R5 are calculated; of course, the calculation process of the difference between the blue feature value and the green feature value of the ith pixel point and the color feature value corresponding to each pixel point in the four neighboring domains is the same as that of the foregoing example, and will not be repeated here.

After the color characteristic value of the ith pixel point is finished and the difference value of the color characteristic value of each corresponding pixel point in the four adjacent domains is calculated, the calculated difference value can be used for calculating three edge characteristic values of the ith pixel point; the calculation process of the three edge feature values is shown in the following steps S43c to S43 g.

S43c, constructing a first edge characteristic value of the ith pixel point based on a difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of each pixel point in the four adjacent domains; in this embodiment, the maximum difference value and the minimum difference value may be selected from the differences between the blue characteristic value of the ith pixel point and the blue characteristic value of each pixel point in the four adjacent domains, so as to be the first difference value and the second difference value respectively; and then, calculating the difference between the absolute value of the first difference and the absolute value of the second difference, and taking the calculated result as the first edge characteristic value.

Similarly, the calculation process of the other two edge feature values is as follows in steps S43d to S43 g.

S43d, screening out a pixel point in the four adjacent domains corresponding to the maximum difference value from the blue characteristic value of the ith pixel point and the difference value between the blue characteristic values of the pixel points in each four adjacent domains to serve as a target pixel point; in the present embodiment, the following describes step S43d with an example, assuming that the coordinates of the ith pixel point areThen, the coordinates of the corresponding four neighboring pixel points are respectively: /(I)，/>，/>，/>Among the 4 blue characteristic value differences, the blue characteristic value and the coordinate of the ith pixel point are/>The difference between the blue characteristic values of the four adjacent pixel points is the largest, and then the coordinate is the/>Is used as the target pixel point.

After the target pixel point is obtained, calculating a second edge characteristic value based on the color characteristic values of the target pixel point and the ith pixel point; the calculation process is as follows in step S43e and step S43 f.

S43e, calculating the difference value between the blue characteristic value, the red characteristic value and the green characteristic value of the ith pixel point and the blue characteristic value, the red characteristic value and the green characteristic value of the target pixel point, and screening out the minimum difference value from the calculated difference values; in this embodiment, a difference between the blue feature value of the i-th pixel and the blue feature value of the target pixel, a difference between the red feature value of the i-th pixel and the red feature value of the target pixel, and a difference between the green feature value of the i-th pixel and the green feature value of the target pixel are calculated; then, screening out the minimum difference value from the three difference values; then, a second edge characteristic value can be calculated based on the screened minimum difference value; the process of calculating the second edge feature value may be, but is not limited to, as shown in step S43f below.

S43f, calculating a second edge characteristic value of the ith pixel point based on a target difference value and the minimum difference value, wherein the target difference value is a difference value between a blue characteristic value of the ith pixel point and a blue characteristic value of the target pixel point; in the implementation, judging whether the target difference value is equal to the minimum difference value; and if the two values are equal, taking the target difference value as the second edge characteristic value, otherwise, taking the second edge characteristic value as 0.

After the second edge feature value of the ith pixel point is obtained, the third edge feature value may be calculated as shown in step S43g below.

S43g, calculating a third edge characteristic value of the ith pixel point according to the difference value between the blue characteristic value, the red characteristic value and the green characteristic value of the ith pixel point and the blue characteristic value, the red characteristic value and the green characteristic value of the target pixel point; in the present embodiment, the following equation (13) may be used for example and not limited to the calculation of the third edge feature value.

（13）

In the above-mentioned formula (13),Representing third edge eigenvalue,/>Sequentially representing the difference value between the red characteristic value of the ith pixel point and the red characteristic value of the target pixel point, the difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of the target pixel point, the difference value between the green characteristic value of the ith pixel point and the green characteristic value of the target pixel point,/>Representing a sign function, wherein when a variable in the sign function is greater than 0,/>=1, When the variable in the sign function is less than 0,/>= -1, When the variable in the sign function is equal to 0,/>=0。

Therefore, based on the formula (13), after calculating the third edge characteristic value of the ith pixel point, the two edge characteristic values can be combined to construct an edge characteristic vector of the ith pixel point; wherein the construction process is as follows.

S43g, constructing an edge feature vector of the ith pixel point by using the first edge feature value, the second edge feature value and the third edge feature value; in the present embodiment, the edge feature vector of the ith pixel point is exemplifiedMay be, but is not limited to: /(I)，/>And sequentially representing the first edge characteristic value, the second edge characteristic value and the third edge characteristic value of the ith pixel point.

From the steps S43a to S43g, the edge feature vector of the ith pixel point can be calculated, and then, the edge feature vector of each other pixel point in any one of the candidate license plate regions can be calculated by the same method; finally, edge detection processing can be carried out on any edge license plate region based on the edge feature vector of each pixel point in any candidate license plate region so as to obtain a corresponding edge detection image; wherein the edge detection process is as shown in step S44 below.

S44, performing edge detection processing on the any one candidate license plate region based on the edge feature vector of each pixel point in the any one candidate license plate region so as to obtain an edge detection image corresponding to the any one candidate license plate region after the edge detection processing; in this embodiment, the points belonging to the license plate and the background points in any candidate license plate region are determined, so as to complete edge detection; the specific detection process is shown in the following step S44a and step S44 b.

S44a, performing edge detection processing on each pixel point in any candidate license plate region according to the edge feature vector of each pixel point in any candidate license plate region and the following formula (7), so as to obtain binarization values of each pixel point in any candidate license plate region after the edge detection processing.

（7）

In the above-mentioned formula (7),Representing the binarization value of the ith pixel point in any candidate license plate area,Representing the edge feature vector of the ith pixel point in the any candidate license plate region,，/>Sequentially representing a first edge characteristic value, a second edge characteristic value and a third edge characteristic value of the ith pixel point, wherein/>Represents an edge detection weight row vector (which contains three weight values, of course, which can be preset),/>Representing a binarization threshold (presettable),/>Represents a transpose operation, and/>。

Based on the above formula (7), the binarized value of each pixel is calculated, and the edge detection image can be extracted as shown in step S44 b.

S44b, determining an edge detection image corresponding to any candidate license plate region based on the binarization value of each pixel point in the any candidate license plate region; in this embodiment, the detection of the edge point is essentially performed on any candidate license plate region, that is, if the binarization value of the ith pixel point is 1, it indicates that it is an edge pixel point (represented as white in the image), otherwise, if it is 0, it indicates that it is a background point (represented as black in the image); thus, each edge pixel point is connected into a connected domain, and an edge detection image can be obtained.

The edge detection processing of each candidate license plate region can be completed through the steps S41 to S44 and the substeps thereof, so that the edge detection image corresponding to each candidate license plate region is obtained; then, the real license plate region can be extracted from each edge detection image, and image recognition is carried out, so that license plate recognition information of the target vehicle can be obtained; the actual license plate region extraction and license plate recognition process may be, but is not limited to, as shown in step S5 below.

S5, determining a real license plate region of the target vehicle from edge detection images of each candidate license plate region, and carrying out image recognition on the real license plate region to obtain license plate recognition information of the target vehicle; in specific application, morphological opening and closing operation can be performed on each edge detection image to extract a real license plate region, wherein the closing operation is used for forming a connected region, and the opening operation is used for removing an interference region, namely, a non-license plate region in each edge detection image is in an isolated irregular point or line, so that the interference region can be removed based on the morphological operation; finally, carrying out license plate shape to extract a final license plate region (namely, solving the minimum circumscribed rectangle of the remaining region after the interference region is removed, then calculating the length-width ratio of the minimum circumscribed rectangle, and taking the region corresponding to the minimum circumscribed rectangle closest to the length-width ratio of the standard vehicle as a real license plate region) to obtain the real license plate region; meanwhile, after the real license plate area is obtained, character segmentation can be performed, and character recognition is performed by utilizing the trained neural network, so that license plate recognition information of the target vehicle is obtained; in this embodiment, the character segmentation of the character image and the character recognition using the neural network are common techniques for license plate recognition, and the principle thereof is not described again.

After license plate identification information of the target vehicle is obtained, electronic license plate information read by ETC equipment can be combined to judge whether the target vehicle is an imposter vehicle, and different parking payment methods are used for parking management of the target vehicle based on a judgment result; if the target vehicle is a fake vehicle, parking payment is carried out by using the following mode I; if the target vehicle is a normal vehicle, parking payment is performed in the following manner.

The first way is to generate a parking payment code and make manual payment, and the specific process is as follows:

Step 1: if the electronic license plate information is inconsistent with the license plate identification information, a first entering license plate library is obtained, wherein each first entering license plate in the first entering license plate library is license plate identification information of each entering vehicle, of which the electronic license plate information is inconsistent with the license plate identification information, when entering a parking lot; in this embodiment, when the vehicle enters, a different entering license plate library is set based on whether the vehicle is an imposter vehicle, that is: when a certain vehicle enters a field, the electronic license plate information is inconsistent with the license plate identified by the image, the license plate information is judged to be a fake license plate, at the moment, the vehicle identified by the image is required to be taken as a real license plate, and the real license plate is recorded into a first entering license plate library so that the real license plate identified by the image can be used for calculating parking cost when the vehicle leaves the field later; further, based on the first entering license plate library, the process of parking charging management of the target vehicle is performed, as shown in the following steps 2 and 3.

Step 2: and judging whether a first entering license plate matched with license plate identification information of the target vehicle exists in the first entering license plate library.

Step 3: if yes, generating a parking pair Fei Ma, and visually displaying at an exit gate of the parking lot; in this embodiment, if a first entering license plate consistent with license plate identification information of a target license plate can be matched in the first entering license plate library, it is indicated that the target vehicle is an exiting vehicle, at this time, it is required to obtain a shooting time of the first entering license plate consistent with the license plate identification information (i.e., an entering time of the target vehicle) and a shooting time of a license plate image (i.e., an exiting time of the target vehicle), and calculate a parking duration of the target vehicle based on the two shooting times; then, calculating the parking cost of the target vehicle based on the parking market and the parking charging rule; then, generating a parking payment code based on parking fees and displaying the parking payment code at an exit gate; thus, when the target vehicle is an imposter vehicle, the driver of the target vehicle is required to manually sweep the code to pay the parking fee, and the payment cannot be directly deducted from the account bound by the ETC equipment corresponding to the target vehicle; based on the method, the problem that economic loss is brought to a real vehicle owner due to the fact that a license plate is used in an imposition mode can be avoided; of course, after the user pays the parking fee, the controller sends a rod lifting instruction to the exit gate, and meanwhile, the license plate identification information of the target vehicle is deleted from the first entering license plate library.

In addition, if the first entering license plate library does not exist the first entering license plate matched with the license plate identification information of the target vehicle; at this time, the target vehicle is taken as a new entering vehicle, and the shooting time of the license plate image is acquired; and finally, taking the shot image of the license plate image as the entering time of the target vehicle, and storing license plate identification information and the entering time of the target vehicle into the first entering license plate library to realize the entering record of the target vehicle.

Similarly, when the electronic license plate information is consistent with the license plate identification information, parking management is performed in a second mode, wherein the second mode is ETC payment, and the specific management process is as follows.

Step 4: if the electronic license plate information is consistent with the license plate identification information, a second entrance license plate library is obtained, wherein each second entrance license plate in the second entrance license plate library is the electronic license plate information of each entrance vehicle with the consistent electronic license plate information and license plate identification information when entering a parking lot; in the embodiment, when a vehicle enters, if the electronic license plate information is consistent with the license plate identification information, the entering vehicle is a normal vehicle, and the problem of fake license plates is avoided; at this time, the electronic license plate information can be directly used as a real license plate of the vehicle and recorded into a second entering license plate library so as to calculate parking cost based on the electronic license plate information when the vehicle leaves the scene later; meanwhile, based on the second entrance license plate library, the process of parking charging management of the target vehicle is performed, as shown in the following steps 5 and 6.

Step 5: and judging whether a second entering license plate matched with the electronic license plate information of the target vehicle exists in the second entering license plate library.

Step 6: if so, calculating parking cost based on the electronic license plate information of the target vehicle and a second entrance license plate matched with the electronic license plate information, and deducting the parking cost from an account bound by ETC equipment of the target vehicle; in this embodiment, the principle of step 5 and step 6 is the same, and if it is determined that the target vehicle is an off-road vehicle, the parking duration is obtained based on the reading time of the electronic license plate information and the reading time of the second entrance license plate matched with the electronic license plate information, and the parking cost is calculated based on the reading time; and then, compared with the first mode requiring manual payment, the second mode can directly deduct the cost from an account bound by ETC equipment of the target vehicle, thereby realizing quick departure.

In addition, when the second entering license plate library does not have the second entering license plate matched with the electronic license plate information of the target vehicle, the target vehicle is a new entering vehicle; at this time, the electronic license plate information and the corresponding reading time of the target vehicle are required to be recorded into the second entrance license plate library so as to realize the entrance record.

The intelligent parking lot management method based on ETC charging is described in detail in the steps S1-S5, and the intelligent parking lot management method based on ETC charging is combined with an image recognition technology and an ETC license plate recognition technology to carry out license plate recognition on the same vehicle, so that electronic license plate information and license plate recognition information of the vehicle are obtained; then, comparing whether the two are the same or not, and adopting different parking payment methods to carry out parking management of the vehicle; if the electronic license plate information is inconsistent with the license plate identification information, indicating that the vehicle has a fake license plate problem, at the moment, calculating parking cost by adopting a real license plate identified by an image, and generating a parking payment code so as to enable a fake user to manually pay the parking cost; if the two are the same, the problem that the license plate of the vehicle is not fraudulent use is solved, and at the moment, ETC equipment can be directly used for deducting fees; therefore, the invention solves the problem that the traditional technology brings economic loss to the real vehicle owner caused by the impossibility of the license plate, improves the reliability of ETC parking charging, and is suitable for large-scale application and popularization in the ETC parking charging field.

As shown in fig. 2, a second aspect of the present embodiment provides a hardware system for implementing the intelligent parking lot management method based on ETC charging according to the first aspect of the present embodiment, including:

The license plate information acquisition unit is used for acquiring electronic license plate information and license plate images of the target vehicle when the target vehicle enters the recognition range of the ETC parking system of the parking lot.

And the feature extraction unit is used for carrying out feature extraction processing on the license plate image so as to obtain a color feature vector of each pixel point in the license plate image.

And the license plate extraction unit is used for carrying out license plate region detection processing on the license plate image according to the color feature vectors of the pixel points so as to extract at least one candidate license plate region from the license plate image.

The license plate extraction unit is further configured to obtain, for each candidate license plate region in the at least one candidate license plate region, a color feature value of each pixel point in each candidate license plate region, and perform edge detection processing on each candidate license plate region based on the color feature value of each pixel point in each candidate license plate region, so as to obtain an edge detection image corresponding to each candidate license plate region after the edge detection processing.

The license plate recognition unit is used for determining the real license plate region of the target vehicle from the edge detection images of each candidate license plate region, and carrying out image recognition on the real license plate region to obtain license plate recognition information of the target vehicle.

The parking management unit is used for acquiring a first entering license plate library when the electronic license plate information is inconsistent with the license plate identification information, wherein each first entering license plate in the first entering license plate library is license plate identification information of each entering vehicle of which the electronic license plate information is inconsistent with the license plate identification information when entering a parking lot.

The parking management unit is used for judging whether a first entering license plate matched with license plate identification information of the target vehicle exists in the first entering license plate library.

If so, the parking management unit is used for generating a parking pair Fei Ma and visually displaying at the exit gate of the parking lot.

The parking management unit is used for acquiring a second entering license plate library when the electronic license plate information is consistent with the license plate identification information, wherein each second entering license plate in the second entering license plate library is the electronic license plate information of each entering vehicle with the consistent electronic license plate information and license plate identification information when entering the parking lot.

The parking management unit is used for judging whether a second entering license plate matched with the electronic license plate information of the target vehicle exists in the second entering license plate library.

The working process, working details and technical effects of the device provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

As shown in fig. 3, a third aspect of the present embodiment provides an intelligent parking lot management device based on ETC charging, taking the device as an electronic device as an example, including: the intelligent parking lot management method based on ETC charging comprises a memory, a processor and a transceiver which are connected in sequence in communication, wherein the memory is used for storing a computer program, the transceiver is used for receiving and transmitting messages, and the processor is used for reading the computer program and executing the intelligent parking lot management method based on ETC charging according to the first aspect of the embodiment.

By way of specific example, the Memory may include, but is not limited to, random access Memory (random access Memory, RAM), read Only Memory (ROM), flash Memory (Flash Memory), first-in-first-Out Memory (First Input First Output, FIFO) and/or first-in-last-Out Memory (FIRST IN LAST Out, FILO), and the like; in particular, the processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, or the like. The processor may be implemented in at least one hardware form of DSP (DIGITAL SIGNAL Processing), FPGA (Field-Programmable gate array), PLA (Programmable Logic Array ), and may also include a main processor and a coprocessor, where the main processor is a processor for Processing data in a wake-up state, and is also called CPU (Central Processing Unit ); a coprocessor is a low-power processor for processing data in a standby state.

In some embodiments, the processor may be integrated with a GPU (Graphics Processing Unit, image processor) for rendering and drawing of content required to be displayed by the display screen, e.g., the processor may not be limited to a microprocessor of the STM32F105 family, a reduced instruction set computer (reduced instruction set computer, RISC) microprocessor, an X86 or other architecture processor, or a processor that integrates an embedded neural network processor (neural-network processing units, NPU); the transceiver may be, but is not limited to, a wireless fidelity (WIFI) wireless transceiver, a bluetooth wireless transceiver, a General Packet Radio Service (GPRS) wireless transceiver, a ZigBee wireless transceiver (low power local area network protocol based on the ieee802.15.4 standard), a 3G transceiver, a 4G transceiver, and/or a 5G transceiver, etc. In addition, the device may include, but is not limited to, a power module, a display screen, and other necessary components.

The working process, working details and technical effects of the electronic device provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

A fourth aspect of the present embodiment provides a storage medium storing instructions including the intelligent parking lot management method based on ETC charging according to the first aspect of the present embodiment, that is, the storage medium storing instructions thereon, when the instructions are executed on a computer, the intelligent parking lot management method based on ETC charging according to the first aspect of the present embodiment is executed.

The storage medium refers to a carrier for storing data, and may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash Memory, a flash disk, and/or a Memory Stick (Memory Stick), where the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable devices.

The working process, working details and technical effects of the storage medium provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

A fifth aspect of the present embodiment provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the ETC billing based intelligent parking lot management method of the first aspect of the embodiment, wherein the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable device.

Finally, it should be noted that: the foregoing description is only of the preferred embodiments of the invention and is not intended to limit the scope of the invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent parking lot management method based on ETC charging is characterized by comprising the following steps:

If so, calculating parking cost based on the electronic license plate information of the target vehicle and a second entrance license plate matched with the electronic license plate information, and deducting the parking cost from an account bound by ETC equipment of the target vehicle;

performing feature extraction processing on the license plate image to obtain a color feature vector of each pixel point in the license plate image, wherein the feature extraction processing comprises the following steps:

（1）

（2）

Wherein, （3）

（4）

（5）

In the above-mentioned formula (5), Color feature vector representing any pixel point,/>Representing a transpose operation;

And carrying out license plate region detection processing on the license plate image according to the color feature vector of each pixel point so as to extract at least one candidate license plate region from the license plate image, wherein the method comprises the following steps:

Performing morphological operation on the binarized image to extract at least one connected domain in the binarized image after the morphological operation, so as to take the extracted at least one connected domain as the at least one candidate license plate region;

Based on the color feature vectors of each pixel, a first probability value of each pixel is calculated, including:

（6）

According to the ratio between the first probability value and the second probability value of each pixel point, carrying out pixel resetting treatment on each pixel point to obtain the binarized image after pixel resetting, wherein if the ratio between the first probability value and the second probability value of any pixel point is greater than or equal to a preset threshold value, the pixel value of any pixel point is set to 1, and if the ratio between the first probability value and the second probability value of any pixel point is less than the preset threshold value, the pixel value of any pixel point is set to 0;

the method for obtaining the color characteristic value of each pixel point in each candidate license plate area comprises the following steps:

Performing edge detection processing on the any one candidate license plate region based on the edge feature vector of each pixel point in the any one candidate license plate region so as to obtain an edge detection image corresponding to the any one candidate license plate region after the edge detection processing;

The color feature value of the ith pixel point comprises a red feature value, a blue feature value and a green feature value, wherein the edge feature vector of each pixel point in any candidate license plate region is constructed based on the color feature value of each pixel point in any candidate license plate region, and the method comprises the following steps:

constructing an edge feature vector of the ith pixel point by using the first edge feature value, the second edge feature value and the third edge feature value;

Based on the edge feature vector of each pixel point in any one candidate license plate region, carrying out edge detection processing on the any one candidate license plate region to obtain an edge detection image corresponding to the any one candidate license plate region after the edge detection processing, wherein the edge detection image comprises the following steps:

（7）

Determining an edge detection image corresponding to any candidate license plate region based on the binarization value of each pixel point in the any candidate license plate region;

based on the difference value between the blue characteristic value of the ith pixel point and the blue characteristic value of each pixel point in the four adjacent domains, a first edge characteristic value of the ith pixel point is constructed, and the method comprises the following steps:

2. The method of claim 1, wherein determining whether a first entering license plate matching license plate identification information of the target vehicle exists in the first entering license plate library comprises:

3. An ETC billing-based intelligent parking lot management system, comprising:

If yes, the parking management unit is further used for calculating parking cost based on the electronic license plate information of the target vehicle and a second entering license plate matched with the electronic license plate information, and deducting the parking cost from an account bound by ETC equipment of the target vehicle;

（1）

（2）

Wherein, （3）

（4）

（5）

（6）

（7）