A regional parking evaluation method based on parking and video big data

Abstract

The application discloses a regional parking evaluation method based on parking and video big data, belongs to the technical field of regional parking big data, and solves the problem that single-factor fuzzy evaluation only reflects the influence of an evaluation object on a single factor, which is far from enough for the overall evaluation of municipal public traffic and simultaneously leads to the problem of large deviation of the overall evaluation result. The fuzzy comprehensive evaluation set is calculated, all factors are subjected to a fuzzy comprehensive evaluation through the collected big data of regional parking resources and various building parking spaces, and thus a comprehensive evaluation set is obtained, the influence of all factors on the evaluation object is comprehensively considered, the fuzzy comprehensive evaluation set can effectively reflect the evaluation result of public regional parking, and thus a good evaluation standard is achieved, so that external personnel have a clear cognition of the building parking spaces for public regional parking.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of regional parking big data, and particularly relates to a regional parking evaluation method based on parking and video big data. BACKGROUND

[0002] With the development of the times, cars have become a necessary means of transportation in people's lives. For the parking of cars, most communities and building construction parking lots are used to orderly park cars.

[0003] Based on the rapid development of urban public transportation, when evaluating regional parking, the general method is to control the problem analysis variable in a single factor, evaluate this single factor, and finally obtain the element membership degree, which is called single factor fuzzy evaluation. However, the single factor fuzzy evaluation only reflects the influence of the object to be evaluated on a single factor, which is far from enough for the overall evaluation of urban public transportation, and also leads to a large deviation of the overall evaluation result. SUMMARY

[0004] In order to solve the problems existing in the above scheme, the application provides a regional parking evaluation method based on parking and video big data.

[0005] The purpose of the application can be achieved by the following technical scheme: a regional parking evaluation method based on parking and video big data, comprising the following steps:

[0006] S1, comprehensively collecting regional parking resources, including parking resource survey data, various types of building parking spaces and public parking space data of the intelligent parking platform, classifying according to basic parking spaces and travel parking spaces, and digitizing and dynamically updating the parking resources;

[0007] S2, analyzing and judging regional parking demand, comprehensively analyzing the vehicle access data of community gates, vehicle operation data collected by regional electric police checkpoints and parking data collected by the intelligent parking platform, using the K-prototype algorithm to construct a parking demand classification model, judging the parking demand type of the vehicle, and then calculating the basic parking demand and travel parking demand in the region;

[0008] S3, according to the real-time collected basic parking space and travel parking space supply quantity, basic parking demand and travel parking demand data, drawing a real-time parking supply and demand relationship curve of different time periods and different types;

[0009] S4, constructing a regional parking supply and demand evaluation index system, comprehensively evaluating the regional parking supply and demand by analyzing the index weight of the parking supply and demand influence.

[0010] Preferably, the parking resource survey data in step S1, various types of building parking spaces and public parking space data of the intelligent parking platform are provided by the intelligent parking platform, parking lot intelligent parking management system, artificial survey and other ways. According to the provided data, the external operator updates in real time, and the updated time is determined as one month update.

[0011] Preferably, the steps of calculating basic parking demand and travel parking demand in step S2 according to the collected data are as follows:

[0012] S21, the gate access data of residential areas in the region is marked as Zi, the gate data of other types of buildings is marked as Qi, the in-road parking data collected by the intelligent parking platform is marked as Li, and the vehicle operation data collected by the electric police corner in the region is marked as Di, wherein i represents different vehicles, i=1, 2, …, n;

[0013] S22, the parking time, parking location and parking duration information of a single parking vehicle are extracted, and the parking time is divided according to date and time period, which is divided into parking date and parking time period;

[0014] Parking date PD i is divided into: weekdays and weekends, that is: PD i =[0, 1], wherein "0" represents weekdays and "1" represents weekends.

[0015] Parking time period PS i is divided into four time periods, wherein parking time period 1 is from 7:00 am to 11:00 am on the same day, represented by "0"; parking time period 2 is from 11:00 am to 14:00 pm on the same day, represented by "1"; parking time period 3 is from 14:00 pm to 18:00 pm on the same day, represented by "2"; parking time period 4 is from 18:00 pm on the same day to 6:00 am on the next day, represented by "3", so the parking time period is divided into four parts, that is: PS i =[0, 1, 2, 3].

[0016] Parking location PL i is divided into: in-road parking and off-road parking. Among them, in-road parking refers to occupying the parking space marked on the public road, represented by "0". Off-road parking mainly refers to the places specially built for parking vehicles, which can be divided into public parking lots and special parking lots. Among them, public parking lots are represented by "1".

[0017] Parking duration Among them, is the time when the i-th vehicle arrives at the parking space detected by the equipment, is the time when the i-th vehicle leaves the parking space detected by the equipment.

[0018] S23, constructing a parking demand classification model by using a K-prototype algorithm, and the specific steps are as follows:

[0019] Input parking feature data X = {x1, x2,..., x i ,...,x n}, wherein x1 represents the parking feature data of the ith vehicle, x i = [x i1 , x i2 , x i3 , x i4 ]. x i1 is a numerical variable, x i1 = PT i . x i2 , x i3 and x i4 are categorical variables, x i2 = PD i , x i3 = PS i , x i4 = PL i ; the number of clustering clusters k = 2.

[0020] Randomly select k initial center points Q l = [q l1 , q l2 ];

[0021] Calculate the distance d(x i , Q l ) between the sample points and the k center points, wherein the numerical variable calculates the Euclidean distance, and the categorical variable calculates the Hamming distance.

[0022]

[0023] wherein the first p is a numerical attribute, p = 1, and the last m is a classification attribute, m = 3, represents the jth numerical attribute of the ith sample, is the jth numerical variable attribute of the center cluster Q, μ l is the weight factor of the categorical variable attribute, is the calculation method of the categorical variable.

[0024]

[0025] wherein, represents the jth categorical variable attribute of the ith sample, is the jth categorical variable attribute of the center cluster Q

[0026] The sample point is divided into the category corresponding to the center point closest to it;

[0027] The category center is updated, the mean value of the sample value of the numerical variable is taken as the feature value of the new prototype, and the mode of the sample value of the category variable is taken as the feature value of the new prototype;

[0028] The objective function of the i-th cycle is calculated

[0029] Wherein,

[0030] The difference ΔF between the i-th and i-1th objective functions is calculated i,i-1 =F i -F -i1 Whether ΔF i,i-1 ≤ε is satisfied, if yes, the cycle is stopped and the clustering result is output; if no, steps S233-S236 are continued until the iteration stopping condition is satisfied.

[0031] Preferably, the parking supply and demand curve graph drawn in step S3 is automatically drawn by the system according to real-time classified parking supply and demand data, and the internal time row and column interval of the drawn curve graph is 1 hour.

[0032] Preferably, the parking supply and demand evaluation index system in step S4 is constructed by the operator himself, the index calculation rule is set, the index is calculated according to the collected data, the evaluation index system can be expanded according to the demand, the weight of the index is analyzed through the role of the index in the parking supply and demand analysis, and the steps are as follows:

[0033] S41, a regional parking evaluation index system U=[u1, u2,..., u j ] is constructed, including: static parking index and dynamic parking index two categories, wherein, the static parking index includes but is not limited to parking berth ratio, total parking capacity / motor vehicle population, parking supply gap, etc., the dynamic parking index mainly reflects the parking operation level, including but not limited to parking turnover rate, berth saturation rate, berth utilization rate, average parking time, free parking proportion, etc.;

[0034] S42, the calculation method of part of the dynamic index is as follows:

[0035] The parking turnover rate is the ratio of the actual total number of parked vehicles in a specific time in a region to the parking capacity of the parking lot;

[0036] The berth utilization rate is the ratio of the number of utilized berths in a region to the total number of parking berths;

[0037] The average parking time Wherein, n is the total number of vehicles in the region, PT i is the parking time of the i-th vehicle;

[0038] The free parking ratio is the ratio of the number of free parking spaces in the region to the total number of parking spaces in the region.

[0039] Preferably, the step of comprehensively evaluating the parking supply and demand in the public area in step S4 is as follows:

[0040] S401, set the parking service level evaluation set V, and divide the parking service level into four levels, i.e. V = [v1, v2, v3, v4]. The higher the level, the lower the service level.

[0041] S402, set the evaluation matrix R. Let the membership degree of the i-th element in the index set U to the j-th element in the evaluation set V be r ij , then the parking service level evaluation matrix R can be represented as:

[0042]

[0043] S403, determine the evaluation index weight set A, A = [a1, a2,..., an], where a j is the j-th evaluation index weight. j

[0044] S404, use the entropy weight method to determine a j , and the calculation steps are as follows:

[0045] Let u ij be the j-th evaluation index value of the i-th parking evaluation area, where i ∈ [1, m] and j ∈ [1, n], and the evaluation index matrix U is as follows:

[0046]

[0047] S4041, data standardization processing, use the minimum-maximum normalization to process the data, as shown in equation 3:

[0048]

[0049] Where u′ ij is the normalized index value, u ij is the index value before normalization, max{u 1j ,u 2j ,..., u mj} and min{u 1j ,u 2j ,..., u mj} are the maximum and minimum values of the j-th evaluation index, respectively.

[0050] ​S4042, calculate the proportion of the i-th evaluation area under the j-th evaluation index, as shown in formula 4.

[0051]

[0052] S4043, calculate the entropy value E of the j-th index j , as shown in formula 5:

[0053]

[0054] , wherein,

[0055] S4044, calculate the weight of each evaluation index, as shown in formula 6:

[0056]

[0057] S405, construct a regional parking service level evaluation model, multiply the evaluation index weight set A and the evaluation matrix R to obtain a fuzzy comprehensive evaluation matrix B, B=A·R, and the final parking service level is the parking service level corresponding to the maximum value in the fuzzy comprehensive evaluation matrix B.

[0058] Preferably, in step S405, the calculated fuzzy comprehensive evaluation is directly transmitted to an external display terminal, and the fuzzy comprehensive evaluation is displayed for the operator to check and accept.

[0059] Compared with the prior art, the beneficial effects of the present application are: when evaluating regional parking, a plurality of different data sets are collected in advance by using a data collection method, the collected factors are evaluated by using a clustering algorithm to obtain the travel regularity value of the vehicle, and then different index systems are selected for different regions to analyze the index weight of the travel regularity value Ei, and the corresponding three sets are analyzed, and then the corresponding three sets are used to calculate the fuzzy comprehensive evaluation set, and all factors are comprehensively evaluated through the collected big data regional parking resources and various building parking spaces, so as to obtain a comprehensive evaluation set, and all factors are comprehensively considered to affect the evaluation object, so that the fuzzy comprehensive evaluation set can effectively reflect the evaluation result of the public regional parking, so as to achieve a good evaluation standard, and the external personnel can clearly understand the parking space allocation in the public regional parking. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only represent some of the embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0061] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0062] The technical solutions of the present application will be described in detail below in conjunction with the embodiments. Obviously, the described embodiments only represent some of the embodiments of the present application, but not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0063] As Figure 1 shown, a regional parking evaluation method based on parking and video big data, comprising the following steps:

[0064] S1, comprehensively collecting parking resources in the region, including parking resource survey data, various types of building parking spaces and public parking space data of the intelligent parking platform, classifying according to basic parking spaces and travel parking spaces, and digitizing and dynamically updating the parking resources;

[0065] S2, analyzing and judging the regional parking demand, comprehensively analyzing the vehicle access data of community gate, vehicle operation data collected by regional electric police checkpoint and parking data collected by intelligent parking platform, using K-prototype algorithm to construct parking demand classification model, judging the parking demand type of vehicle, and then calculating the basic parking demand and travel parking demand in the region;

[0066] S3, according to the real-time collected basic parking space and travel parking space supply quantity, basic parking demand and travel parking demand data, drawing real-time parking supply and demand relationship curve of different time periods and different types;

[0067] S4, constructing regional parking supply and demand evaluation index system, comprehensively evaluating the regional parking supply and demand by analyzing the index weight of parking supply and demand influence.

[0068] The parking resource survey data, various types of building parking spaces and public parking space data of the intelligent parking platform in step S1 are provided by the intelligent parking platform, parking lot intelligent parking management system, artificial survey and other ways. According to the provided data, external operators update in real time, and the updated time is determined as one month update.

[0069] The steps of calculating the basic parking demand and the trip parking demand according to the collected data in step S2 are as follows:

[0070] S21, marking the gate access data of residential communities in the region as Zi, marking the gate data of other types of buildings as Qi, marking the in-road parking data collected by the intelligent parking platform as Li, and marking the vehicle operation data collected by the electric police corner in the region as Di, wherein i represents different vehicles, i = 1, 2, …, n;

[0071] S22, extracting the parking time, parking location and parking duration information of a single parking vehicle, and dividing the parking time according to date and time period, into parking date and parking time period;

[0072] Parking date PD i is divided into weekdays and weekends, that is, PD i =[0, 1], wherein "0" represents weekdays and "1" represents weekends.

[0073] Parking time period PS i is divided into four time periods, wherein time period 1 is from 7:00 am to 11:00 am on the same day, represented by "0"; time period 2 is from 11:00 am to 14:00 pm on the same day, represented by "1"; time period 3 is from 14:00 pm to 18:00 pm on the same day, represented by "2"; and time period 4 is from 18:00 pm on the same day to 6:00 am on the next day, represented by "3", thus the parking time period is divided into four parts, that is, PS i =[0, 1, 2, 3].

[0074] Parking location PL i is divided into in-road parking and off-road parking, wherein in-road parking refers to occupying a parking space delineated on a public road, represented by "0"; off-road parking mainly refers to places specially built for parking vehicles, which can be divided into public parking lots and special parking lots. Among them, the public parking lot is represented by "1", mainly including traffic parking lot, cultural and sports parking lot, commercial parking lot and freight vehicle parking lot, and the special parking lot is represented by "2", mainly including residential parking lot and working parking lot, thus the parking location can be divided into three parts, that is, PL i =[0, 1, 2].

[0075] Parking duration , wherein is the time when vehicle i arrives at the parking space detected by the equipment, is the time when vehicle i leaves the parking space detected by the equipment.

[0076] S23, a parking demand classification model is constructed by using a K-prototype algorithm, and the specific steps are as follows:

[0077] S231, input parking feature data X={x1, x2,..., x i ,...,x n}, wherein x1 represents the parking feature data of the ith vehicle, x i =[x i1 ,x i2 ,x i3 ,x i4 ]. x i1 is a numerical variable, x i1 =PT i . x i2 , x i3 and x i4 are category variables, x i2 =PD i , x i3 =PS i , x i4 =PL i ; the number of clustering clusters k=2.

[0078] S232, randomly select k initial center points Q l =[q l1 ,q l2 ];

[0079] S233, calculate the distance d(x i ,Q l ) between the sample points and the k center points, wherein the numerical variable calculates the Euclidean distance, and the category variable calculates the Hamming distance.

[0080]

[0081] wherein the first p is a numerical attribute, p=1, and the last m is a classification attribute, m=3, represents the jth numerical attribute of the ith sample, is the jth numerical variable attribute of the center cluster Q, μ l is the weight factor of the category variable attribute, is the calculation method of the category variable.

[0082]

[0083] wherein, represents the jth category variable attribute of the ith sample, is the jth category variable attribute of the center cluster Q

[0084] S234, the sample point is divided into the category corresponding to the center point closest to it;

[0085] S235, the category center is updated, the mean value of the sample value of the numerical variable is taken as the feature value of the new prototype, and the mode of the sample value of the category variable is taken as the feature value of the new prototype;

[0086] S236, the objective function of the ith cycle is calculated

[0087] wherein,

[0088] S237, the difference ΔF of the ith and (i-1)th objective functions is calculated i,i-1 = F i -F i-1 , whether ΔF i,i-1 ≤ε is satisfied, if yes, the cycle is stopped and the clustering result is output; if no, steps S233-S236 are continued until the iteration stopping condition is satisfied.

[0089] The parking supply and demand curve chart drawn in step S3 is automatically drawn by the system according to real-time classified parking supply and demand data. The internal time row and column interval of the drawn curve chart is 1 hour.

[0090] The parking supply and demand evaluation index system in step S4 is constructed by the operator himself, the index calculation rule is set, the index is calculated according to the collected data, the evaluation index system can be expanded according to the demand, the weight of the index is analyzed through the role of the evaluation index in the parking supply and demand analysis, and the steps are as follows:

[0091] S41, a regional parking evaluation index system U=[u1, u2,..., u j ] is constructed, including: static parking index and dynamic parking index two categories, wherein, the static parking index includes but is not limited to parking berth ratio, total parking capacity / motor vehicle population, parking supply gap, etc., the dynamic parking index mainly reflects the parking operation level, including but not limited to parking turnover rate, berth saturation rate, berth utilization rate, average parking time, free parking proportion, etc.;

[0092] S42, the calculation method of part of the dynamic index is as follows:

[0093] The parking turnover rate is the ratio of the actual total number of parked vehicles in a specific time in a region to the parking capacity of the parking lot;

[0094] The berth utilization rate is the ratio of the number of utilized berths in a region to the total number of parking berths;

[0095] The average parking time Wherein, n is the total number of vehicles in the region, PT iParking duration of the ith vehicle;

[0096] The free parking ratio is the ratio of the number of free parking spaces in the area to the total number of parking spaces in the area.

[0097] The step of comprehensively evaluating the parking supply and demand of the public area in step S4 is as follows:

[0098] S401, set the parking service level evaluation set V, and divide the parking service level into four levels, i.e. V = [v1, v2, v3, v4]. The higher the level, the lower the service level.

[0099] S402, set the evaluation matrix R. Let the membership degree of the ith element in the index set U to the jth element in the evaluation set V be r ij The parking service level evaluation matrix R can be represented as:

[0100]

[0101] S403, determine the evaluation index weight set A, A = [a1, a2,..., an], where a j is the jth evaluation index weight. j

[0102] S404, use the entropy weight method to determine a j , and the calculation steps are as follows:

[0103] Let u ij be the jth evaluation index value of the ith parking evaluation area, where i ∈ [1, m] and j ∈ [1, n], and the evaluation index matrix U is as follows:

[0104]

[0105] S4041, data standardization processing, use the minimum-maximum normalization to process the data, as shown in equation 3:

[0106]

[0107] Where u′ ij is the normalized index value, u ij is the index value before normalization, max{u 1j ,u 2j ,..., u mj} and min{u 1j ,u 2j ,..., u mj} are the maximum and minimum values of the jth evaluation index respectively;

[0108] ​S4042, calculate the proportion of the i-th evaluation area under the j-th evaluation index, as shown in formula 4.

[0109]

[0110] S4043, calculate the entropy value E of the j-th index j , as shown in formula 5:

[0111]

[0112] , wherein,

[0113] S4044, calculate the weight of each evaluation index, as shown in formula 6:

[0114]

[0115] S405, construct a regional parking service level evaluation model, multiply the evaluation index weight set A and the evaluation matrix R to obtain a fuzzy comprehensive evaluation matrix B, B=A·R, and the final parking service level is the parking service level corresponding to the maximum value in the fuzzy comprehensive evaluation matrix B.

[0116] In step S405, the calculated fuzzy comprehensive evaluation is directly transmitted to an external display terminal, and the fuzzy comprehensive evaluation is displayed for the operator to check and accept.

[0117] The above formulas are all calculated by removing the dimension and taking the numerical value, the formula is obtained by software simulation of a large amount of data to obtain a formula closest to the actual situation, and the preset parameters and the preset threshold in the formula are set by the person skilled in the art according to the actual situation or obtained by a large amount of data simulation.

[0118] The working principle of the application is: when evaluating regional parking based on parking and video big data, a plurality of different data are collected in advance by data collection, the collected factors are evaluated by clustering algorithm, the travel regularity value of the vehicle is obtained, different index systems are selected for different regions, the index weight analysis of the travel regularity value is performed, the corresponding three sets are analyzed by weight, the fuzzy comprehensive evaluation set calculation is performed through the corresponding three sets, the parking resources in the collected big data region and various building parking spaces are evaluated, and a fuzzy comprehensive evaluation of all factors is performed, so that the comprehensive evaluation set is obtained.

[0119] Finally, it should be noted that the above examples are only used to illustrate the technical method of the application and are not limiting, although the application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the application.

Claims

1.A regional parking evaluation method based on parking and video big data, characterized in that, Comprise the following steps: S1, comprehensive collection of parking resources in the region, including parking resource survey data, various types of building parking spaces and public parking space data of intelligent parking platform, classified according to basic parking spaces and travel parking spaces, and the parking resources are digitized and dynamically updated; S2, regional parking demand analysis and judgment, comprehensive analysis of the vehicle access data of community gate, vehicle operation data collected by regional electric police portal and parking data collected by intelligent parking platform, using K-prototype algorithm to construct parking demand classification model, judging the parking demand type of vehicle, and then calculating the basic parking demand and travel parking demand in the region; The steps of calculating the basic parking demand and travel parking demand in the region are as follows: S21, mark the data of community gate in residential area as Zi, mark the data of gate of other types of buildings as Qi, mark the in-road parking data collected by intelligent parking platform as Li, and mark the vehicle operation data collected by regional electric police portal as Di, wherein i represents different vehicles, i=1, 2, …, n; S22, extract the parking time, parking location and parking duration information of single parking vehicle, divide the parking time according to date and time period, and divide it into parking date and parking time period; Parking date Is divided into: weekdays and weekends, namely: , "0" represents weekdays, "1" represents weekends; Parking period The parking period is divided into four periods, wherein the first parking period is represented by "0"; the second parking period is represented by "1"; the third parking period is represented by "2"; and the fourth parking period is represented by "3". Thus, the parking period is divided into four parts, namely: ; Parking position Divided into: parking in the road, parking outside the road, wherein the parking in the road refers to occupying the parking space delineated by the public road, represented by "0", the parking outside the road refers to the place specially for parking vehicles which needs to be invested and built, the parking outside the road is divided into public parking lot and special parking lot, wherein the public parking lot is represented by "1", the special parking lot is represented by "2", the parking position is divided into three parts, namely: ; Parking duration wherein, is the time at which the i-th vehicle arrives at the parking space as detected by the device, is the time at which the i-th vehicle departs from the parking space as detected by the device; S23, use K-prototype algorithm to construct parking demand classification model, the specific steps are as follows: S231、input parking feature data wherein, representing parking feature data of the i-th vehicle, , is a numerical variable, , , and is a categorical variable, , , number of clustering clusters k = 2; S232, randomly select 2 initial center points ; S233. Calculate the distance between the sample point and the k center points. Among them, Euclidean distance is calculated for numerical variables and Hemingway distance is calculated for categorical variables; ； where the first p are numerical attributes, p = 1, and the last m are categorical attributes, m = 3, represents the jth numerical attribute of the ith sample, is the jth numerical variable attribute of the central cluster Q, is the weight factor of the categorical variable attribute, is the calculation method of the categorical variable. ; wherein, represents the jth categorical attribute of the ith sample, is the jth categorical attribute of the central cluster Q; S234, divide the sample points into the category corresponding to the nearest center point; S235, update the category center, the mean value of the numerical variable sample value as the new prototype feature value, and the mode of the category variable sample value as the new prototype feature value; S236, calculate the objective function of the i-th loop , wherein ; S237, calculate the difference of the ith objective function and the i-1th objective function , judge whether the iteration stopping condition is met, if yes, stop the loop and output the clustering result; if no, continue steps S233-S236 until the iteration stopping condition is met; S3, according to the real-time collected basic parking space and travel parking space supply quantity, basic parking demand and travel parking demand, draw the real-time parking supply and demand relationship curve of different time periods and different types; S4, construct the regional parking supply and demand evaluation index system, analyze the index weight of parking supply and demand influence, and comprehensively evaluate the regional parking supply and demand. 2.The regional parking evaluation method based on parking and video big data according to claim 1, wherein, The parking resource survey data, various types of building parking spaces and public parking space data of intelligent parking platform in step S1 are provided by intelligent parking platform, parking lot intelligent parking management system and artificial survey method. According to the provided data, external operators update in real time, and the updated time is set to one month. 3.The regional parking evaluation method based on parking and video big data according to claim 1, characterized in that, The parking supply and demand curve in step S3 is drawn by the system according to the real-time classification parking supply and demand data. The internal time row and column interval of the drawn curve is 1 hour. 4.The regional parking evaluation method based on parking and video big data according to claim 1, wherein, The parking supply and demand evaluation index system in step S4 is constructed by the operator, the index calculation rule is set, the index is calculated according to the collected data, the evaluation index system can be expanded according to the demand, the weight of the index is analyzed through the role of the evaluation index in the parking supply and demand analysis, and the steps are as follows: S41, constructing a regional parking evaluation index system wherein, denotes the bth evaluation index in the index set U, and N denotes the number of evaluation indexes, including two categories of static parking indexes and dynamic parking indexes, and the dynamic parking indexes reflect the parking operation level; S42, the calculation method of part of dynamic parking index is as follows: Parking turnover rate is the ratio of actual parking total number in a certain time in a region to parking capacity of parking lot; Parking space utilization rate is the ratio of used parking space number to total parking space number; average parking duration where n is the total number of vehicles in the region, is the parking duration of the i-th vehicle; The free parking ratio is the ratio of the number of free parking spaces in the area to the total number of parking spaces in the area. 5.The regional parking evaluation method based on parking and video big data according to claim 4, wherein, The step of comprehensively evaluating the parking supply and demand of the public area in step S4 is as follows: S401, set the parking service level evaluation set V, divide the parking service level into first, second, third and fourth levels, that is , the higher the level, the lower the service level represents; S402、Set the evaluation matrix R, set the membership degree of the bth element in the index set U to the cth element in the evaluation set V as The parking service level evaluation matrix R is represented as: ； S403, determine the evaluation index weight set W, W wherein, is the bth evaluation index weight; S404, determining by using an entropy weight method The calculation steps are as follows: Set the number of parking evaluation areas, M, and the number of evaluation indexes, N, and the evaluation index value of the i-th evaluation index of the j-th parking evaluation area, Cji, as follows. b The evaluation index matrix C is as follows.​​​ ; S4041, data standardization processing, using minimum-maximum normalization to process the data, as shown in formula 3: （3）； wherein, is the normalized index value, is the index value before normalization, and are the maximum and minimum values of the bth evaluation index, respectively. S4042, calculating the proportion of the a-th evaluation area in the b-th evaluation index, as shown in formula 4: （4）； S4043、calculate the entropy value of the bth index As shown in Equation 5: （5）； wherein ; S4044, calculating the weight of each evaluation index, as shown in formula 6: （6）； S405, the regional parking service level evaluation model is constructed, the evaluation index weight set W and the evaluation matrix R are multiplied to obtain a fuzzy comprehensive evaluation matrix D, D The final parking service level is the parking service level grade corresponding to the maximum value in the fuzzy comprehensive evaluation matrix D. 6.The regional parking evaluation method based on parking and video big data according to claim 5, wherein, In step S405, the calculated fuzzy comprehensive evaluation is directly transmitted to an external display terminal, and the fuzzy comprehensive evaluation is displayed for the operator to check and accept.

Images