Parking guidance system and method based on image recognition

Abstract

The present invention provides a parking space guidance system and method based on image recognition, and relates to the technical field of parking space guidance. The system of the present invention comprises a data acquisition module for collecting real-time images of parking spaces, the number of vehicles entering a parking lot entrance, and the length of a road from a parking space to an entrance and exit of the parking lot, a parking space evaluation coefficient calculation module for generating a parking space evaluation coefficient according to the above data, and a parking space recommendation module for sorting the parking space evaluation coefficients of each parking space from large to small, and recommending the parking space number and the parking lot entrance corresponding to the parking space evaluation coefficient greater than a preset threshold to a driver. The present invention comprehensively considers the parking convenience of the parking space, the convenience of leaving the parking lot, and the convenience of driving to the parking space through the parking space evaluation coefficient calculation module, so as to achieve the effect of guiding the driver to find the best parking space.

Description

Technical Field

[0001] The present invention relates to the technical field of parking guidance, and in particular to a parking guidance system and method based on image recognition. Background Art

[0002] As the number of cars increases, the problem of parking difficulties arises. Many vehicles encounter the problem of finding parking spaces after entering the garage. The current parking guidance system commonly uses ultrasonic sensors, geomagnetic detectors and other sensors and monitors the parking status of each parking space in the parking lot in real time based on its own working principle, and collects information on the parking space usage status of the parking lot. In recent years, there has also been a method of using cameras to perform single-point detection of parking spaces. The single-point detection method has the advantages of accurate detection data and high real-time performance. Most parking lots in China use this data collection method.

[0003] In the prior art, a parking guidance method and system based on image processing and pattern recognition is disclosed with publication number "CN107067813A", which includes: receiving an image to be detected acquired by a camera, the image to be detected including: a marker pole and a plurality of parking spaces, the marker pole being arranged at four vertices or two points of a diagonal line of a rectangular area corresponding to the plurality of parking spaces; identifying the marker pole in the parking lot image, and determining the position of the marker pole in the parking space image according to the coordinates of the marker pole; identifying vehicles within the range of the marker pole, and determining the number of vehicles; comparing the number of vehicles with the number of parking spaces within the range, if the number of vehicles is equal to the number of parking spaces, determining that there are no empty parking spaces, if the number of vehicles is less than the number of parking spaces, determining the position of the empty parking spaces according to the coordinates of the vehicles, and realizing parking guidance based on image processing and pattern recognition by adopting the existing surveillance camera in combination with the marker pole, thereby reducing the cost of the system and facilitating later maintenance, debugging and upgrading.

[0004] However, the existing technology still has major deficiencies, such as: the existing technology only analyzes and determines whether there are empty parking spaces, but does not analyze the quality of the empty parking spaces. For example, empty parking spaces near the parking lot exit are more convenient for drivers to leave than empty parking spaces in the middle of the parking lot. Another example is that empty parking spaces near the parking lot entrance are more convenient for drivers to drive to the parking spaces than empty parking spaces in the middle of the parking lot. Another example is that multiple consecutive adjacent empty parking spaces are more convenient for drivers to park than a single empty parking space. Another example is that empty parking spaces in areas with less traffic are more convenient for drivers to drive to the parking spaces than empty parking spaces in areas with more traffic, etc. That is, the existing technology only roughly analyzes and determines whether there are empty parking spaces, but cannot help drivers find the best parking space to park, and there are certain defects and deficiencies. Summary of the invention

[0005] The object of the present invention is to provide a parking guidance system and method based on image recognition to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A parking guidance system based on image recognition, comprising:

[0008] A data acquisition module, the data acquisition module is used to collect real-time images of parking spaces, the number of vehicles entering the parking lot entrance, and the length of the road from the parking space to the parking lot entrance and exit;

[0009] A data processing module, the data processing module is used to process the real-time image at the parking space to obtain information on whether the parking space is an empty space, and to use the ratio of the number of vehicles entering the previous time period and the time period length as the vehicle flow per unit time;

[0010] A parking convenience coefficient calculation module, the parking convenience coefficient calculation module is used to generate a parking convenience coefficient according to information on whether a parking space is an empty space;

[0011] A departure convenience coefficient calculation module, the departure convenience coefficient calculation module is used to generate a departure convenience coefficient according to information on whether the parking space is an empty space and the length of the road from the parking space to the parking lot exit;

[0012] A driving convenience coefficient calculation module, the driving convenience coefficient calculation module is used to generate a driving convenience coefficient according to information on whether the parking space is an empty space, the length of the road from the parking space to the entrance of the parking lot, and the traffic volume per unit time;

[0013] A parking space evaluation coefficient calculation module, the parking space evaluation coefficient calculation module is used to generate a parking space evaluation coefficient according to a parking convenience coefficient, a driving convenience coefficient, and a driving convenience coefficient;

[0014] The parking space recommendation module is used to sort the parking space evaluation coefficients of various parking spaces from large to small, and recommend the parking space numbers and parking lot entrances corresponding to the parking space evaluation coefficients greater than a preset threshold to the driver.

[0015] Furthermore, the road length from the i-th parking space to the j-th parking lot entrance is calibrated as The length of the road from the i-th parking space to the k-th parking lot exit is calibrated as The vehicle flow per unit time at the entrance of the jth parking lot is calibrated as CL j , the information of whether the i-th parking space is an empty parking space is marked as TCW i , and when the i-th parking space is empty, let TCW i = 1, and when the i-th parking space is not an empty space, let TCW i= 0, calibrate the parking convenience coefficient of the i-th parking space as TCbl i , calibrate the departure convenience coefficient of the i-th parking space as SLbl i , calibrate the driving convenience coefficient corresponding to the i-th parking space as Wherein:

[0016] i represents the numbers of different parking spaces, and i = 1, 2, 3, ……, m, and m ∈ N + , j represents the numbers of different entrances in the parking lot, and j = 1, 2, 3, ……, X, and X ∈ N + , k represents the numbers of different exits in the parking lot, and k = 1, 2, 3, ……, Y, and Y ∈ N + .

[0017] Furthermore, the calculation formula of the parking convenience coefficient is as follows:

[0018] TCbl i = TCW i * f(i)

[0019] Wherein, f(i) is the correction coefficient corresponding to the i-th parking space, and 0 < f(i) ≤ 1, and the more empty spaces there are in the parking spaces adjacent to the i-th parking space, the larger f(i) is.

[0020] Furthermore, when there are no empty spaces in the parking spaces adjacent to the i-th parking space, let f(i) = 0.3; when there is one empty space in the parking spaces adjacent to the i-th parking space, let f(i) = 0.6; when there are two empty spaces in the parking spaces adjacent to the i-th parking space, let f(i) = 1.

[0021] Furthermore, the departure convenience coefficient calculation module includes a first assignment unit and a first calculation unit, wherein:

[0022] The first assignment unit is used to generate a departure coefficient according to the road length from the parking space to the exit of the parking lot, assign a value to the departure coefficient, and generate a departure influence value between 0 and 1 and decreasing as the departure coefficient increases. The calculation formula of the departure coefficient is as follows:

[0023]

[0024] Wherein, d2 max represents the maximum road length from the parking space to the exit of the parking lot, d2 min represents the minimum road length from the parking space to the exit of the parking lot, represents the road length from the i-th parking space to the nearest exit of the parking lot, SL i represents the departure coefficient corresponding to the i-th parking space;

[0025] The first calculation unit is used to generate a departure convenience coefficient according to information about whether the parking space is an empty parking space and the departure impact value, and the calculation formula is as follows:

[0026] SLb i =TCW i *f(SL i )

[0027] Among them, f(SL i ) represents the departure impact value corresponding to the i-th parking space.

[0028] Furthermore, the departure impact value corresponding to the i-th parking space is calibrated as f(SL i ), the departure influence value f(SL i ) is as follows:

[0029] When SL is met i = 0, the departure impact value f(SL i )=1;

[0030] When 0 is satisfied <SL i ≤0.3, the driving-off impact value f(SL i )=0.7;

[0031] When 0.3 is satisfied <SL i ≤0.6, the driving-off impact value f(SL i )=0.5;

[0032] When 0.6 is satisfied <SL i <1, the departure impact value f(SL i )=0.3;

[0033] When SL is met i =1, the departure impact value f(SL i )=0.1.

[0034] Furthermore, the driving convenience coefficient calculation module includes a second assignment unit and a second calculation unit, wherein:

[0035] The second assignment unit is used to generate a driving path coefficient according to the road length from the parking space to the parking lot entrance, and assign a value to the driving path coefficient to generate a first driving impact value between 0 and 1 that decreases as the driving coefficient increases, and assign a value to the vehicle flow per unit time to generate a second driving impact value between 0 and 1 that decreases as the vehicle flow per unit time increases, wherein the calculation formula of the driving path coefficient is as follows:

[0036]

[0037] Among them, d1 max Indicates the maximum length of the road from the parking space to the parking lot entrance, d1 min Indicates the minimum length of the road from the parking space to the parking lot entrance. represents the driving path coefficient corresponding to the jth parking lot entrance to the i-th parking space;

[0038] The second calculation unit is used to generate a driving convenience coefficient according to the first driving impact value, the second driving impact value, and information about whether the parking space is an empty space, and the calculation formula is as follows:

[0039]

[0040] in, represents the driving coefficient from the jth parking lot entrance to the i-th parking space, represents the first driving impact value from the jth parking lot entrance to the ith parking space, f(CL j ) represents the vehicle flow per unit time CL at the entrance of the jth parking lot j The corresponding second driving influence value, λ1, λ2 are both preset proportional coefficients, and λ1+λ2=1, and 0<λ2<λ1<1, the driving convenience coefficient It is the maximum value of the driving coefficients corresponding to the Y parking lot entrances to the i-th parking space.

[0041] Furthermore, the first driving impact value The assignment criteria are as follows:

[0042] When JS is met i j =0, the first driving impact value

[0043] When 0 is satisfied <JS i j ≤0.3, the first driving impact value

[0044] When 0.3 is satisfied <JS i j ≤0.6, the first driving impact value

[0045] When 0.6 is satisfied <JS i j <1, the first driving impact value

[0046] When JS is met i j =1, the first driving impact value

[0047] The assignment standard of the second driving influence value f(CL j ) is as follows:

[0048] When CJ j = 0, the second driving influence value f(CL j ) = 1;

[0049] When 0 < CJ j ≤ 5, the second driving influence value f(CL j ) = 0.8;

[0050] When 5 < CJj ≤ 12, the second driving influence value f(CL j ) = 0.5;

[0051] When 12 < CJ j < 20, the second driving influence value f(CL j ) = 0.2;

[0052] When CJ j > 20, the second driving influence value f(CL j ) = 0.1.

[0053] Furthermore, the calculation formula of the parking space evaluation coefficient is as follows:

[0054]

[0055] where CWpj i represents the parking space evaluation coefficient of the i-th parking space, τ1, τ2, and τ3 are all preset proportionality coefficients, and τ1 + τ2 + τ3 = 1, and 0 < τ2 < τ3 < τ1 < 1.

[0056] A parking space guiding method based on image recognition, which is used for the above-mentioned parking space guiding system based on image recognition, includes the following steps:

[0057] S1, Collect the real-time image at the parking space, the number of vehicles entering at the entrance of the parking lot, and obtain the road length from the parking space to the entrance and exit of the parking lot;

[0058] S2, Process the real-time image at the parking space to obtain information on whether the parking space is empty, and use the ratio of the number of vehicles entering in the previous time period to the time period duration when the real-time image is collected as the traffic flow per unit time;

[0059] S3, Generate a parking convenience coefficient according to the information on whether the parking space is empty;

[0060] S4, Generate a departure convenience coefficient according to the information on whether the parking space is empty and the road length from the parking space to the exit of the parking lot;

[0061] S5, generating a driving convenience coefficient according to information on whether the parking space is an empty space, the length of the road from the parking space to the entrance of the parking lot, and the traffic volume per unit time;

[0062] S6, generating a parking space evaluation coefficient according to the parking convenience coefficient, the exit convenience coefficient, and the driving convenience coefficient;

[0063] S7, sorting the parking space evaluation coefficients of the parking spaces from large to small, and recommending the parking space numbers and parking lot entrances corresponding to the parking space evaluation coefficients greater than a preset threshold to the driver.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] The parking guidance system and method based on image recognition of the present invention generates a parking convenience coefficient, a driving convenience coefficient, and a driving convenience coefficient by analyzing the real-time image of the parking space, the number of vehicles entering the parking lot entrance, and the length of the road from the parking space to the parking lot entrance and exit, and then combines the parking convenience coefficient, the driving convenience coefficient, and the driving convenience coefficient to generate a parking space evaluation coefficient, thereby comprehensively considering the parking convenience of the parking space, the convenience of leaving the parking lot, and the convenience of driving to the parking space, and recommending the parking space number and the parking lot entrance corresponding to the parking space evaluation coefficient to the driver in a descending order, thereby achieving the effect of guiding the driver to find the best parking space. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a module diagram of the parking guidance system based on image recognition in the present invention;

[0067] Figure 2 It is a unit diagram of the departure convenience coefficient calculation module in the present invention;

[0068] Figure 3 It is a unit diagram of the driving convenience coefficient calculation module in the present invention;

[0069] Figure 4 It is a flow chart of the parking guidance method based on image recognition in the present invention. DETAILED DESCRIPTION

[0070] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0071] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0072] Embodiment 1:

[0073] See also Figure 1-3 The present invention provides a parking guidance system based on image recognition, comprising:

[0074] A data acquisition module is used to collect real-time images of parking spaces, the number of vehicles entering the parking lot entrance, and the length of the road from the parking space to the parking lot entrance and exit;

[0075] Among them, the road length from the i-th parking space to the j-th parking lot entrance is calibrated as The length of the road from the i-th parking space to the k-th parking lot exit is calibrated as i represents the number of different parking spaces, and i = 1, 2, 3, ..., m, and m represents the total number of parking spaces in the parking lot, and m∈N + , j represents the number of different entrances in the parking lot, and j = 1, 2, 3, ..., X, and X represents the total number of entrances in the parking lot, and X∈N + , k represents the number of different exits in the parking lot, and k = 1, 2, 3, ..., Y, and Y represents the total number of entrances in the parking lot, and Y∈N + ;

[0076] It should be noted that the data acquisition module is composed of multiple camera devices installed at the parking space and the parking lot entrance. The camera device can be an existing surveillance camera. The camera device installed at the parking space takes a picture of the parking space to obtain a real-time image of the parking space. The parking space can be photographed intermittently. The photographing time interval can be 1 minute, 3 minutes, 5 minutes, 10 minutes, etc., and there is no limitation here. The camera device installed at the parking lot entrance continuously photographs the parking lot entrance to obtain the number of vehicles entering. The specific number of vehicles entering can be obtained by counting software, which will not be elaborated here.

[0077] It should be noted that the road length between the parking space and the entrance / exit of the parking lot can be obtained through a distance measurement software (such as a map navigation software, etc.), which will not be elaborated here.

[0078] A data processing module, which is wirelessly connected to the data acquisition module, is used to process the real-time image at the parking space to obtain information on whether the parking space is empty, and to use the ratio of the number of vehicles entering in the previous time period to the time period duration during real-time image acquisition as the vehicle flow per unit time;

[0079] Among them, the vehicle flow per unit time at the j-th parking lot entrance is calibrated as CL j , and the information on whether the i-th parking space is empty is calibrated as TCW i , and when the parking space is empty, let TCW i = 1, and when the parking space is not empty, let TCW i = 0;

[0080] It should be noted that the duration of the previous time period during real-time image acquisition can be 15 minutes, 30 minutes, 60 minutes, etc., which is not restricted here. The unit of the vehicle flow per unit time can be the number of vehicles entering per hour;

[0081] It should be noted that to determine whether the parking space is empty by processing the image at the parking space, existing image recognition technologies can be used. For example, a large number of parking space images (parking space images include non-empty parking space images with vehicles parked above and empty parking space images without vehicles parked above) are first imported into a neural network learning model for training to generate an empty parking space judgment model, and then the image at the parking space collected by the data acquisition module is imported into the empty parking space judgment model for analysis and judgment to determine whether the parking space is empty.

[0082] A parking convenience coefficient calculation module, which is electrically connected to the data processing module, is used to generate a parking convenience coefficient according to the information on whether the parking space is empty. The calculation formula is as follows:

[0083] TCbl i = TCW i * f(i)

[0084] Among them, TCbl i represents the parking convenience coefficient of the i-th parking space, f(i) is the correction coefficient corresponding to the i-th parking space, and 0 < f(i) ≤ 1. The more empty parking spaces there are among the parking spaces adjacent to the i-th parking space, the larger f(i) is;

[0085] As an implementation mode, when there is no empty parking space in the parking space adjacent to the i-th parking space, let f(i)=0.3, when there is one empty parking space in the parking space adjacent to the i-th parking space, let f(i)=0.6, when there are two empty parking spaces in the parking space adjacent to the i-th parking space, let f(i)=1;

[0086] It should be noted that when determining the correction coefficient corresponding to the i-th parking space, first determine the parking space number adjacent to the i-th parking space according to the actual location of the parking space in the parking lot, and then search for information on whether the parking space is an empty parking space according to the number, so as to know the number of empty parking spaces in the parking spaces adjacent to the i-th parking space;

[0087] It should be noted that TCbl i represents the parking convenience coefficient of the i-th parking space, which is determined by the information TCW of whether the i-th parking space is an empty space. i , and the correction coefficient f(i). When the i-th parking space is empty, the more empty parking spaces adjacent to the parking space, the more parking spaces available for drivers to choose, and the more convenient it is for drivers to park. The parking convenience coefficient TCbl i The larger the parking space is, the greater the parking convenience coefficient TCbl is. i All are 0 to avoid recommending non-empty parking spaces to drivers in the future.

[0088] The departure convenience coefficient calculation module is electrically connected to the data acquisition module and the data processing module, and is used to generate the departure convenience coefficient according to the information of whether the parking space is an empty parking space and the road length from the parking space to the parking lot exit. The departure convenience coefficient calculation module includes a first assignment unit and a first calculation unit, wherein:

[0089] The first assignment unit is electrically connected to the data acquisition module, and is used to generate a departure coefficient according to the length of the road from the parking space to the parking lot exit, and assign a value to the departure coefficient to generate a departure influence value between 0 and 1 that decreases as the departure coefficient increases, wherein the calculation formula of the departure coefficient is as follows:

[0090]

[0091] Among them, d2 max Indicates the maximum length of the road from the parking space to the parking lot exit, d2 min Indicates the minimum length of the road from the parking space to the parking lot exit. represents the length of the road from the i-th parking space to the nearest parking lot exit, d2 max d2 min ,as well as All of them can be obtained by using existing data processing software, which will not be described in detail here;

[0092] Among them, SL i represents the departure coefficient corresponding to the i-th parking space, and the road length from the i-th parking space to the nearest parking lot exit The larger the value, the longer the path the i-th parking space has to travel when leaving the parking lot, that is, the more inconvenient it is for the driver to leave the parking lot. i The bigger it is;

[0093] As an implementation method, the departure impact value corresponding to the i-th parking space is calibrated as f(SL i ), driving away influence value f(SL i ) is as follows:

[0094] <![CDATA[Departure coefficient SL i > <![CDATA[Driving-away influence value f(SL i )]]> <![CDATA[SL i =0]]> 1 <![CDATA[0<SL i ≤0.3]]> 0.7 <![CDATA[0.3<SL i ≤0.6]]> 0.5 <![CDATA[0.6<SL i <1]]> 0.3 <![CDATA[SL i =1]]> 0.1

[0095] It should be noted that the road length from the i-th parking space to the nearest parking lot exit is The larger the value, the longer the path the i-th parking space has to travel when leaving the parking lot, that is, the more inconvenient it is for the driver to leave the parking lot. i The larger the value, the corresponding departure influence value f(SL i ) is smaller;

[0096] The first calculation unit is electrically connected to the first assignment unit and the data processing module, and is used to generate a departure convenience coefficient according to information about whether the parking space is an empty space and the departure impact value. The calculation formula is as follows:

[0097] SLb i =TCW i *f(SL i )

[0098] Among them, SLbl i represents the exit convenience coefficient of the i-th parking space, which is determined by the information TCW of whether the i-th parking space is an empty space. i , and the departure impact value f(SL i ), when the i-th parking space is empty, the driving-away impact value f(SL i ) is larger, it means that it is more convenient for drivers to leave the parking lot, and the departure convenience coefficient SLbl i The larger the value, the greater the impact value f(SL i ) value, departure convenience coefficient SLbl i All are 0 to avoid recommending non-empty parking spaces to drivers in the future.

[0099] A driving convenience coefficient calculation module, which is electrically connected to the data acquisition module and the data processing module, is used to generate a driving convenience coefficient according to information on whether the parking space is an empty space, the length of the road from the parking space to the entrance of the parking lot, and the traffic volume per unit time. The driving convenience coefficient calculation module includes a second assignment unit and a second calculation unit, wherein:

[0100] The second assignment unit is electrically connected to the data acquisition module and the data processing module, and is used to generate a driving path coefficient according to the road length from the parking space to the parking lot entrance, and assign a value to the driving path coefficient to generate a first driving impact value between 0 and 1 that decreases as the driving coefficient increases, and assign a value to the vehicle flow per unit time to generate a second driving impact value between 0 and 1 that decreases as the vehicle flow per unit time increases, wherein the calculation formula of the driving path coefficient is as follows:

[0101]

[0102] Among them, d1 max Indicates the maximum length of the road from the parking space to the parking lot entrance, d1 min Indicates the minimum length of the road from the parking space to the parking lot entrance, d1 max and d1 min All of them can be obtained by using existing data processing software, which will not be described in detail here;

[0103] in, represents the driving path coefficient from the jth parking lot entrance to the i-th parking space, and the road length from the i-th parking space to the jth parking lot entrance The larger the driving path coefficient is, the longer the path from the entrance of the jth parking lot to the ith parking space is, that is, the more inconvenient it is for the driver to drive to the parking space. The bigger it is;

[0104] As an implementation method, the first driving impact value corresponding to the j-th parking lot entrance to the i-th parking space is calibrated as First driving impact value The assignment criteria are as follows:

[0105]

[0106] It should be noted that the length of the road from the i-th parking space to the j-th parking lot entrance is The larger the driving path coefficient is, the longer the path from the entrance of the jth parking lot to the ith parking space is, that is, the more inconvenient it is for the driver to drive to the parking space. The larger the value, the corresponding first driving impact value The smaller it is;

[0107] As an implementation method, the vehicle flow per unit time CL at the entrance of the j-th parking lot is j The corresponding second driving influence value is calibrated as f(CL j ), the second driving influence value f(CL j ) is as follows:

[0108] <![CDATA[Traffic flow per unit time CL j > <![CDATA[Second driving influence value f(CL j )]]> <![CDATA[CJ j =0]]> 1 <![CDATA[0<CJ j ≤5]]> 0.8 <![CDATA[5<CJ j ≤12]]> 0.5 <![CDATA[12<CJ j <20]]> 0.2 <![CDATA[CJ j >20]]> 0.1

[0109] It should be noted that the vehicle flow per unit time CL at the entrance of the jth parking lot is j The larger the value, the more difficult it is to drive from the parking lot entrance to the parking space, that is, the more inconvenient it is for the driver to drive to the parking space. The corresponding second driving influence value f(CL j ) is smaller;

[0110] The second calculation unit is electrically connected to the second value assignment unit and the data processing module, and is used to generate a driving convenience coefficient according to the first driving impact value, the second driving impact value, and information about whether the parking space is an empty space. The calculation formula is as follows:

[0111]

[0112] in, It represents the driving coefficient corresponding to the jth parking lot entrance to the ith parking space. λ1 and λ2 are both preset proportional coefficients. λ1 represents the first driving influence value. The driving coefficient The weight in the calculation, λ2 represents the second driving influence value f(CL j ) in the driving coefficient The weights in the calculation, and λ1+λ2=1, and 0<λ2<λ1<1, the specific values ​​of λ1, λ2, and λ3 are generally determined by those skilled in the art according to actual conditions;

[0113] As an implementation mode, the value range of λ1 is 0.6-0.7, and the value range of λ2 is 0.3-0.4;

[0114] It should be noted that the driving coefficient The first driving impact value The second driving influence value f(CL j ), and whether the parking space is vacant TCW i When the i-th parking space is empty, the first driving impact value The second driving influence value f(CL j ) is larger, it means that it is more convenient to drive from the jth parking lot entrance to the ith parking space, and the driving coefficient The larger the first driving impact value is, the larger the first driving impact value is. The second driving influence value f(CL j ) value, driving coefficient All are 0, to avoid recommending non-empty parking spaces to drivers in the future;

[0115] in, represents the driving convenience coefficient corresponding to the i-th parking space, which is the maximum value of the driving coefficients corresponding to the Y parking lot entrances to the i-th parking space. Existing data processing software can be used for acquisition, which will not be described in detail here.

[0116] The parking space evaluation coefficient calculation module is electrically connected to the parking convenience coefficient calculation module, the driving convenience coefficient calculation module, and the driving convenience coefficient calculation module, and is used to generate a parking space evaluation coefficient according to the parking convenience coefficient, the driving convenience coefficient, and the driving convenience coefficient. The calculation formula is as follows:

[0117]

[0118] Among them, CWpj i represents the parking space evaluation coefficient of the i-th parking space, τ1, τ2, τ3 are all preset proportional coefficients, τ1 represents the parking convenience coefficient TCbl i Evaluation coefficient of parking space CWpj i The weight in the calculation, τ2 represents the departure convenience coefficient SLbl i Evaluation coefficient of parking space CWpj i The weight in the calculation, τ3 represents the driving convenience coefficient Evaluation coefficient of parking space CWpj i The weights in the calculation, and τ1+τ2+τ3=1, and 0<τ2<τ3<τ1<1, the specific values ​​of τ1, τ2, τ3 are generally determined by those skilled in the art according to actual conditions;

[0119] As an implementation mode, the value range of τ1 is 0.5-0.6, the value range of τ2 is 0.1-0.2, and the value range of τ3 is 0.2-0.4;

[0120] It should be noted that the parking space evaluation coefficient CWpj i By parking convenience coefficient TCbl i , departure convenience coefficient SLbl i , and driving convenience factor The parking convenience of the parking space, the convenience of leaving the parking lot and the convenience of driving to the parking space are considered together, and the parking convenience coefficient TCbl i, departure convenience coefficient SLbl i , and driving convenience factor The larger the value, the more convenient it is for the driver to drive to the parking space, park in the parking space, and leave the parking space. i The bigger it is.

[0121] A parking space recommendation module, which is electrically connected to the parking space evaluation coefficient calculation module, is used to sort the parking space evaluation coefficients of each parking space from large to small, and recommend the parking space number and parking lot entrance corresponding to the parking space evaluation coefficient greater than a preset threshold to the driver;

[0122] It should be noted that when the parking space number and parking lot entrance corresponding to the parking space evaluation coefficient greater than the preset threshold are recommended to the driver, the parking lot entrance is the parking lot entrance corresponding to the driving convenience coefficient used in calculating the parking space evaluation coefficient;

[0123] As an implementation mode, the preset threshold value may be 0, and when making a recommendation, the parking space numbers and parking lot entrances corresponding to the largest three, five, or ten parking space evaluation coefficients are recommended to the driver.

[0124] Embodiment 2:

[0125] See also Figure 4 The present invention provides a parking guidance method based on image recognition according to an embodiment, which is used in the parking guidance system based on image recognition in embodiment 1, and comprises the following steps:

[0126] S1, collect real-time images of parking spaces, the number of vehicles entering the parking lot entrance, and obtain the length of the road from the parking space to the parking lot entrance and exit;

[0127] S2, processing the real-time image at the parking space to obtain information on whether the parking space is an empty space, and taking the ratio of the number of vehicles entering the previous time period and the time period length when the real-time image is collected as the vehicle flow per unit time;

[0128] S3, generating a parking convenience coefficient according to information on whether the parking space is an empty space;

[0129] S4, generating a departure convenience coefficient according to information on whether the parking space is an empty space and the length of the road from the parking space to the parking lot exit;

[0130] S5, generating a driving convenience coefficient according to information on whether the parking space is an empty space, the length of the road from the parking space to the entrance of the parking lot, and the traffic volume per unit time;

[0131] S6, generating a parking space evaluation coefficient according to the parking convenience coefficient, the exit convenience coefficient, and the driving convenience coefficient;

[0132] S7, sorting the parking space evaluation coefficients of the parking spaces from large to small, and recommending the parking space numbers and parking lot entrances corresponding to the parking space evaluation coefficients greater than a preset threshold to the driver.

[0133] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0134] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A parking guidance system based on image recognition, characterized in that: include: A data acquisition module, the data acquisition module is used to collect real-time images of parking spaces, the number of vehicles entering the parking lot entrance, and the length of the road from the parking space to the parking lot entrance and exit; A data processing module, the data processing module is used to process the real-time image at the parking space to obtain information on whether the parking space is an empty space, and to use the ratio of the number of vehicles entering the previous time period and the time period length as the vehicle flow per unit time; A parking convenience coefficient calculation module, the parking convenience coefficient calculation module is used to generate a parking convenience coefficient according to information on whether a parking space is an empty space; A departure convenience coefficient calculation module, the departure convenience coefficient calculation module is used to generate a departure convenience coefficient according to information on whether the parking space is an empty space and the length of the road from the parking space to the parking lot exit; A driving convenience coefficient calculation module, the driving convenience coefficient calculation module is used to generate a driving convenience coefficient according to information on whether the parking space is an empty space, the length of the road from the parking space to the entrance of the parking lot, and the traffic volume per unit time; A parking space evaluation coefficient calculation module, the parking space evaluation coefficient calculation module is used to generate a parking space evaluation coefficient according to a parking convenience coefficient, a driving convenience coefficient, and a driving convenience coefficient; A parking space recommendation module, which is used to sort the parking space evaluation coefficients of various parking spaces from large to small, and recommend parking space numbers and parking lot entrances corresponding to parking space evaluation coefficients greater than a preset threshold to the driver; The length of the road from the i-th parking space to the j-th parking lot entrance is calibrated as , and the road length from the i-th parking space to the k-th parking lot exit is calibrated as , the vehicle flow per unit time at the entrance of the jth parking lot is calibrated as , the information of whether the i-th parking space is an empty parking space is marked as , and when the i-th parking space is empty, , and when the i-th parking space is not empty, , the parking convenience coefficient of the i-th parking space is calibrated as , the departure convenience coefficient of the i-th parking space is calibrated as , the driving convenience coefficient corresponding to the i-th parking space is calibrated as ,in: i represents the number of different parking spaces, and i=1, 2, 3, ..., m, and , j represents the number of different entrances in the parking lot, and j=1, 2, 3, ..., X, and , k represents the number of different exits in the parking lot, and k=1, 2, 3, ..., Y, and ; The parking convenience coefficient calculation formula is as follows: in, is the correction coefficient corresponding to the i-th parking space, and , and the more empty parking spaces there are adjacent to the i-th parking space, The bigger; The departure convenience coefficient calculation module includes a first assignment unit and a first calculation unit, wherein: The first assignment unit is used to generate a departure coefficient according to the length of the road from the parking space to the parking lot exit, and assign a value to the departure coefficient to generate a departure influence value between 0 and 1 that decreases as the departure coefficient increases. The calculation formula of the departure coefficient is as follows: in, Indicates the maximum length of the road from the parking space to the parking lot exit. Indicates the minimum length of the road from the parking space to the parking lot exit. represents the length of the road from the i-th parking space to the nearest parking lot exit, represents the departure coefficient corresponding to the i-th parking space; The first calculation unit is used to generate a departure convenience coefficient according to information about whether the parking space is an empty parking space and the departure impact value, and the calculation formula is as follows: in, represents the departure impact value corresponding to the i-th parking space; The driving convenience coefficient calculation module includes a second assignment unit and a second calculation unit, wherein: The second assignment unit is used to generate a driving path coefficient according to the road length from the parking space to the parking lot entrance, and assign a value to the driving path coefficient to generate a first driving impact value between 0 and 1 that decreases as the driving coefficient increases, and assign a value to the vehicle flow per unit time to generate a second driving impact value between 0 and 1 that decreases as the vehicle flow per unit time increases, wherein the calculation formula of the driving path coefficient is as follows: in, Indicates the maximum length of the road from the parking space to the parking lot entrance. Indicates the minimum length of the road from the parking space to the parking lot entrance. represents the driving path coefficient corresponding to the jth parking lot entrance to the i-th parking space; The second calculation unit is used to generate a driving convenience coefficient according to the first driving impact value, the second driving impact value, and information about whether the parking space is an empty space, and the calculation formula is as follows: in, represents the driving coefficient from the jth parking lot entrance to the i-th parking space, represents the first driving impact value corresponding to the j-th parking lot entrance to the i-th parking space, Represents the vehicle flow per unit time at the entrance of the jth parking lot The corresponding second driving influence value, are preset scale factors, and ,and , driving convenience coefficient is the maximum value of the driving coefficients corresponding to the Y parking lot entrances to the i-th parking space; The calculation formula of the parking space evaluation coefficient is as follows: in, represents the parking space evaluation coefficient of the i-th parking space, are preset scale factors, and ,and .

2. The parking guidance system based on image recognition according to claim 1, characterized in that: When there is no empty parking space in the parking space adjacent to the i-th parking space, let , when there is an empty parking space in the parking space adjacent to the i-th parking space, let , when there are two empty parking spaces adjacent to the i-th parking space, let .

3. The parking guidance system based on image recognition according to claim 1, characterized in that: The departure influence value corresponding to the i-th parking space is calibrated as , the departure impact value The assignment criteria are as follows: When satisfied When the driving away impact value ; When satisfied When the driving away impact value ; When satisfied When the driving away impact value ; When satisfied When the driving away impact value ; When satisfied When the driving away impact value .

4. The parking guidance system based on image recognition according to claim 1, characterized in that: The first driving impact value The assignment criteria are as follows: When satisfied When the first driving impact value ; When satisfied When the first driving impact value ; When satisfied When the first driving impact value ; When satisfied When the first driving impact value ; When satisfied When the first driving impact value ; The second driving impact value The assignment criteria are as follows: When satisfied When the second driving influence value ; When satisfied When the second driving influence value ; When satisfied When the second driving influence value ; When satisfied When the second driving influence value ; When satisfied When the second driving influence value .

5. A parking guidance method based on image recognition, used in the parking guidance system based on image recognition according to any one of claims 1 to 4, characterized in that: The steps include: S1, collecting real-time images of parking spaces, the number of vehicles entering the parking lot entrance, and obtaining the length of the road from the parking space to the parking lot entrance and exit; S2, processing the real-time image at the parking space to obtain information on whether the parking space is an empty space, and taking the ratio of the number of vehicles entering the previous time period and the time period length when the real-time image is collected as the vehicle flow per unit time; S3, generating a parking convenience coefficient according to information on whether the parking space is an empty space; S4, generating a departure convenience coefficient according to information on whether the parking space is an empty space and the length of the road from the parking space to the parking lot exit; S5, generating a driving convenience coefficient according to information on whether the parking space is an empty space, the length of the road from the parking space to the entrance of the parking lot, and the traffic volume per unit time; S6, generating a parking space evaluation coefficient according to the parking convenience coefficient, the exit convenience coefficient, and the driving convenience coefficient; S7, sorting the parking space evaluation coefficients of the parking spaces from large to small, and recommending the parking space numbers and parking lot entrances corresponding to the parking space evaluation coefficients that are greater than a preset threshold to the driver.

Images