Intelligent parking system based on Internet of Things

Abstract

The invention discloses an intelligent parking system based on the Internet of Things, and relates to the technical field of the Internet of Things, and the system comprises a management and control module which is connected with a database, a service terminal, an anti-collision module and an early warning module. The database is used for storing the parking space condition of the parking lot and generating a corresponding early warning signal according to the parking space condition; the anti-collision module is used for acquiring a parking image, judging whether a vehicle has the possibility of collision in a parking process according to four-corner coordinates of a license plate in the parking image, judging whether a parking position of a user is safe according to the four-corner coordinates of the license plate, and generating a corresponding early warning signal; the early warning module is used for correspondingly processing the generated early warning signal; according to the system, the parking image is obtained through the anti-collision unit, whether a collision possibility exists in the parking process of a user is judged through the license plate area in the parking image, and meanwhile whether the parking position is safe or not is judged through the license plate area.

Description

Technical Field

[0001] The present invention relates to the technical field of the Internet of Things, and specifically to an intelligent parking system based on the Internet of Things. Background Art

[0002] With the development of social economy, the number of automobiles has increased sharply, and the requirements for parking lots, toll management, and security management have also been increasing day by day. Using electronic information technology to achieve safe and efficient intelligent management has become the main development direction of vehicle management.

[0003] In the prior art, vehicle-mounted radars are only arranged in the four corners of the vehicle, and the anti-collision area is limited. Therefore, an intelligent parking system based on the Internet of Things is provided. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent parking system based on the Internet of Things.

[0005] The purpose of the present invention can be achieved by the following technical solutions: An intelligent parking system based on the Internet of Things includes a control module, and the control module is connected to a database, a service terminal, an anti-collision module, and an early warning module;

[0006] The database is used to store the parking space conditions of the parking lot and generate corresponding early warning signals according to the parking space conditions;

[0007] The service terminal is used to provide parking spaces and parking routes for users;

[0008] The anti-collision module determines whether there is a possibility of collision during the parking process of the vehicle by obtaining the parking image, and determines whether the parking position of the user is safe according to the four-corner coordinates of the license plate in the parking image, and generates a corresponding early warning signal;

[0009] The early warning module is used to perform corresponding processing on the generated early warning signals.

[0010] Further, the parking space conditions include parking space numbers and parking states; the parking states include "yes", "no", and "to be parked", and the initial values of the parking states are all "no";

[0011] Obtain the number of parking space numbers with the vehicle state of "yes", denoted as C yes ; Obtain the number of parking space numbers with the vehicle state of "to be parked", denoted as C wait ; Obtain the total number of vehicle numbers, denoted as C all ;

[0012] When and , generate a first-level early warning signal; when , generate a second-level early warning signal.

[0013] Furthermore, a pressure sensor, a camera, and an image processing unit are provided inside the anti-collision module;

[0014] The pressure sensor is used to determine whether there is a vehicle performing a parking action in the parking space;

[0015] The camera is used to capture the parking image of the vehicle;

[0016] The image processing unit is used to process the parking image, obtain the four-corner coordinates of the license plate, so as to judge whether there is a possibility of colliding with foreign objects during the parking process of the user, and judge whether the parking position of the user is safe.

[0017] Furthermore, the process by which the pressure sensor determines whether there is a vehicle performing a parking action in the parking space is as follows:

[0018] The pressure sensor is placed on the vehicle entry section of the parking space;

[0019] Set thresholds T s and M s ;

[0020] When the time when the pressure sensor senses pressure is greater than T s , and the sensed pressure continues to be greater than or equal to M s , it is determined that there is a vehicle performing a parking action in this parking space, the camera is activated, and the parking image is captured in real time.

[0021] Furthermore, after receiving the parking image, the image processing unit obtains the histogram of the parking image;

[0022] Set threshold D p ;

[0023] Obtain the total number C of pixel points of the parking image ap ;

[0024] Respectively obtain the pixel values of each pixel point, mark the pixel points with pixel values less than or equal to 50 as dark pixel points, and obtain the number C of dark pixel points dp ;

[0025] When is greater than or equal to D p , it is determined that the tone of the parking image is too dark, and the parking image is sharpened.

[0026] Furthermore, the process of obtaining the four-corner coordinates of the license plate is as follows:

[0027] Identify the license plate in the parking image through deep learning to obtain the license plate area;

[0028] Set pixel value Dc1 and D c2 and D c1 is less than D c2 ;

[0029] When the pixel value of a pixel point in the license plate area is greater than or equal to D c1 and less than or equal to D c2 , then mark this pixel point as a suspected border point;

[0030] Taking the lower left corner of the license plate area as the origin, establish a two-dimensional coordinate system, and taking one pixel point as a unit, obtain the coordinates of the suspected border points;

[0031] Set the threshold D L and D W according to the number of suspected border points, where D L is greater than D W ;

[0032] When there are greater than or equal to D L suspected border points that satisfy the function y 1 = ax + b 1 or y' 1 = a'x + b' 1 = a' 1 x + b' 1 1 or y' 1 , then mark the suspected border points that satisfy the function y

[0033] Then, according to the y-values of the long border points, judge the up-down relationship between the function y 1 and the function y' 1 ;

[0034] When there are greater than or equal to D W and less than D L suspected border points that satisfy the function y 2 = ax + b 2 or y' 2 = a'x + b' 2 = a' 2 x + b' 2 2 or y' 2 , then mark the suspected border points that satisfy the function y

[0035] Then, according to the x-values of the wide border points, judge the left-right relationship between the function y 2 and the function y' 2 ;

[0036] Obtain the function y 1 , the function y' 1 , the function y 2 and the function y'​​2 The intersection points therebetween, i.e., obtaining the four corner coordinates of the license plate, denoted as point T, point B, point R, and point L.

[0037] Further, the process by which the image processing unit determines whether there is a possible collision during the parking process of the user is as follows:

[0038] Taking the lower left corner of the parking image as the origin, establishing a two-dimensional coordinate system with one pixel point as the unit;

[0039] Using the two-dimensional coordinate system of the parking image to replace the two-dimensional coordinate system of the license plate area;

[0040] Setting a safety area and threshold A s1 and threshold A s2 , and threshold A s1 is less than threshold A s2 ;

[0041] When the license plate area is completely within the safety area, the parking action of the vehicle is safe and there is no collision situation;

[0042] Obtaining the border coordinates of the safety area and generating a coordinate set;

[0043] Assuming that the initial positions of point T, point B, point R, and point L are all within the safety area;

[0044] Respectively obtaining the distances and movement directions of point T, point B, point R, and point L to the coordinate set, and marking the point with the distance being 0 first as the critical point;

[0045] Obtaining the distance and movement direction of the critical point in real time;

[0046] When the movement direction remains unchanged, the distance is greater than 0 and less than or equal to threshold A s1 at this time, generating a level three warning signal, when the distance is greater than threshold A s1 and less than or equal to threshold A s2 at this time, generating a level four warning signal, when the distance is greater than threshold A s2 at this time, generating a level five warning signal.

[0047] Further, the process by which the image processing unit determines whether the parking position of the user is safe is as follows:

[0048] After the parking action of the vehicle ends, obtaining b 1 and b' 1 ;

[0049] Setting threshold g 1 and g 2 ;

[0050] When |b 1 - b' 1 | is greater than or equal to g1 When it is, an inclination warning signal is generated;

[0051] When b 1 is greater than g 2 When it is, a protrusion warning signal is generated.

[0052] Furthermore, the process of the warning module processing the warning signal is as follows:

[0053] When a first-level warning signal is received, the warning signal is sent to the administrator to inform the administrator that the parking spaces in the parking lot are about to be full, and the administrator makes corresponding handling;

[0054] When a second-level warning signal is received, the warning signal is sent to the administrator to inform the administrator that the parking spaces in the parking lot are full, and the administrator makes corresponding handling;

[0055] When a third-level warning signal is received, the warning signal is sent to the user to inform the user to drive carefully as the vehicle may collide;

[0056] When a fourth-level warning signal is received, the warning signal is sent to the user to inform the user that the vehicle is about to collide;

[0057] When a fifth-level warning signal is received, the warning signal is sent to the user to inform the user that the vehicle has collided;

[0058] When an inclination warning signal is received, the warning signal is sent to the user to inform the user that the parking position of the vehicle is too inclined and a scratching event may occur later, and the user makes corresponding handling;

[0059] When a protrusion warning signal is received, the warning signal is sent to the user to inform the user that the front or rear of the vehicle exceeds the parking line and a scratching event may occur later, and the user makes corresponding handling.

[0060] Compared with the prior art, the beneficial effects of the present invention are: This system determines whether there is a possibility of collision during the vehicle parking process by obtaining the parking image and judging the four-corner coordinates of the license plate in the parking image, and after the user parks the vehicle, determines whether the parking position of the user is safe according to the four-corner coordinates of the license plate. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is the schematic diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0062] As Figure 1 shown, an Internet of Things-based intelligent parking system includes a control module, and the control module is connected to a database, a service terminal, an anti-collision module, and a warning module;

[0063] The database is used to store the status of parking spaces in the parking lot. The specific process is as follows:

[0064] The status of the parking space includes the parking space number and the parking status. The parking status includes "Yes", "No", and "Reserved". If the parking status shows "Yes", it means there is a vehicle on this parking space and parking cannot be carried out. If the parking status shows "No", it means there is no vehicle on this parking space and parking can be carried out. If the parking status shows "Reserved", it means this parking space has been reserved by others and parking cannot be carried out. The initial value of the parking status is "No".

[0065] Obtain the number of parking space numbers with the vehicle status of "Yes", denoted as C yes ; Obtain the number of parking space numbers with the vehicle status of "Reserved", denoted as C wait ; Obtain the total number of vehicle numbers, denoted as C all ;

[0066] When and , generate a first-level warning signal; when , generate a second-level warning signal, and submit the generated warning signal to the warning module for corresponding processing by the warning module;

[0067] The service terminal is used to provide parking spaces and parking routes for users. The specific process is as follows:

[0068] An application port and a path generation unit are set in the service terminal;

[0069] The user scans the QR code at the entrance of the parking lot through the application port to generate a usage application and submit the usage application to the control center;

[0070] After receiving the usage application, the control center queries the database;

[0071] If there is no parking space with the parking status of "No" in the database, the user will be informed that parking is not available;

[0072] If there is only one parking space with the parking status of "No" in the database, directly obtain the parking space number of this parking space, and send the parking space number to the path generation unit. After the path generation unit plans the driving path, send the driving path to the user to guide the user to drive to the parking space; at the same time, generate change information according to the parking space number and submit the change information to the database to change the parking status of the parking space number to "Reserved";

[0073] If there are multiple parking spaces in the database with a parking status of "No", the "Automatic Allocation" and "Self-Selection" functions are provided to the user; if the user selects "Automatic Allocation", the parking space number that is the closest to the parking lot entrance and has a parking status of "No" is obtained, and the parking space number is submitted to the path generation unit. After the path generation unit generates a driving path, the driving path is sent to the user; if the user selects "Self-Selection", a query application is submitted to the database to obtain the parking space numbers with a parking status of "No", and a distribution map of available parking spaces is generated through GIS technology. The distribution map of available parking spaces is sent to the user, and the user selects a parking space for parking. The parking space number selected by the user is obtained and submitted to the path generation unit. After the path generation unit generates a driving path, the driving path is sent to the user; meanwhile, change information is generated according to the parking space number and submitted to the database to change the parking status of the parking space number to "To Be Parked".

[0074] The anti-collision module is used to prevent the user from colliding with foreign objects during parking. The specific process is as follows:

[0075] A pressure sensor, a camera, and an image processing unit are provided in the anti-collision module;

[0076] The pressure sensor is used to determine whether there is a vehicle performing a parking action in the parking space;

[0077] The camera is used to capture the parking image of the vehicle;

[0078] The image processing unit is used to process the parking image, obtain the four corner coordinates of the license plate, and determine whether there is a possibility of colliding with foreign objects during the user's parking process;

[0079] The pressure sensor is placed on the vehicle entry section of the parking space;

[0080] Set threshold values T s and M s ;

[0081] When the time when the pressure sensor senses pressure is greater than T s , and the sensed pressure continuously remains greater than or equal to M s , it is determined that there is a vehicle performing a parking action in this parking space, the camera is started, and the parking image is captured in real time;

[0082] Obtain the pressure value sensed by the pressure sensor, denoted as D s ;

[0083] When D s is greater than or equal to M s , start timing. During the timing process, because there is a possibility of vehicle movement, the value of D s will show a fluctuating change;

[0084] Set the timing time to T' s ;

[0085] Within the time range of T' s if D s always remains greater than or equal to M s , then this T' s is valid timing. If within the time range of T' s , the situation where D s is less than M s occurs, then this T' s is invalid timing and needs to be timed again;

[0086] When the valid timing T' s is greater than T s , it is determined that there is a vehicle performing a parking action in this parking space, and the camera will be activated to start capturing the parking image of the vehicle;

[0087] The camera transmits the captured parking image to the image processing unit in real time, and the image processing unit makes corresponding processing;

[0088] The process of the image processing unit processing the parking image is as follows:

[0089] After receiving the parking image, obtain the histogram of the parking image;

[0090] Set the threshold D p ;

[0091] Obtain the total number of pixel points of the parking image, denoted as C ap ;

[0092] Respectively obtain the pixel values of each pixel point, mark the pixel points with pixel values less than or equal to 50 as dark pixel points, and obtain the number of dark pixel points, denoted as C dp ;

[0093] When is greater than or equal to the threshold D p , it is determined that the tone of this parking image is too dark, and the parking image is sharpened;

[0094] If the overall tone of the parking image is dull due to light problems, the position of the license plate in the parking image cannot be accurately obtained. Therefore, the parking image needs to be sharpened to accurately obtain the position information of the license plate in the parking image;

[0095] The process of the sharpening is as follows:

[0096] Obtain the pixel values of pixel points in sequence. On the basis of the original pixel values, add a constant to generate new pixel values, and replace the original pixel values with the new pixel values;

[0097] Set the segmentation point D 1 and D 2 , and set the pixel value as x;

[0098] Adopt different transformation functions for the pixel points in interval A, interval B, and interval C respectively;

[0099] The said interval A is 0 ≤ x < D 1 , the said interval B is D 1 ≤ x < D 2 , the said interval C is D 2 ≤ x ≤ 255;

[0100] Then use gamma correction to generate a new parking image, and cover the original parking image with the new parking image;

[0101] Identify the license plate in the parking image through deep learning to obtain the license plate area;

[0102] Set the pixel value D c1 and D c2 , and D c1 is less than D c2 ;

[0103] When the pixel value of the pixel point in the license plate area is greater than or equal to D c1 and less than or equal to D c2 , then it is determined that the pixel point may be a pixel point of the license plate border, and mark it as a suspected border point;

[0104] Taking the lower left corner of the license plate area as the origin, establish a two-dimensional coordinate system, and take one pixel point as the unit to obtain the coordinates of the suspected border points;

[0105] Set the threshold D L and D W according to the number of suspected border points, where D L is greater than D W ;

[0106] When there are greater than or equal to D L suspected border points satisfying the function y 1 = a 1 x + b 1 or y' 1 = a' 1 x + b' 1 , then mark the suspected border points satisfying the function y 1 or y' 1 as long border points;

[0107] Then, based on the y-value of the long border point, determine the function y 1 and the function y' 1 for their up and down relationship;

[0108] If the y-value of the long border point coordinate on the function y 1 is greater than the y-value of the long border point coordinate on the function y' 1 , then the function y 1 represents the upper border of the license plate, and the function y' 1 represents the lower border of the license plate;

[0109] If the y-value of the long border point coordinate on the function y 1 is less than the y-value of the long border point coordinate on the function y' 1 , then the function y' 1 represents the upper border of the license plate, and the function y 1 represents the lower border of the license plate;

[0110] When there are greater than or equal to D W and less than D L suspected border points that satisfy the function y 2 = a 2 x + b 2 or y' 2 = a' 2 x + b' 2 , then mark the suspected border points that satisfy the function y 2 or y' 2 as wide border points;

[0111] Then, based on the x-value of the wide border point, determine the left and right relationship between the function y 2 and the function y' 2 ;

[0112] If the x-value of the wide border point coordinate on the function y 2 is less than the x-value of the wide border point coordinate on the function y' 2 , then the function y 2 represents the left border of the license plate, and the function y' 2 represents the right border of the license plate;

[0113] If the x-value of the wide border point coordinate on the function y 2 is greater than the x-value of the wide border point coordinate on the function y' 2 , then the function y' 2 represents the left border of the license plate, and the function y 2 represents the right border of the license plate;

[0114] Obtain the function y 1 , the function y' 1 , the function y 2 and the function y'2 The intersection points therebetween, i.e., obtain the four corner coordinates of the license plate, denoted as point T, point B, point R, and point L;

[0115] Take the lower left corner of the parking image as the origin to establish a two-dimensional coordinate system with one pixel point as the unit;

[0116] Fuse the two-dimensional coordinate system of the license plate area into the two-dimensional coordinate system of the parking image, and use the two-dimensional coordinate system of the parking image to replace the two-dimensional coordinate system of the license plate area; since the units are the same, there is no case of scale distortion, only the coordinates of the original license plate area have corresponding changes;

[0117] Set a safety area and threshold A s1 and threshold A s2 , and threshold A s1 is less than threshold A s2 ;

[0118] When the license plate area is completely within the safety area, the parking action of the vehicle is safe and there is no collision;

[0119] Obtain the border coordinates of the safety area and generate a coordinate set K = {k 1 , k 2 , ……, k n}, where n is a positive integer;

[0120] Assume that the initial positions of point T, point B, point R, and point L are all within the safety area;

[0121] Respectively obtain the distances and movement directions of point T, point B, point R, and point L to the coordinate set K, and mark the point with the distance being 0 first as the critical point;

[0122] Obtain the distance and movement direction of the critical point in real time;

[0123] When the movement direction remains unchanged, the distance is greater than 0 and less than or equal to threshold A s1 , generate a level 3 warning signal, when the distance is greater than threshold A s1 and less than or equal to threshold A s2 , generate a level 4 warning signal, when the distance is greater than threshold A s2 , generate a level 5 warning signal; submit the generated warning signal to the warning module, and the warning module makes corresponding processing;

[0124] Set time T m ;

[0125] When the coordinates of point T, point B, point R, and point L do not change after time T m , it is determined that the parking action of the vehicle ends;

[0126] Generate change information according to the parking space number, submit the change information to the database, and change the parking status of the parking space number to "Yes";

[0127] After the parking action of the vehicle ends, obtain b 1 and b' 1 ;

[0128] Set thresholds g 1 and g 2 ;

[0129] When |b 1 -b' 1 | is greater than or equal to g 1 , then generate a tilt warning signal;

[0130] When b 1 is greater than g 2 , then generate a protrusion warning signal;

[0131] Send the generated warning signal to the warning module, and the warning module makes corresponding processing;

[0132] The warning module is used to process warning signals. The specific process is as follows:

[0133] When receiving a first-level warning signal, send the warning signal to the administrator to inform the administrator that the parking spaces in the parking lot are about to be full, and the administrator makes corresponding processing;

[0134] When receiving a second-level warning signal, send the warning signal to the administrator to inform the administrator that the parking spaces in the parking lot are full, and the administrator makes corresponding processing;

[0135] When receiving a third-level warning signal, send the warning signal to the user to inform the user to drive carefully as the vehicle may collide;

[0136] When receiving a fourth-level warning signal, send the warning signal to the user to inform the user that the vehicle is about to collide;

[0137] When receiving a fifth-level warning signal, send the warning signal to the user to inform the user that the vehicle has collided;

[0138] When receiving a tilt warning signal, send the warning signal to the user to inform the user that the parking position of the vehicle is too tilted and a scratching event may occur later, and the user makes corresponding processing;

[0139] When receiving a protrusion warning signal, send the warning signal to the user to inform the user that the front of the vehicle or the parking space exceeds the parking line and a scratching event may occur later, and the user makes corresponding processing.

[0140] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent parking system based on the Internet of Things, including a control module, characterized in that: The control module is connected to a database, a service terminal, an anti-collision module and an early warning module; The database is used to store the parking space status of the parking lot and generate corresponding warning signals according to the parking space status; The service terminal is used to provide users with parking spaces and parking routes; The anti-collision module obtains the parking image, determines whether the vehicle is likely to collide during parking according to the coordinates of the four corners of the license plate in the parking image, and determines whether the user's parking position is safe according to the coordinates of the four corners of the license plate, and generates a corresponding warning signal; The early warning module is used to process the generated early warning signal accordingly.

2. According to claim 1, the intelligent parking system based on the Internet of Things is characterized in that: The parking space status includes the parking space number and the parking status; the parking status includes "yes", "no" and "waiting to park", and the initial value of the parking status is "no"; Get the number of parking spaces with a vehicle status of "yes", denoted as C yes ; Get the number of parking spaces with a vehicle status of "waiting to park", denoted as C wait ; Get the total number of vehicle numbers, recorded as C all ; when and When A secondary warning signal is generated.

3. The intelligent parking system based on the Internet of Things according to claim 2 is characterized in that: The anti-collision module is provided with a pressure sensor, a camera and an image processing unit; The pressure sensor is used to determine whether there is a vehicle in the parking space that is parking; The camera is used to capture a parking image of the vehicle; The image processing unit is used to process the parking image and obtain the coordinates of the four corners of the license plate to determine whether there is a possibility of collision with foreign objects during the parking process and whether the parking position of the user is safe.

4. The intelligent parking system based on the Internet of Things according to claim 3 is characterized in that: The process of the pressure sensor determining whether there is a vehicle in the parking space and parking is as follows: The pressure sensor is placed at the vehicle entrance section of the parking space; Set the threshold T s and M s ; When the pressure sensor senses pressure for a time greater than T s , and the pressure felt is continuously greater than or equal to M s , it is determined that there is a vehicle parking in the parking space, and the camera is started to capture the parking image in real time.

5. The intelligent parking system based on the Internet of Things according to claim 4 is characterized in that: After receiving the parking image, the image processing unit obtains a histogram of the parking image; Set the threshold D p ; Get the total number of pixels C of the parking image ap ; Get the pixel value of each pixel respectively, mark the pixel with a pixel value less than or equal to 50 as a dark pixel, and get the number of dark pixels C dp ; when Greater than or equal to D p When the parking image is dark in tone, the parking image is judged to be dark in tone, and the parking image is clarified.

6. The intelligent parking system based on the Internet of Things according to claim 5 is characterized in that: The process of obtaining the coordinates of the four corners of the license plate is: Recognize the license plate in the parking image through deep learning and obtain the license plate area; Set pixel value D c1 and D c2 , and D c1 Less than D c2 ; When the pixel value of the pixel point in the license plate area is greater than or equal to D c1 and less than or equal to D c2 , then the pixel is marked as a suspected border point; Take the lower left corner of the license plate area as the origin, establish a two-dimensional coordinate system, and obtain the coordinates of the suspected border points in units of one pixel; According to the number of suspected border points, set the threshold D L and D W , where D L Greater than D W ; When there is greater than or equal to D L When a suspected border point satisfies the function y1=a1x+b1 or y'1=a'1x+b'1, the suspected border point satisfying the function y1 or y'1 is marked as a long border point; Then, according to the y value of the long border point, determine the upper and lower relationship between function y1 and function y'1; When there is greater than or equal to D W and less than D L When a suspected border point satisfies the function y2=a2x+b2 or y'2=a'2x+b'2, the suspected border point satisfying the function y2 or y'2 is marked as a wide border point; Then, according to the x value of the wide border point, determine the left-right relationship between function y2 and function y'2; Obtain the intersection points among function y1, function y'1, function y2 and function y'2, that is, obtain the coordinates of the four corners of the license plate, recorded as point T, point B, point R and point L.

7. The intelligent parking system based on the Internet of Things according to claim 6 is characterized in that: The process of the image processing unit determining whether there is a possibility of collision during the parking process of the user is as follows: Take the lower left corner of the parking image as the origin and establish a two-dimensional coordinate system with one pixel as the unit; Use the two-dimensional coordinate system of the parking image to replace the two-dimensional coordinate system of the license plate area; Set the safety zone and threshold A s1 and threshold A s2 , and the threshold A s1 Less than threshold A s2 ; When the license plate area is completely within the safe area, the parking action of the vehicle is safe and there is no collision. Get the border coordinates of the safe area and generate a coordinate set; Assume that the initial positions of points T, B, R, and L are all within the safe area; The distances and movement directions of point T, point B, point R and point L to the coordinate set are obtained respectively, and the point whose distance is first 0 is marked as a dangerous point; Obtain the distance and movement direction of the danger point in real time; When the direction of movement remains unchanged, the distance is greater than 0 and less than or equal to the threshold A s1 When the distance is greater than the threshold A, a third-level warning signal is generated. s1 and is less than or equal to the threshold A s2 When the distance is greater than the threshold A, a level 4 warning signal is generated. s2 A level 5 warning signal is generated.

8. The intelligent parking system based on the Internet of Things according to claim 7 is characterized in that: The process of the image processing unit determining whether the user's parking position is safe is as follows: When the vehicle's parking action is completed, b1 and b'1 are obtained; Set thresholds g1 and g2; When |b1-b'1| is greater than or equal to g1, a tilt warning signal is generated; When b1 is greater than g2, a prominent warning signal is generated.

9. The intelligent parking system based on the Internet of Things according to claim 8, characterized in that: The process of the early warning module processing the early warning signal is as follows: When a level 1 warning signal is received, the warning signal is sent to the administrator, informing the administrator that the parking spaces in the parking lot are about to be full, and the administrator will take corresponding measures; When a level 2 warning signal is received, the warning signal is sent to the administrator, informing the administrator that the parking spaces in the parking lot are full, and the administrator will take corresponding measures; When a level 3 warning signal is received, a warning signal is sent to the user to inform the user to drive carefully as the vehicle may collide; When a level 4 warning signal is received, a warning signal is sent to the user to inform the user that the vehicle is about to collide; When a level 5 warning signal is received, a warning signal is sent to the user to inform the user that the vehicle has collided; When a tilt warning signal is received, a warning signal is sent to the user to inform the user that the vehicle is parked too tilted and a scratch may occur in the future, and the user can take corresponding measures; When a protrusion warning signal is received, a warning signal is sent to the user to inform the user that the front of the vehicle or the parking space exceeds the parking line and a scratch incident may occur subsequently, and the user is required to take corresponding measures.

Images