Berth Status Analysis Method and System Based on Cloud-Edge Integration

Abstract

The present application discloses a berth status analysis method and system based on cloud-edge integration. The method includes: determining whether there is corresponding historical berth cache information for a berth number; if not, transforming the berth calibration coordinates according to a three-dimensional projection algorithm to obtain a three-dimensional berth projection box corresponding to the berth number, and calculating the parking status of each vehicle based on the three-dimensional berth projection box and each vehicle detection box; if so, determining whether each vehicle detection box exists in the historical berth cache information; if so, invoking the parking status of the vehicle corresponding to each vehicle detection box according to the historical berth cache information; if not, calculating the parking status of the vehicle corresponding to the vehicle detection box based on the three-dimensional berth projection box and the vehicle detection box; determining whether there is a vehicle in the berth corresponding to the berth number according to the parking status of each vehicle; if so, there is a vehicle in the berth corresponding to the berth number; if not, there is no vehicle in the berth corresponding to the berth number.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent traffic management, and in particular, to a berth status analysis method and system based on cloud-edge fusion. Background Art

[0002] With the rapid development of China's economy and society, the urbanization process is accelerating, the urban population is constantly concentrated, and at the same time, the number of various motor vehicles in the city is increasing year by year, which brings a series of urban traffic management problems. Berth management and vehicle parking difficulties are one of the problems. From the perspective of berth management, the management department is concerned about how many berths are distributed in the city, where the positions of each berth are on the map, which berths are being occupied by vehicles, which are vacant berths, and how high the frequency of berth use is. Similarly, from the perspective of vehicle users, car owners are also concerned about where there are parking spaces at a certain location and whether the actual use of the parking spaces is idle.

[0003] However, due to the large number of urban berths and the large amount of image or video data transmission in the traditional method, it is easy to have a backlog of computing tasks and drag down the cloud service, resulting in inaccurate berth status detection and low detection efficiency. Summary of the Invention

[0004] The purpose of the present application is to solve the technical problems of inaccurate berth status detection and low detection efficiency in the traditional method. To achieve the above purpose, the present application provides a berth status analysis method and system based on cloud-edge fusion.

[0005] The present application provides a berth status analysis method based on cloud-edge fusion, including:

[0006] Obtaining the vehicle detection frame of each vehicle corresponding to the berth panoramic image of each camera and the berth calibration coordinates of the berth number corresponding to each camera;

[0007] Judging whether there is corresponding historical berth cache information for the berth number;

[0008] If there is no corresponding historical berth cache information for the berth number, then transform the berth calibration coordinates according to the three-dimensional projection algorithm to obtain a three-dimensional berth projection frame corresponding to the berth number, and calculate the docking state of each vehicle according to the three-dimensional berth projection frame and each vehicle detection frame;

[0009] If there is corresponding historical berth cache information for the berth number, then judge whether each vehicle detection frame exists in the historical berth cache information;

[0010] If so, then call the docking state of the vehicle corresponding to each vehicle detection frame according to the historical berth cache information;

[0011] If not, calculate the parking state of the vehicle corresponding to the vehicle detection frame according to the three-dimensional berth projection frame and the vehicle detection frame;

[0012] Judge whether there is a vehicle in the berth corresponding to the berth number according to the parking state of each vehicle;

[0013] If so, there is a vehicle in the berth corresponding to the berth number;

[0014] If not, there is no vehicle in the berth corresponding to the berth number;

[0015] Form new berth cache information with the vehicle parking calculation result corresponding to the berth number for the next berth state determination.

[0016] This application provides a berth state analysis system based on cloud-edge fusion, including:

[0017] A camera data acquisition module, configured to acquire the vehicle detection frame of each vehicle corresponding to the panoramic view of the berth of each camera and the berth calibration coordinates of the berth number corresponding to each camera;

[0018] A historical cache judgment module, configured to judge whether there is corresponding historical berth cache information for the berth number;

[0019] A calculation module, configured to, if there is no corresponding historical berth cache information for the berth number, transform the berth calibration coordinates according to a three-dimensional projection algorithm to obtain a three-dimensional berth projection frame corresponding to the berth number, and calculate the parking state of each vehicle according to the three-dimensional berth projection frame and each vehicle detection frame;

[0020] A vehicle detection frame cache judgment module, configured to, if there is corresponding historical berth cache information for the berth number, judge whether each vehicle detection frame exists in the historical berth cache information;

[0021] A call cache module, configured to, if so, call the parking state of the vehicle corresponding to each vehicle detection frame according to the historical berth cache information;

[0022] A vehicle parking calculation module, configured to, if not, calculate the parking state of the vehicle corresponding to the vehicle detection frame according to the three-dimensional berth projection frame and the vehicle detection frame;

[0023] A vehicle-in-berth judgment module, configured to judge whether there is a vehicle in the berth corresponding to the berth number according to the parking state of each vehicle;

[0024] A vehicle-in-berth judgment result module one, configured to, if so, there is a vehicle in the berth corresponding to the berth number;

[0025] The vehicle judgment result module two in the berth is used to determine that if the answer is no, there is no vehicle in the berth corresponding to the berth number.

[0026] The cache module is used to form new berth cache information from the vehicle docking calculation result corresponding to the berth number for the next berth status determination.

[0027] In the above-mentioned berth status analysis method and system based on cloud-edge integration, data is detected and obtained by edge devices and transmitted to the cloud service for data fusion, calculation, and caching. The docking status of the vehicle can be calculated in real time based on the three-dimensional berth projection frame and the vehicle detection frame of the berth, or the docking status of the vehicle corresponding to the vehicle detection frame can be found and called in the historical berth cache information, so as to judge whether each vehicle is in the berth, and thus determine whether there is a vehicle in the berth corresponding to the berth number. Through the berth status analysis method based on cloud-edge integration provided by this application, the berth status can be analyzed accurately in real time for different situations, the detection of the berth status can be realized, the detection efficiency can be improved, and the analysis of the berth status in the scenario of large-scale edge device access, high frequency, and high real-time data processing can be solved, providing data support for each business. Description of the Drawings

[0028] Figure 1 It is a schematic flow chart of the steps of the berth status analysis method based on cloud-edge integration provided by this application. Figure 2 It is a schematic structural diagram of the circumscribed rectangle of the vehicle detection frame, the three-dimensional berth projection frame, the three-dimensional berth projection frame, and the intersection area in an embodiment provided by this application. Figure 3 It is a schematic structural diagram of each parameter of the vehicle detection frame and the three-dimensional berth projection frame in an embodiment provided by this application. Figure 4 It is a schematic structural diagram of the berth status analysis system based on cloud-edge integration provided by this application. Detailed Embodiment

[0029] The technical solution of this application will be further described in detail below through the drawings and embodiments.

[0030] Please refer to Figure 1 , this application provides a berth status analysis method based on cloud-edge integration, including:

[0031] S10. Obtain the vehicle detection frame of each vehicle corresponding to the panoramic view of the berth of each camera and the berth calibration coordinates of the berth number corresponding to each camera.

[0032] S20. Judge whether there is corresponding historical berth cache information for the berth number.

[0033] S30. If there is no corresponding historical berth cache information for the berth number, transform the berth calibration coordinates according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection box corresponding to the berth number, and calculate the parking state of each vehicle based on the three-dimensional berth projection box and each vehicle detection box.

[0034] S40. If there is corresponding historical berth cache information for the berth number, determine whether each vehicle detection box exists in the historical berth cache information.

[0035] S50. If so, call the parking state of the vehicle corresponding to each vehicle detection box according to the historical berth cache information.

[0036] S60. If not, calculate the parking state of the vehicle corresponding to the vehicle detection box based on the three-dimensional berth projection box and the vehicle detection box.

[0037] S70. Based on the parking state of each vehicle, determine whether there is a vehicle in the berth corresponding to the berth number.

[0038] S80. If so, there is a vehicle in the berth corresponding to the berth number.

[0039] S90. If not, there is no vehicle in the berth corresponding to the berth number.

[0040] S100. Form new berth cache information from the vehicle parking calculation results corresponding to the berth number for the next berth state determination.

[0041] In this embodiment, the edge device (which can also be called a high-position camera) takes a panoramic view of the berth, and uses the deep learning algorithm built into the edge device to detect the coordinates of all vehicles in the panoramic view, obtaining the vehicle detection box, detection time, berth number corresponding to each camera, and berth calibration coordinates corresponding to the berth number of each vehicle, and pushing the vehicle detection box and the berth calibration coordinates corresponding to the berth number to the cloud service for berth state analysis. The edge device has a deep learning algorithm built in, undertaking the target detection task in the panoramic view of the berth, saving a part of the computing resources for the cloud service. The edge device combines the obtained camera number, berth number, vehicle detection box, camera longitude and latitude, vehicle detection time, detection coordinates of the berth detection box, etc., and transfers them to the cloud service through the http call method, and the cloud service determines whether there is a vehicle in the berth to realize the analysis of the berth state.

[0042] Query on the cloud service according to the berth number whether there is corresponding historical berth cache information for the berth number. If there is no corresponding historical berth cache information for the berth number, it means that a newly entered vehicle is in the camera's field of view, and the relationship between the vehicle detection box and the three-dimensional berth projection box can be calculated. By judging the relationship between the vehicle detection box and the three-dimensional berth projection box, it is determined whether there is a vehicle parked in the berth.

[0043] The berth calibration coordinates form a berth detection frame. The berth has four vertices on the ground (e.g., upper left, upper right, lower left, lower right), and the four vertices of the berth in the air can be deduced. The eight vertices form a three-dimensional frame, forming a three-dimensional berth projection frame, realizing the conversion from two-dimensional berth calibration coordinates to a three-dimensional berth projection frame. Set the height. Based on the two-dimensional berth calibration coordinates and using a three-dimensional projection algorithm, the coordinates of the four vertices of the berth in the air can be calculated to obtain the three-dimensional berth projection frame. The three-dimensional berth projection frame represents the position of the berth corresponding to a berth number in the picture. The vehicle detection frame represents the position of the vehicle in the picture. Based on the three-dimensional berth projection frame and the vehicle detection frame, the relative position relationship between the vehicle and the berth in the picture can be judged in real time, and then it can be calculated whether each vehicle is parked in the berth, so as to obtain the parking state of each vehicle.

[0044] If there is corresponding historical berth cache information for the berth number, it means that the data detected by the camera is sent to the cloud service again, and it can be checked whether there is a historical vehicle detection frame in the cache information that is the same as the vehicle detection frame. All historical vehicle detection frames correspond to a result, that is, whether the corresponding vehicle is parked on the berth. After the search is completed, if there is a same one, the calculated parking result of the vehicle detection frame corresponding to the historical berth cache information can be directly called to obtain the parking state of the corresponding vehicle. In one embodiment, the current vehicle detection frame is compared with the historical vehicle detection frame in the historical berth cache information. If the upper, lower, left, and right boundaries of the current vehicle detection frame and the historical vehicle detection frame differ within the range of 0 to 20 pixel values, it is considered that the current vehicle detection frame and the historical vehicle detection frame are the same vehicle detection frame. Thus, if the vehicle detection frame exists in the historical berth cache information, the parking state of the vehicle corresponding to the vehicle detection frame can be directly called based on the historical berth cache information. If the vehicle detection frame does not exist in the historical berth cache information, the vehicle corresponding to the vehicle detection frame is a newly entered vehicle within the camera's field of view. The berth calibration coordinates can continue to be transformed according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and based on the three-dimensional berth projection frame and the vehicle detection frame, the parking state of the vehicle is calculated. The calculation process is the same as the calculation process in S30.

[0045] Steps S10 to S100 are continuously repeated to combine the berth number, detection time, vehicle detection frame, and the parking status of the vehicle corresponding to each vehicle detection frame into a data set, and cache it in the cloud based on the berth number as the query basis. When the next time point arrives, the calculation result of the parking status of the above steps can be reused, saving a part of the calculation amount. The cloud can provide an external http query interface to support real-time query of whether there is a vehicle on the berth under all cameras. By calculating or calling the parking status of each vehicle, it can be determined whether there is a vehicle in the berth corresponding to the berth number, realizing the analysis of the berth status in real time. At the same time, the calculation result of the parking status of each vehicle is stored to form a new berth cache information for caching, which can be used for the determination of the next berth status analysis, and the berth cache information is updated in real time to make the query result more efficient and accurate.

[0046] Through the berth status analysis method based on cloud-edge fusion provided by this application, data is detected and acquired based on edge devices and transmitted to the cloud service for data fusion, calculation, and caching. It can calculate the parking status of the vehicle in real time according to the three-dimensional berth projection frame and the vehicle detection frame of the berth, or search and call the parking status of the vehicle corresponding to the vehicle detection frame in the historical berth cache information, so as to judge whether each vehicle is in the berth, and thus judge whether there is a vehicle in the berth corresponding to the berth number. Through the berth status analysis method based on cloud-edge fusion provided by this application, the berth status can be analyzed accurately in real time for different situations, realizing the detection of the berth status, improving the detection efficiency, and solving the analysis of the berth status in the scenarios of large-scale edge device access, high frequency, and high real-time data processing, providing data support for each business.

[0047] In one embodiment, the berth status analysis method based on cloud-edge fusion can be run on a browser management platform, a mobile app, or other cloud services.

[0048] In one embodiment, S30 includes:

[0049] S301, calculate the intersection area between the circumscribed rectangle of the three-dimensional berth projection frame and each vehicle detection frame, and obtain the ratio of the intersection area to the vehicle detection frame area according to the intersection area and the vehicle detection frame area;

[0050] S302, obtain the ratio of the intersection area to the berth according to the intersection area and the area of the circumscribed rectangle of the three-dimensional berth projection frame;

[0051] S303, draw a parallel line parallel to the ground lower width boundary of the three-dimensional berth projection frame according to the midpoint of the ground boundary of the vehicle detection frame, and obtain the ground intersection point of the parallel line and the ground road-facing long boundary of the three-dimensional berth projection frame;

[0052] S304. Calculate the ground intersection-top difference from the abscissa of the ground intersection point to the abscissa of the vertex of the ground road length boundary, and the ground bottom-top difference from the abscissa of the bottom point of the ground road length boundary to the abscissa of the vertex of the ground road length boundary, and divide the ground intersection-top difference by the ground bottom-top difference to obtain the vehicle bottom value;

[0053] S305. Draw a parallel line parallel to the upper width boundary in the air of the three-dimensional berth projection frame according to the midpoint of the far ground boundary of the vehicle detection frame, and obtain the air intersection point of the parallel line and the air road length boundary of the three-dimensional berth projection frame;

[0054] S306. Calculate the air intersection-top difference from the abscissa of the air intersection point to the abscissa of the vertex of the air road length boundary, and the air bottom-top difference from the abscissa of the bottom point of the air road length boundary to the abscissa of the vertex of the air road length boundary, and divide the air intersection-top difference by the air bottom-top difference to obtain the vehicle top value;

[0055] S307. Calculate the first abscissa difference between the abscissa of the road boundary of the vehicle detection frame and the abscissa of the vertex of the ground road length boundary, calculate the second abscissa difference between the abscissa of the bottom point of the ground road length boundary and the abscissa of the vertex of the ground road length boundary, and divide the first abscissa difference by the second abscissa difference to obtain the vehicle left value;

[0056] S308. Calculate the third abscissa difference between the abscissa of the far road boundary of the vehicle detection frame and the abscissa of the vertex of the ground far road length boundary of the three-dimensional berth projection frame, calculate the fourth abscissa difference between the abscissa of the bottom point of the ground far road length boundary and the abscissa of the vertex of the ground far road length boundary, and divide the third abscissa difference by the fourth abscissa difference to obtain the vehicle right value;

[0057] S309. Calculate the boundary distances between the four boundaries of the vehicle detection frame and the four boundaries of the berth panoramic view respectively. If the boundary distance is less than the pixel threshold, the vehicle is a corner vehicle;

[0058] S310. If the vehicle detection frame has an intersection with the surrounding vehicle detection frames, and the ground boundary of the vehicle detection frame is above the ground boundary of the intersection vehicle detection frame, the vehicle is an occluded vehicle;

[0059] S311. Divide the height of the vehicle detection frame by the height of the circumscribed rectangle frame of the three-dimensional berth projection frame to obtain the vehicle-berth height ratio;

[0060] S312. If the intersection area ratio to the vehicle is less than the first vehicle ratio threshold, and the intersection area ratio to the berth is less than the first berth ratio threshold, the vehicle is not in the berth.

[0061] In this embodiment, in S301, the circumscribed rectangle of the three-dimensional berth projection frame can be understood as the circumscribed rectangle of 8 points after the three-dimensional projection of the berth, as can be seen in Figure 2 shown. Divide the intersection area by the area of the vehicle detection frame to obtain the ratio of the intersection area to the vehicle. In S302, divide the intersection area by the area of the circumscribed rectangle of the three-dimensional berth projection frame to obtain the ratio of the intersection area to the berth. In S303, the midpoint of the ground boundary of the vehicle detection frame can be understood as the midpoint of the lower boundary of the vehicle detection frame (taking Figure 3 the up, down, left, and right of the picture in as a reference). The four ground points of the three-dimensional berth projection frame can be understood as four points on the berth ground. Through the midpoint of the ground boundary of the vehicle detection frame, draw a parallel line to the lower-width boundary formed by the four points on the berth ground, as in Figure 3 the parallel line L shown. The lower-width boundary of the three-dimensional berth projection frame on the ground can be understood as the lower side on the berth ground. Take the upper and lower boundaries of the obtained panoramic berth map as a reference to determine the upper and lower positioning. The long boundary of the three-dimensional berth projection frame on the ground close to the road can be understood as the long boundary of the berth formed by the four points on the berth ground close to the vehicle driving road, and can also be understood as the left side on the berth ground. The ground intersection point is as shown by point m in Figure 3 .

[0062] In S304, calculate the difference between the abscissa of the ground intersection point m and the abscissa of the vertex a of the long boundary of the three-dimensional berth projection frame on the ground close to the road to obtain the ground intersection-vertex difference x m -x a . Calculate the difference between the abscissa of the bottom point b of the long boundary of the three-dimensional berth projection frame on the ground close to the road and the abscissa of the vertex a to obtain the ground bottom-vertex difference x b -x a . Divide x m -x a by x b -x a to obtain the vehicle bottom value. In S305, the midpoint of the far ground boundary of the vehicle detection frame can be understood as the midpoint of the upper boundary of the vehicle detection frame. The upper-width boundary of the three-dimensional berth projection frame in the air can be understood as the upper boundary of the quadrilateral formed by four points in the air, and can also be understood as the upper side of the quadrilateral formed by the berth in the air. The long boundary of the three-dimensional berth projection frame in the air close to the road can be understood as the long boundary of the quadrilateral formed by four points in the air of the berth close to the vehicle driving road side, that is, the left side of the quadrilateral formed by four points in the air of the berth. Draw a parallel line to the upper-width boundary of the three-dimensional berth projection frame in the air through the midpoint of the far ground boundary of the vehicle detection frame, as in Figure 3 the parallel line I shown, and the intersection point in the air is shown by point n in Figure 3 .

[0063] In S306, calculate the difference between the abscissa of the intersection point n in the air and the abscissa of the vertex a' of the long boundary of the three-dimensional berth projection frame in the air close to the road to obtain the air intersection-vertex difference x n-x a’ Calculate the aerial bottom-top difference x from the abscissa of the bottom point b' of the aerial length boundary along the road to the abscissa of the vertex a'. b’ -x a’ x n -x a’ Divide by x b’ -x a’ Obtain the roof value. In S307, the road-side boundary of the vehicle detection frame can be understood as the left side, which is a straight line parallel to the vertical coordinate axis. The ground road-side length boundary can be understood as the left side of the ground berth close to the vehicle driving road. Subtract the abscissa of the vertex a of the ground road-side length boundary from the abscissa of the road-side boundary of the vehicle detection frame to obtain the first abscissa difference. Subtract the abscissa of the vertex a of the ground road-side length boundary from the abscissa of the bottom point b of the ground road-side length boundary to obtain the second abscissa difference. Divide the first abscissa difference by the second abscissa difference to obtain the left-side value of the vehicle.

[0064] In S308, the far-road boundary of the vehicle detection frame can be understood as the right boundary of the vehicle detection frame. The ground far-road length boundary can be understood as the right side of the ground berth far from the vehicle driving road. Subtract the abscissa of the ground far-road length boundary of the three-dimensional berth projection frame from the abscissa of the far-road boundary of the vehicle detection frame to obtain the third abscissa difference. Subtract the abscissa of the vertex of the ground far-road length boundary from the abscissa of the bottom point of the ground far-road length boundary to obtain the fourth abscissa difference. Divide the third abscissa difference by the fourth abscissa difference to obtain the right-side value of the vehicle. In S309, if the distance differences between the left, right, upper, and lower boundary values of the vehicle detection frame and the left, right, upper, and lower distances of the panoramic view of the berth captured by the camera are less than the pixel threshold, it can be understood that the distance between the left boundary of the vehicle detection frame and the left boundary of the panoramic view of the berth captured by the camera is less than the pixel threshold, or the distance between the right boundary of the vehicle detection frame and the right boundary of the panoramic view of the berth captured by the camera is less than the pixel threshold, or the distance between the upper boundary of the vehicle detection frame and the upper boundary of the panoramic view of the berth captured by the camera is less than the pixel threshold, or the distance between the lower boundary of the vehicle detection frame and the lower boundary of the panoramic view of the berth captured by the camera is less than the pixel threshold, then the vehicle is considered a corner vehicle. In one embodiment, the pixel threshold can be set between 15 and 25 pixels, and the specific value can be set according to the actual situation.

[0065] In S310, if the vehicle detection frame has an intersection with other vehicle detection frames and there are overlapping parts between them. If the ground boundary of the vehicle detection frame, that is, the lower boundary, is above the ground boundary of the vehicle detection frame with an intersection, then the vehicle is considered to be blocked. In S311, refer to Figure 3, the height h1 of the vehicle detection frame, which can also be understood as the length of the left boundary of the vehicle detection frame, is divided by the height h2 of the circumscribed rectangle of the three-dimensional berth projection frame to obtain the vehicle-berth height ratio. In 312, the first vehicle occupancy ratio threshold and the first berth occupancy ratio threshold can be defined according to the actual situation of the application scenario. In one embodiment, the values of the first vehicle occupancy ratio threshold and the first berth occupancy ratio threshold are the same and can be set in the range of 0.49 to 0.51. In one embodiment, both the first vehicle occupancy ratio threshold and the first berth occupancy ratio threshold are set to 0.5, that is, if the ratio of the intersection area to the vehicle is less than 0.5 and the ratio of the intersection area to the berth is less than 0.5, then the vehicle is not in the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0066] In one embodiment, if the left side value of the vehicle is greater than the first lower left threshold and less than the first upper left threshold, and the ratio of the intersection area to the vehicle is greater than the second vehicle occupancy ratio threshold, and the roof value of the vehicle is greater than the first lower roof threshold and less than the first upper roof threshold, then the vehicle is in the berth.

[0067] The first lower left threshold and the first upper left threshold can be set according to the actual application scenario. In one embodiment, the first lower left threshold can be set in the range of -0.41 to -0.39, and the first upper left threshold can be set in the range of 0.39 to 0.41. In one embodiment, the first lower left threshold is set to -0.4 and the first upper left threshold is set to 0.4. The second vehicle occupancy ratio threshold can be set according to the actual application scenario. In one embodiment, the second vehicle occupancy ratio threshold can be set in the range of 0.89 to 0.91. In one embodiment, the second vehicle occupancy ratio threshold can be set to 0.9. The first lower roof threshold and the first upper roof threshold can be set according to the actual application scenario. In one embodiment, the first lower roof threshold can be set in the range of -0.41 to -0.39, and the first upper roof threshold can be set in the range of 0.39 to 0.41. In one embodiment, the first lower roof threshold is set to -0.4 and the first upper roof threshold is set to 0.4.

[0068] In one embodiment, if the left side value of the vehicle is greater than -0.4 and less than 0.4, and the ratio of the intersection area to the vehicle is greater than 0.9, and the roof value of the vehicle is greater than -0.4 and less than 0.4, then the vehicle is in the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0069] In one embodiment, if the left side value of the vehicle is greater than the first lower left threshold and less than the first upper left threshold, and the ratio of the intersection area to the berth is greater than the second berth ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, then the vehicle is within the berth. The second berth ratio threshold can be set according to the actual application scenario. In one embodiment, the second berth ratio threshold can be set within the range of 0.69 to 0.71. In one embodiment, the second berth ratio threshold can be set to 0.7.

[0070] In one embodiment, if the left side value of the vehicle is greater than -0.4 and less than 0.4, and the ratio of the intersection area to the berth is greater than 0.7, and the roof value is greater than -0.4 and less than 0.4, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0071] In one embodiment, if the right side value of the vehicle is greater than the first lower right threshold and less than the first upper right threshold, and the ratio of the intersection area to the vehicle is greater than the second vehicle ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, then the vehicle is within the berth.

[0072] The first lower right threshold and the first upper right threshold can be set according to the actual application scenario. In one embodiment, the first lower right threshold can be set within the range of -0.41 to -0.39, and the first upper right threshold can be set within the range of 0.39 to 0.41. In one embodiment, the first lower right threshold is set to -0.4, and the first upper right threshold is set to 0.4. The first lower roof threshold and the first upper roof threshold can be set according to the actual application scenario. In one embodiment, the first lower roof threshold can be set within the range of -0.41 to -0.39, and the first upper roof threshold can be set within the range of 0.39 to 0.41. In one embodiment, the first lower roof threshold is set to -0.4, and the first upper roof threshold is set to 0.4.

[0073] In one embodiment, if the right side value of the vehicle is greater than -0.4 and less than 0.4, and the ratio of the intersection area to the vehicle is greater than 0.9, and the roof value is greater than -0.4 and less than 0.4, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0074] In one embodiment, if the right side value of the vehicle is greater than the first lower right threshold and less than the first upper right threshold, and the ratio of the intersection area to the berth is greater than the second berth ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, then the vehicle is within the berth.

[0075] In one embodiment, if the right side value of the vehicle is greater than -0.4 and less than 0.4, and the ratio of the intersection area to the berth is greater than 0.7, and the roof value is greater than -0.4 and less than 0.4, then the vehicle is within the berth.

[0076] In one embodiment, if the left side value of the vehicle is greater than the first upper left threshold and less than the second left threshold, and the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, and the ratio of the intersection area to the berth is greater than the second berth ratio threshold, and the vehicle is not a corner vehicle, then the vehicle is within the berth.

[0077] The second left threshold can be set according to the actual application scenario. In one embodiment, the second left threshold can be set within the range of 1.59 to 1.61. In one embodiment, the second left threshold is set to 1.6. The first lower bottom threshold and the first upper bottom threshold can be set according to the actual application scenario. In one embodiment, the first lower bottom threshold can be set within the range of 0.49 to 0.51. The first upper bottom threshold can be set within the range of 1.49 to 1.51. In one embodiment, the first lower bottom threshold is set to 0.5 and the first upper bottom threshold is set to 1.5. The vehicle not being a corner vehicle, i.e., a non-corner vehicle, can be understood as not meeting the conditions of step S07, that is, it is a non-corner vehicle.

[0078] In one embodiment, if the left side value of the vehicle is greater than 0.4 and less than 1.6, and the bottom value of the vehicle is greater than 0.5 and less than 1.5, and the ratio of the intersection area to the berth is greater than 0.7, and the vehicle is not a corner vehicle, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0079] In one embodiment, if the right side value of the vehicle is greater than the first upper right threshold and less than the second right threshold, and the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, and the ratio of the intersection area to the berth is greater than the second berth ratio threshold, and the vehicle is not a corner vehicle, then the vehicle is within the berth.

[0080] The second right threshold can be set according to the actual application scenario. In one embodiment, the second right threshold can be set within the range of 1.59 to 1.61. In one embodiment, the second right threshold is set to 1.6.

[0081] In one embodiment, if the right side value of the vehicle is greater than 0.4 and less than 1.6, and the bottom value of the vehicle is greater than 0.5 and less than 1.5, and the ratio of the intersection area to the berth is greater than 0.7, and the vehicle is not a corner vehicle, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0082] In one embodiment, if the left side value of the vehicle is greater than the third lower left threshold and less than the third upper left threshold, and the right side value of the vehicle is greater than the third lower right threshold and less than the third upper right threshold, and the bottom value of the vehicle is greater than the second lower bottom threshold and less than the second upper bottom threshold, then the vehicle is within the berth.

[0083] The third left lower threshold and the third left upper threshold can be set according to the actual application scenario. In one embodiment, the third left lower threshold is set to 0. The third left upper threshold can be set within the range of 0.99 to 1.01. In one embodiment, the third left upper threshold is set to 1. The third right lower threshold and the third right upper threshold can be set according to the actual application scenario. In one embodiment, the third right lower threshold is set to 0. The third right upper threshold can be set within the range of 0.99 to 1.01. In one embodiment, the third right upper threshold is set to 1. The second underbody lower threshold and the second underbody upper threshold can be set according to the actual application scenario. In one embodiment, the second underbody lower threshold can be set within the range of 0.29 to 0.31, and the second underbody upper threshold can be set within the range of 0.99 to 1.01. In one embodiment, the second underbody lower threshold is set to 0.3, and the second underbody upper threshold is set to 1.

[0084] In one embodiment, if the left value of the vehicle is greater than 0 and less than 1, and the right value of the vehicle is greater than 0 and less than 1, and the underbody value is greater than 0.3 and less than 1, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0085] In one embodiment, if the left value of the vehicle is greater than the third left lower threshold and less than the third left upper threshold, and the right value of the vehicle is greater than the third right lower threshold and less than the third right upper threshold, and the underbody value is greater than the third underbody threshold, and the vehicle is an occluded vehicle, then the vehicle is within the parking space. The third underbody threshold can be set according to the actual application scenario. In one embodiment, the third underbody threshold can be set within the range of -0.049 to -0.051. In one embodiment, the third underbody threshold is set to -0.05.

[0086] In one embodiment, if the left value of the vehicle is greater than 0 and less than 1, and the right value of the vehicle is greater than 0 and less than 1, and the underbody value is greater than -0.05, and the vehicle is an occluded vehicle, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0087] In one embodiment, if the intersection area ratio to the parking space is greater than the third ratio threshold to the parking space, and the vehicle height ratio to the parking space is greater than the first vehicle height threshold to the parking space, then the vehicle is within the parking space.

[0088] The third ratio threshold to the parking space can be set according to the actual application scenario. In one embodiment, the third ratio threshold to the parking space can be set within the range of 0.89 to 0.91. In one embodiment, the third ratio threshold to the parking space is set to 0.9. The first vehicle height threshold to the parking space can be set according to the actual application scenario. In one embodiment, the first vehicle height threshold to the parking space can be set within the range of 1.49 to 1.51. In one embodiment, the first vehicle height threshold to the parking space is set to 1.5. The vehicle is a large bus or a truck.

[0089] In one embodiment, if the ratio of the intersection area to the berth is greater than 0.9 and the vehicle-berth height ratio is greater than 1.5, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0090] In one embodiment, if the ratio of the intersection area to the berth is less than the fourth berth ratio threshold and the vehicle-berth height ratio is greater than the first vehicle-berth height threshold, then the vehicle is within the berth. The fourth berth ratio threshold can be set according to the actual application scenario. In one embodiment, the fourth berth ratio threshold can be set within the range of 0.59 to 0.61. In one embodiment, the fourth berth ratio threshold is set to 0.6. In one embodiment, if the ratio of the intersection area to the berth is less than 0.6 and the vehicle-berth height ratio is greater than 1.5, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0091] In one embodiment, if the ratio of the intersection area to the vehicle is less than the third vehicle ratio threshold, the ratio of the intersection area to the berth is greater than the fifth berth ratio threshold, the difference between the ordinate of the outer far-ground boundary of the circumscribed rectangle of the three-dimensional berth projection frame and the ordinate of the far-ground boundary of the vehicle detection frame is greater than the ordinate threshold, the vehicle-berth height ratio is greater than the second vehicle-berth height threshold, and the vehicle bottom value is greater than the fourth vehicle bottom threshold, then the vehicle is within the berth.

[0092] The third vehicle occupancy ratio threshold can be set according to the actual application scenario. In one embodiment, the third vehicle occupancy ratio threshold can be set between 0.74 and 0.76. In one embodiment, the third vehicle occupancy ratio threshold is set to 0.75. The fifth berth occupancy ratio threshold can be set according to the actual application scenario. In one embodiment, the fifth berth occupancy ratio threshold can be set between 0.74 and 0.76. In one embodiment, the fifth berth occupancy ratio threshold is set to 0.75. The outer far ground boundary of the circumscribed rectangle of the three-dimensional berth projection frame can be understood as the upper boundary of the circumscribed rectangle. The far ground boundary of the vehicle detection frame can be understood as the upper boundary of the vehicle detection frame. Subtract the ordinate of the far ground boundary of the vehicle detection frame from the ordinate of the outer far ground boundary of the circumscribed rectangle of the three-dimensional berth projection frame, and determine whether the difference is greater than the ordinate threshold. The ordinate threshold can be set according to the actual application scenario. The ordinate threshold can be set according to the actual application scenario. In one embodiment, the ordinate threshold can be set between 9.9 and 10.1. In one embodiment, the ordinate threshold is set to 10. The second vehicle-berth height threshold can be set according to the actual application scenario. In one embodiment, the second vehicle-berth height threshold is set between 1.19 and 1.21. In one embodiment, the second vehicle-berth height threshold is set to 1.2. The fourth vehicle bottom threshold can be set according to the actual application scenario. In one embodiment, the fourth vehicle bottom threshold is set between 0.79 and 0.81. In one embodiment, the fourth vehicle bottom threshold is set to 0.8. The vehicle is a medium-sized bus or truck.

[0093] In one embodiment, if the ratio of the intersection area to the vehicle is less than 0.75, the ratio of the intersection area to the berth is greater than 0.75, the difference between the ordinate of the outer far ground boundary of the circumscribed rectangle of the three-dimensional berth projection frame and the ordinate of the far ground boundary of the vehicle detection frame is greater than 10, the vehicle-berth height ratio is greater than 1.2, and the vehicle bottom value is greater than 0.8, then the vehicle is in the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0094] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the third vehicle occupancy ratio threshold, and the vehicle bottom value is greater than the fifth lower vehicle bottom threshold and less than the second upper vehicle bottom threshold, then the vehicle is in the berth. The fifth lower vehicle bottom threshold can be set according to the actual application scenario. In one embodiment, the fifth lower vehicle bottom threshold is set between 0.189 and 0.191. In one embodiment, the fifth lower vehicle bottom threshold is set to 0.19.

[0095] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.75, and the vehicle bottom value is greater than 0.19 and less than 1, then the vehicle is in the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0096] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the roof value is greater than the second roof lower threshold and less than the second roof upper threshold, and the ordinate of the far ground boundary of the vehicle detection frame is greater than the ordinate threshold, and the bottom value is less than the sixth bottom threshold, then the vehicle is within the parking space. The second roof lower threshold and the second roof upper threshold can be set according to the actual application scenario. In one embodiment, the second roof lower threshold is set to 0. The second roof upper threshold can be set according to the actual application scenario. In one embodiment, the second roof upper threshold is set within the range of 0.99 to 1.01. In one embodiment, the second roof upper threshold is set to 1. The ordinate of the far ground boundary of the vehicle detection frame can be understood as the ordinate of the upper boundary of the vehicle detection frame. The ordinate threshold is the same as the ordinate threshold in the above embodiment and can be set to 10. The sixth bottom threshold is set to 0.

[0097] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.75, and the roof value is greater than 0 and less than 1, and the ordinate of the far ground boundary of the vehicle detection frame is greater than 10, and the bottom value is less than 0, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0098] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the roof value is greater than the second roof lower threshold and less than the third roof threshold, and the bottom value is greater than the seventh bottom lower threshold and less than the seventh bottom upper threshold, then the vehicle is within the parking space.

[0099] The third roof threshold can be set according to the actual application scenario. In one embodiment, the third roof threshold can be set between 0.60 and 0.62. In one embodiment, the third roof threshold is set to 0.61. The seventh bottom lower threshold and the seventh bottom upper threshold can be set according to the actual application scenario. In one embodiment, the seventh bottom lower threshold can be set between 0.46 and 0.48. In one embodiment, the seventh bottom lower threshold is set to 0.47. In one embodiment, the seventh bottom upper threshold can be set between 1.27 and 1.29. In one embodiment, the seventh bottom upper threshold is set to 1.28.

[0100] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.75, and the roof value is greater than 0 and less than 0.61, and the bottom value is greater than 0.47 and less than 1.28, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0101] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the roof value is greater than the third roof threshold and less than or equal to the fourth roof threshold, and the bottom value is greater than the second upper bottom threshold and less than the eighth bottom threshold, and the absolute value of the slope of the ground road-side long boundary of the three-dimensional berth projection frame is greater than the first slope threshold, then the vehicle is within the berth.

[0102] The fourth roof threshold can be set according to the actual application scenario. In one embodiment, the fourth roof threshold can be set between 0.89 and 0.91. In one embodiment, the fourth roof threshold is set to 0.9. The eighth bottom threshold can be set according to the actual application scenario. In one embodiment, the eighth bottom threshold can be set between 1.09 and 1.11. In one embodiment, the eighth bottom threshold is set to 1.1. The first slope threshold can be set according to the actual application scenario. In one embodiment, the first slope threshold can be set between 6.9 and 7.1. In one embodiment, the first slope threshold is set to 7.

[0103] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.75, and the roof value is greater than 0.61 and less than or equal to 0.9, and the bottom value is greater than 1 and less than 1.1, and the absolute value of the slope of the ground road-side long boundary of the three-dimensional berth projection frame is greater than 7, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0104] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the absolute value of the slope of the ground road-side long boundary is less than the second slope threshold, and the roof value is greater than the fifth roof threshold and less than the second upper roof threshold, and the bottom value is greater than the sixth bottom threshold and less than the second upper bottom threshold, then the vehicle is within the berth.

[0105] The second slope threshold can be set according to the actual application scenario. In one embodiment, the second slope threshold can be set between 0.44 and 0.46. In one embodiment, the second slope threshold is set to 0.45. The fifth roof threshold can be set according to the actual application scenario. In one embodiment, the fifth roof threshold can be set between -0.11 and -0.09. In one embodiment, the fifth roof threshold is set to -0.1.

[0106] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.75, and the absolute value of the slope of the ground road-side long boundary is less than 0.45, and the roof value is greater than -0.1 and less than 1, and the bottom value is greater than 0 and less than 1, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0107] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the roof value is greater than the second lower roof threshold and less than the sixth roof threshold, and the bottom value is greater than the sixth bottom threshold and less than the second upper bottom threshold, then the vehicle is within the parking space. The sixth roof threshold can be set according to the actual application scenario. In one embodiment, the sixth roof threshold can be set between 0.77 and 0.79. In one embodiment, the sixth roof threshold is set to 0.78.

[0108] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.75, and the roof value is greater than 0 and less than 0.78, and the bottom value is greater than 0 and less than 1, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0109] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the fifth vehicle ratio threshold, and the ratio of the intersection area to the parking space is greater than the seventh parking space ratio threshold, and the bottom value is greater than the tenth bottom threshold and less than the fifth upper bottom threshold, then the vehicle is within the parking space. The fifth vehicle ratio threshold can be set according to the actual application scenario. In one embodiment, the fifth vehicle ratio threshold can be set between 0.64 and 0.66. In one embodiment, the fifth vehicle ratio threshold is set to 0.65. The seventh parking space ratio threshold can be set according to the actual application scenario. In one embodiment, the seventh parking space ratio threshold can be set between 0.64 and 0.66. In one embodiment, the seventh parking space ratio threshold is set to 0.65, which has the same value as the fifth vehicle ratio threshold. The tenth bottom threshold can be set according to the actual application scenario. In one embodiment, the tenth bottom threshold can be set between 0.49 and 0.51. In one embodiment, the tenth bottom threshold is set to 0.5.

[0110] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.65, and the ratio of the intersection area to the parking space is greater than 0.65, and the bottom value is greater than 0.5 and less than 1, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0111] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold, and the roof value is greater than the seventh lower roof threshold and less than the seventh upper roof threshold, then the vehicle is within the parking space. The fourth vehicle ratio threshold can be set according to the actual application scenario. In one embodiment, the fourth vehicle ratio threshold can be set between 0.84 and 0.86. In one embodiment, the fourth vehicle ratio threshold is set to 0.85. The seventh lower roof threshold and the seventh upper roof threshold can be set according to the actual application scenario. In one embodiment, the seventh lower roof threshold can be set between -0.21 and -0.19. In one embodiment, the seventh lower roof threshold is set to -0.2. In one embodiment, the seventh upper roof threshold can be set between 0.19 and 0.21. In one embodiment, the seventh upper roof threshold is set to 0.2.

[0112] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.85, and the roof value is greater than -0.2 and less than 0.2, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0113] In one embodiment, if the ratio of the intersection area to the parking space is greater than the sixth parking space ratio threshold, and the roof value is greater than the seventh lower roof threshold and less than the seventh upper roof threshold, then the vehicle is within the parking space. The sixth parking space ratio threshold can be set according to the actual application scenario. In one embodiment, the sixth parking space ratio threshold can be set between 0.84 and 0.86. In one embodiment, the sixth parking space ratio threshold is set to 0.85.

[0114] In one embodiment, if the ratio of the intersection area to the parking space is greater than 0.85, and the roof value is greater than -0.2 and less than 0.2, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0115] In one embodiment, if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold, and the bottom value is greater than the ninth lower bottom threshold and less than the ninth upper bottom threshold, then the vehicle is within the parking space. The ninth lower bottom threshold and the ninth upper bottom threshold can be set according to the actual application scenario. In one embodiment, the ninth lower bottom threshold can be set in the range between 0.79 and 0.81. In one embodiment, the ninth lower bottom threshold is set to 0.8. In one embodiment, the ninth upper bottom threshold can be set in the range between 1.19 and 1.21. In one embodiment, the ninth upper bottom threshold is set to 1.2.

[0116] In one embodiment, if the ratio of the intersection area to the vehicle is greater than 0.85, and the bottom value is greater than 0.8 and less than 1.2, then the vehicle is within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0117] In one embodiment, if the ratio of the intersection area to the berth is greater than the sixth berth ratio threshold, and the vehicle bottom value is greater than the ninth upper vehicle bottom threshold and less than the ninth lower vehicle bottom threshold, then the vehicle is within the berth.

[0118] In one embodiment, if the left side value of the vehicle is greater than the fourth left threshold and less than the third upper left threshold, and the right side value of the vehicle is greater than the fourth right threshold and less than the third upper right threshold, and the roof value of the vehicle is greater than the eighth roof threshold and less than the second upper roof threshold, and the vehicle bottom value is greater than the third vehicle bottom threshold and less than the second upper vehicle bottom threshold, then the vehicle is within the berth.

[0119] The fourth left threshold can be set according to the actual application scenario. In one embodiment, the fourth left threshold can be set between -0.06 and -0.04. In one embodiment, the fourth left threshold is set to -0.05. The fourth right threshold can be set according to the actual application scenario. In one embodiment, the fourth right threshold can be set between -0.06 and -0.04. In one embodiment, the fourth right threshold is set to -0.05. The eighth roof threshold can be set according to the actual application scenario. In one embodiment, the eighth roof threshold can be set between -0.06 and -0.04. In one embodiment, the eighth roof threshold is set to -0.05. The third vehicle bottom threshold can be set according to the actual application scenario. In one embodiment, the third vehicle bottom threshold can be set between -0.06 and -0.04. In one embodiment, the third vehicle bottom threshold is set to -0.05.

[0120] In one embodiment, if the left side value of the vehicle is greater than -0.05 and less than 1, and the right side value of the vehicle is greater than -0.05 and less than 1, and the roof value of the vehicle is greater than -0.05 and less than 1, and the vehicle bottom value of the vehicle is greater than -0.05 and less than 1, then the vehicle is within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0121] In one embodiment, if the vehicle detection frame has no intersection with other vehicle detection frames, which can also be understood as having no intersection relationship, then it is considered that the vehicle is not blocked, and if the vehicle bottom value is less than the eleventh vehicle bottom threshold, then the vehicle is not within the berth. The eleventh vehicle bottom threshold can be set according to the actual application scenario. In one embodiment, the eleventh vehicle bottom threshold can be set between 0.19 and 0.21. In one embodiment, the eleventh vehicle bottom threshold is set to 0.2.

[0122] In one embodiment, if the left - hand side value of the vehicle is greater than the fifth left - hand side threshold, and the right - hand side value of the vehicle is greater than the fifth right - hand side threshold, and the roof value is greater than the ninth roof threshold, then the vehicle is not within the parking space. The fifth left - hand side threshold can be set according to the actual application scenario. In one embodiment, the fifth left - hand side threshold can be set between 1.29 and 1.31. In one embodiment, the fifth left - hand side threshold is set to 1.3. The fifth right - hand side threshold can be set according to the actual application scenario. In one embodiment, the fifth right - hand side threshold can be set between 0.94 and 0.96. In one embodiment, the fifth right - hand side threshold is set to 0.95. The ninth roof threshold can be set according to the actual application scenario. In one embodiment, the ninth roof threshold can be set between 0.79 and 0.81. In one embodiment, the ninth roof threshold is set to 0.8.

[0123] In one embodiment, if the left - hand side value of the vehicle is greater than 1.3, and the right - hand side value of the vehicle is greater than 0.95, and the roof value is greater than 0.8, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0124] In one embodiment, if the left - hand side value of the vehicle is greater than the third upper left - hand side threshold, and the right - hand side value of the vehicle is greater than the third upper right - hand side threshold, and the bottom value of the vehicle is greater than the first upper bottom threshold, then the vehicle is not within the parking space.

[0125] In one embodiment, if the left - hand side value of the vehicle is greater than 1, and the right - hand side value of the vehicle is greater than 1, and the bottom value of the vehicle is greater than 1.5, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0126] In one embodiment, if the left - hand side value of the vehicle is less than the sixth left - hand side threshold, and the right - hand side value of the vehicle is less than the sixth right - hand side threshold, then the vehicle is not within the parking space. The sixth left - hand side threshold can be set according to the actual application scenario. In one embodiment, the sixth left - hand side threshold can be set between - 2.1 and - 1.9. In one embodiment, the sixth left - hand side threshold is set to - 2. The sixth right - hand side threshold can be set according to the actual application scenario. In one embodiment, the sixth right - hand side threshold can be set between - 1.1 and - 0.9. In one embodiment, the sixth right - hand side threshold is set to - 1.

[0127] In one embodiment, if the left - hand side value of the vehicle is less than - 2, and the right - hand side value of the vehicle is less than - 1, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0128] In one embodiment, if the left - hand side value of the vehicle is greater than the third upper left - hand side threshold, and the right - hand side value of the vehicle is greater than the third upper right - hand side threshold, then the vehicle is not within the parking space.

[0129] In one embodiment, if the value on the left side of the vehicle is greater than 1 and the value on the right side of the vehicle is greater than 1, then the vehicle is not in the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0130] In one embodiment, the ratio of the ground boundary to the road-side boundary of the vehicle detection frame can be understood as the ratio of the lower boundary to the left boundary being greater than the first width-to-height ratio threshold, and the vehicle bottom value being less than the sixth vehicle bottom threshold, then the vehicle is not in the parking space. The first width-to-height ratio threshold can be set according to the actual application scenario. In one embodiment, the first width-to-height ratio threshold can be set between 0.79 and 0.81. In one embodiment, the first width-to-height ratio threshold is set to 0.8.

[0131] In one embodiment, the ratio of the ground boundary to the road-side boundary of the vehicle detection frame can be understood as the ratio of the lower boundary to the left boundary being greater than 0.8, and the vehicle bottom value being less than 0, then the vehicle is not in the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0132] In one embodiment, if the vehicle is a non-corner vehicle, and the ratio of the ground boundary to the road-side boundary of the vehicle detection frame is less than the second width-to-height ratio threshold, and the intersection area ratio to the parking space is less than the second parking space ratio threshold, then the vehicle is not in the parking space. The second width-to-height ratio threshold can be set according to the actual application scenario. In one embodiment, the second width-to-height ratio threshold can be set between 0.9 and 1.1. In one embodiment, the second width-to-height ratio threshold is set to 1.

[0133] In one embodiment, if the vehicle is a non-corner vehicle, and the ratio of the ground boundary to the road-side boundary of the vehicle detection frame is less than 1, and the intersection area ratio to the parking space is less than 0.7, then the vehicle is not in the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0134] In one embodiment, if the roof value is greater than the tenth roof threshold and the vehicle bottom value is greater than the seventh vehicle bottom threshold, then the vehicle is not in the parking space. The tenth roof threshold can be set according to the actual application scenario. In one embodiment, the tenth roof threshold can be set between 0.94 and 0.96. In one embodiment, the tenth roof threshold is set to 0.95. The seventh vehicle bottom threshold can be set according to the actual application scenario. In one embodiment, the seventh vehicle bottom threshold can be set between 1.79 and 1.81. In one embodiment, the seventh vehicle bottom threshold is set to 1.8.

[0135] In one embodiment, if the roof value is greater than 0.95 and the vehicle bottom value is greater than 1.8, then the vehicle is not in the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0136] In one embodiment, if the left side value of the vehicle is greater than the seventh lower left threshold and less than the seventh upper left threshold, and the bottom value of the vehicle is greater than the seventh bottom threshold, then the vehicle is not within the parking space. The seventh lower left threshold and the seventh upper left threshold can be set according to the actual application scenario. In one embodiment, the seventh lower left threshold can be set between 0.79 and 0.81. In one embodiment, the seventh lower left threshold is set to 0.8. The seventh upper left threshold can be set between 1.19 and 1.21. In one embodiment, the seventh upper left threshold is set to 1.2. In one embodiment, if the left side value of the vehicle is greater than 0.8 and less than 1.2, and the bottom value of the vehicle is greater than 1.8, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0137] In one embodiment, if the roof value is less than the second lower roof threshold, the bottom value of the vehicle is less than the sixth bottom threshold, the left side value of the vehicle is less than the third lower left threshold, and the right side value of the vehicle is less than the third lower right threshold, then the vehicle is not within the parking space.

[0138] In one embodiment, if the roof value is less than 0, the bottom value of the vehicle is less than 0, the left side value of the vehicle is less than 0, and the right side value of the vehicle is less than 0, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0139] In one embodiment, if the roof value is greater than the eleventh roof threshold, then the vehicle is not within the parking space. The eleventh roof threshold can be set according to the actual application scenario. In one embodiment, the eleventh roof threshold is set between 0.85 and 0.87. In one embodiment, the eleventh roof threshold is set to 0.86. In one embodiment, if the roof value is greater than 0.86, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0140] In one embodiment, if the absolute value of the slope of the ground road-length boundary of the three-dimensional parking space projection frame is greater than the third slope threshold, and the roof value is greater than the twelfth roof threshold, then the vehicle is not within the parking space. The third slope threshold can be set according to the actual application scenario. In one embodiment, the third slope threshold is set between 8.9 and 9.1. In one embodiment, the third slope threshold is set to 9. The twelfth roof threshold can be set according to the actual application scenario. In one embodiment, the twelfth roof threshold is set between 0.59 and 0.61. In one embodiment, the twelfth roof threshold is set to 0.6. In one embodiment, if the absolute value of the slope of the ground road-length boundary of the three-dimensional parking space projection frame is greater than 9, and the roof value is greater than 0.6, then the vehicle is not within the parking space, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0141] In one embodiment, if the ratio of the intersection area to the vehicle is less than the sixth vehicle ratio threshold and the ratio of the intersection area to the berth is less than the eighth berth ratio threshold, then the vehicle is not within the berth. The sixth vehicle ratio threshold can be set according to the actual application scenario. In one embodiment, the sixth vehicle ratio threshold can be set between 0.19 and 0.21. In one embodiment, the sixth vehicle ratio threshold is set to 0.2. The eighth berth ratio threshold can be set according to the actual application scenario. In one embodiment, the eighth berth ratio threshold can be set between 0.19 and 0.21. In one embodiment, the eighth berth ratio threshold is set to 0.2. In one embodiment, if the ratio of the intersection area to the vehicle is less than 0.2 and the ratio of the intersection area to the berth is less than 0.2, then the vehicle is not within the berth, and the parking state of the vehicle corresponding to the vehicle detection frame is obtained.

[0142] Please refer to Figure 4 , this application provides a berth status analysis system based on cloud-edge fusion 01, including a camera data acquisition module 10, a historical cache judgment module 20, a calculation module 30, a vehicle detection frame cache judgment module 40, a call cache module 50, a vehicle parking calculation module 60, a vehicle-in-berth judgment module 70, a first vehicle-in-berth judgment result module 80, a second vehicle-in-berth judgment result module 90, and a cache module 100. The camera data acquisition module 10 is used to acquire the vehicle detection frame of each vehicle corresponding to the panoramic view of the berth of each camera and the berth calibration coordinates of the berth number corresponding to each camera. The historical cache judgment module 20 is used to judge whether there is corresponding historical berth cache information for the berth number. The calculation module 30 is used to, if there is no corresponding historical berth cache information for the berth number, transform the berth calibration coordinates according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and calculate the parking state of each vehicle based on the three-dimensional berth projection frame and each vehicle detection frame. The vehicle detection frame cache judgment module 40 is used to, if there is corresponding historical berth cache information for the berth number, judge whether each vehicle detection frame exists in the historical berth cache information. The call cache module 50 is used to, if so, call the parking state of the vehicle corresponding to each vehicle detection frame according to the historical berth cache information. The vehicle parking calculation module 60 is used to, if not, calculate the parking state of the vehicle corresponding to the vehicle detection frame based on the three-dimensional berth projection frame and the vehicle detection frame. The vehicle-in-berth judgment module 70 is used to judge whether there is a vehicle within the berth corresponding to the berth number according to the parking state of each vehicle. The first vehicle-in-berth judgment result module 80 is used to, if so, there is a vehicle within the berth corresponding to the berth number. The second vehicle-in-berth judgment result module 90 is used to, if not, there is no vehicle within the berth corresponding to the berth number. The cache module 100 is used to form new berth cache information from the vehicle parking calculation results corresponding to the berth number for the next berth status determination.

[0143] In this embodiment, the relevant description of the camera data acquisition module 10 can refer to the relevant description of S10 in the above embodiment. The relevant description of the historical cache judgment module 20 can refer to the relevant description of S20 in the above embodiment. The relevant description of the calculation module 30 can refer to the relevant description of S30 in the above embodiment. The relevant description of the vehicle detection box cache judgment module 40 can refer to the relevant description of S40 in the above embodiment. The relevant description of the call cache module 50 can refer to the relevant description of S50 in the above embodiment. The relevant description of the vehicle parking calculation module 60 can refer to the relevant description of S60 in the above embodiment. The relevant description of the in-berth vehicle judgment module 70 can refer to the relevant description of S70 in the above embodiment. The relevant description of the in-berth vehicle judgment result module one 80 can refer to the relevant description of S80 in the above embodiment. The relevant description of the in-berth vehicle judgment result module two 90 can refer to the relevant description of S90 in the above embodiment. The relevant description of the cache module 100 can refer to the relevant description of S100 in the above embodiment.

[0144] In one embodiment, the calculation module includes an intersection area to vehicle ratio module, an intersection area to berth ratio module, a ground intersection point acquisition module, a vehicle bottom value calculation module, an air intersection point acquisition module, a vehicle top value calculation module, a vehicle left side value calculation module, a vehicle right side value calculation module, a corner vehicle judgment module, an occluded vehicle judgment module, a vehicle-berth height ratio calculation module, and a first judgment module. The intersection area to vehicle ratio module is used to calculate the intersection area between the circumscribed rectangle of the three-dimensional berth projection box and each vehicle detection box, and obtain the intersection area to vehicle ratio based on the intersection area and the area of the vehicle detection box. The intersection area to berth ratio module is used to obtain the intersection area to berth ratio based on the intersection area and the area of the circumscribed rectangle of the three-dimensional berth projection box. The ground intersection point acquisition module is used to draw a parallel line parallel to the ground lower width boundary of the three-dimensional berth projection box according to the midpoint of the ground boundary of the vehicle detection box, and obtain the ground intersection point of the parallel line and the ground road-facing long boundary of the three-dimensional berth projection box. The vehicle bottom value calculation module is used to calculate the ground intersection top difference from the abscissa of the ground intersection point to the abscissa of the vertex of the ground road-facing long boundary and the ground bottom top difference from the abscissa of the bottom point of the ground road-facing long boundary to the abscissa of the vertex of the ground road-facing long boundary, and divide the ground intersection top difference by the ground bottom top difference to obtain the vehicle bottom value. The air intersection point acquisition module is used to draw a parallel line parallel to the air upper width boundary of the three-dimensional berth projection box according to the midpoint of the far ground boundary of the vehicle detection box, and obtain the air intersection point of the parallel line and the air road-facing long boundary of the three-dimensional berth projection box.

[0145] The roof value calculation module is used to calculate the aerial intersection top difference from the abscissa of the aerial intersection point to the abscissa of the vertex of the aerial road length boundary, and the aerial bottom-top difference from the abscissa of the bottom point of the aerial road length boundary to the abscissa of the vertex of the aerial road length boundary, and divide the aerial intersection top difference by the aerial bottom-top difference to obtain the roof value. The vehicle left side value calculation module is used to calculate the first abscissa difference between the abscissa of the road-side boundary of the vehicle detection frame and the abscissa of the vertex of the ground road length boundary, calculate the second abscissa difference between the abscissa of the bottom point of the ground road length boundary and the abscissa of the vertex of the ground road length boundary, and divide the first abscissa difference by the second abscissa difference to obtain the vehicle left side value. The vehicle right side value calculation module is used to calculate the third abscissa difference between the abscissa of the far-road boundary of the vehicle detection frame and the abscissa of the vertex of the ground far-road length boundary of the three-dimensional berth projection frame, calculate the fourth abscissa difference between the abscissa of the bottom point of the ground far-road length boundary and the abscissa of the vertex of the ground far-road length boundary, and divide the third abscissa difference by the fourth abscissa difference to obtain the vehicle right side value. The corner vehicle judgment module is used to calculate the boundary distances between the four boundaries of the vehicle detection frame and the four boundaries of the berth panoramic view respectively. If the boundary distance is less than the pixel threshold, the vehicle is a corner vehicle. The occluded vehicle judgment module is used to determine that the vehicle is an occluded vehicle if the vehicle detection frame has an intersection with the surrounding vehicle detection frames and the ground boundary of the vehicle detection frame is above the ground boundary of the intersection vehicle detection frames. The vehicle-to-berth height ratio calculation module is used to divide the height of the vehicle detection frame by the height of the circumscribed rectangle of the three-dimensional berth projection frame to obtain the vehicle-to-berth height ratio. The first judgment module is used to determine that the vehicle is not in the berth if the intersection area ratio to the vehicle is less than the first vehicle ratio threshold and the intersection area ratio to the berth is less than the first berth ratio threshold.

[0146] In this embodiment, the relevant descriptions of the intersection area ratio to vehicle module, the intersection area ratio to berth module, the ground intersection point acquisition module, the vehicle bottom value calculation module, the aerial intersection point acquisition module, the roof value calculation module, the vehicle left side value calculation module, the vehicle right side value calculation module, the corner vehicle judgment module, the occluded vehicle judgment module, the vehicle-to-berth height ratio calculation module, and the first judgment module can refer to the relevant descriptions of the method in the above embodiment.

[0147] In one embodiment, the calculation module further includes a second judgment module, a third judgment module, a fourth judgment module, and a fifth judgment module. The second judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the first lower left threshold and less than the first upper left threshold, and the intersection area ratio to the vehicle is greater than the second vehicle ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold. The third judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the first lower left threshold and less than the first upper left threshold, and the intersection area ratio to the parking space is greater than the second parking space ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold. The fourth judgment module is used to determine that the vehicle is in the parking space if the right side value of the vehicle is greater than the first lower right threshold and less than the first upper right threshold, and the intersection area ratio to the vehicle is greater than the second vehicle ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold. The fifth judgment module is used to determine that the vehicle is in the parking space if the right side value of the vehicle is greater than the first lower right threshold and less than the first upper right threshold, and the intersection area ratio to the parking space is greater than the second parking space ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold. In this embodiment, the relevant descriptions of the second judgment module, the third judgment module, the fourth judgment module, and the fifth judgment module can refer to the relevant descriptions of the method in the above embodiment.

[0148] In one embodiment, the calculation module further includes a sixth judgment module and a seventh judgment module. The sixth judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the first upper left threshold and less than the second left threshold, and the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, and the intersection area ratio to the parking space is greater than the second parking space ratio threshold, and the vehicle is not a corner vehicle. The seventh judgment module is used to determine that the vehicle is in the parking space if the right side value of the vehicle is greater than the first upper right threshold and less than the second right threshold, and the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, and the intersection area ratio to the parking space is greater than the second parking space ratio threshold, and the vehicle is not a corner vehicle. In this embodiment, the relevant descriptions of the sixth judgment module and the seventh judgment module can refer to the relevant descriptions of the method in the above embodiment.

[0149] In one embodiment, the calculation module further includes an eighth determination module and a ninth determination module. The eighth determination module is configured to determine that the vehicle is within the parking space if the left-side value of the vehicle is greater than the third lower left-side threshold and less than the third upper left-side threshold, the right-side value of the vehicle is greater than the third lower right-side threshold and less than the third upper right-side threshold, and the bottom value of the vehicle is greater than the second lower bottom threshold and less than the second upper bottom threshold. The ninth determination module is configured to determine that the vehicle is within the parking space if the left-side value of the vehicle is greater than the third lower left-side threshold and less than the third upper left-side threshold, the right-side value of the vehicle is greater than the third lower right-side threshold and less than the third upper right-side threshold, the bottom value of the vehicle is greater than the third bottom threshold, and the vehicle is an occluded vehicle. In this embodiment, for the relevant descriptions of the eighth determination module and the ninth determination module, reference may be made to the relevant descriptions of the method in the above embodiment.

[0150] In one embodiment, the calculation module further includes a tenth determination module and an eleventh determination module. The tenth determination module is configured to determine that the vehicle is within the parking space if the ratio of the intersection area to the parking space is greater than the third ratio-to-parking-space threshold and the vehicle-parking-height ratio is greater than the first vehicle-parking-height threshold. The eleventh determination module is configured to determine that the vehicle is within the parking space if the ratio of the intersection area to the parking space is less than the fourth ratio-to-parking-space threshold and the vehicle-parking-height ratio is greater than the first vehicle-parking-height threshold. In this embodiment, for the relevant descriptions of the tenth determination module and the eleventh determination module, reference may be made to the relevant descriptions of the method in the above embodiment.

[0151] In one embodiment, the calculation module further includes a twelfth determination module, a thirteenth determination module, a fourteenth determination module, a fifteenth determination module, a sixteenth determination module, a seventeenth determination module, an eighteenth determination module, and a nineteenth determination module. The twelfth determination module is configured to determine that the vehicle is within the parking space if the ratio of the intersection area to the vehicle is less than the third ratio-to-vehicle threshold, the ratio of the intersection area to the parking space is greater than the fifth ratio-to-parking-space threshold, the difference between the ordinate of the outer far-ground boundary of the circumscribed rectangle of the three-dimensional parking-space projection frame and the ordinate of the far-ground boundary of the vehicle detection frame is greater than the ordinate threshold, the vehicle-parking-height ratio is greater than the second vehicle-parking-height threshold, and the bottom value of the vehicle is greater than the fourth bottom threshold. The thirteenth determination module is configured to determine that the vehicle is within the parking space if the ratio of the intersection area to the vehicle is greater than the third ratio-to-vehicle threshold and the bottom value of the vehicle is greater than the fifth lower bottom threshold and less than the second upper bottom threshold. The fourteenth determination module is configured to determine that the vehicle is within the parking space if the ratio of the intersection area to the vehicle is greater than the third ratio-to-vehicle threshold, the roof value of the vehicle is greater than the second lower roof threshold and less than the second upper roof threshold, the ordinate of the far-ground boundary of the vehicle detection frame is greater than the ordinate threshold, and the bottom value of the vehicle is less than the sixth bottom threshold. The fifteenth determination module is configured to determine that the vehicle is within the parking space if the ratio of the intersection area to the vehicle is greater than the third ratio-to-vehicle threshold, the roof value of the vehicle is greater than the second lower roof threshold and less than the third upper roof threshold, and the bottom value of the vehicle is greater than the seventh lower bottom threshold and less than the seventh upper bottom threshold.

[0152] The sixteenth determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the roof value is greater than the third roof threshold and less than or equal to the fourth roof threshold, the bottom value is greater than the second bottom upper threshold and less than the eighth bottom threshold, and the absolute value of the slope of the ground road length boundary of the three-dimensional berth projection frame is greater than the first slope threshold. The seventeenth determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the absolute value of the slope of the ground road length boundary is less than the second slope threshold, the roof value is greater than the fifth roof threshold and less than the second roof upper threshold, and the bottom value is greater than the sixth bottom threshold and less than the second bottom upper threshold. The eighteenth determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the roof value is greater than the second roof lower threshold and less than the sixth roof threshold, and the bottom value is greater than the sixth bottom threshold and less than the second bottom upper threshold. The nineteenth determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the fifth vehicle ratio threshold, the ratio of the intersection area to the berth is greater than the seventh berth ratio threshold, and the bottom value is greater than the tenth bottom threshold and less than the fifth bottom upper threshold.

[0153] In this embodiment, the relevant descriptions of the twelfth determination module, the thirteenth determination module, the fourteenth determination module, the fifteenth determination module, the sixteenth determination module, the seventeenth determination module, the eighteenth determination module, and the nineteenth determination module can refer to the description of the method in the above embodiment.

[0154] In one embodiment, the calculation module further includes a twentieth determination module, a twenty-first determination module, a twenty-second determination module, and a twenty-third determination module. The twentieth determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold and the roof value is greater than the seventh roof lower threshold and less than the seventh roof upper threshold. The twenty-first determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the berth is greater than the sixth berth ratio threshold and the roof value is greater than the seventh roof lower threshold and less than the seventh roof upper threshold. The twenty-second determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold and the bottom value is greater than the ninth bottom lower threshold and less than the ninth bottom upper threshold. The twenty-third determination module is used to determine that the vehicle is within the berth if the ratio of the intersection area to the berth is greater than the sixth berth ratio threshold and the bottom value is greater than the ninth bottom upper threshold and less than the ninth bottom lower threshold. In this embodiment, the relevant descriptions of the twentieth determination module, the twenty-first determination module, the twenty-second determination module, and the twenty-third determination module can refer to the description of the method in the above embodiment.

[0155] In one embodiment, the calculation module further includes a twenty-fourth determination module. The twenty-fourth determination module is configured to determine that the vehicle is within the parking space if the left-side value of the vehicle is greater than the fourth left-side threshold and less than the third upper left-side threshold, and the right-side value of the vehicle is greater than the fourth right-side threshold and less than the third upper right-side threshold, and the roof value is greater than the eighth roof threshold and less than the second upper roof threshold, and the bottom value of the vehicle is greater than the third bottom threshold and less than the second upper bottom threshold. In this embodiment, for the related description of the twenty-fourth determination module, reference may be made to the description of the method in the above embodiment.

[0156] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present application. It should be understood that the above description is only the specific embodiments of the present application and is not used to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A berth status analysis method based on cloud-edge integration, characterized in that, Including: Obtaining the vehicle detection frame of each vehicle corresponding to the panoramic view of the berth of each camera and the berth calibration coordinates of the berth number corresponding to each camera; Judging whether there is corresponding historical berth cache information for the berth number; If there is no corresponding historical berth cache information for the berth number, transform the berth calibration coordinates according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and calculate the parking state of each vehicle according to the three-dimensional berth projection frame and each vehicle detection frame; If there is corresponding historical berth cache information for the berth number, judge whether each vehicle detection frame exists in the historical berth cache information; If so, call the parking state of the vehicle corresponding to each vehicle detection frame according to the historical berth cache information; If not, calculate the parking state of the vehicle corresponding to the vehicle detection frame according to the three-dimensional berth projection frame and the vehicle detection frame; Judge whether there is a vehicle in the berth corresponding to the berth number according to the parking state of each vehicle; If so, there is a vehicle in the berth corresponding to the berth number; If not, there is no vehicle in the berth corresponding to the berth number; Form new berth cache information from the vehicle parking calculation results corresponding to the berth number for the next berth state determination; The step that if there is no corresponding historical berth cache information for the berth number, transform the berth calibration coordinates according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and calculate the parking state of each vehicle according to the three-dimensional berth projection frame and each vehicle detection frame includes: Calculating the intersection area between the circumscribed rectangle of the three-dimensional berth projection frame and each vehicle detection frame, and obtaining the ratio of the intersection area to the vehicle according to the intersection area and the area of the vehicle detection frame; Obtaining the ratio of the intersection area to the berth according to the intersection area and the area of the circumscribed rectangle of the three-dimensional berth projection frame; Drawing a parallel line parallel to the ground lower width boundary of the three-dimensional berth projection frame according to the midpoint of the ground boundary of the vehicle detection frame, and obtaining the ground intersection point of the parallel line and the ground road-facing long boundary of the three-dimensional berth projection frame; Calculating the ground intersection difference from the abscissa of the ground intersection point to the abscissa of the vertex of the ground road-facing long boundary and the ground bottom-top difference from the abscissa of the bottom point of the ground road-facing long boundary to the abscissa of the vertex of the ground road-facing long boundary, and dividing the ground intersection difference by the ground bottom-top difference to obtain the vehicle bottom value; Drawing a parallel line parallel to the air upper width boundary of the three-dimensional berth projection frame according to the midpoint of the far ground boundary of the vehicle detection frame, and obtaining the air intersection point of the parallel line and the air road-facing long boundary of the three-dimensional berth projection frame; Calculating the air intersection difference from the abscissa of the air intersection point to the abscissa of the vertex of the air road-facing long boundary and the air bottom-top difference from the abscissa of the bottom point of the air road-facing long boundary to the abscissa of the vertex of the air road-facing long boundary, and dividing the air intersection difference by the air bottom-top difference to obtain the vehicle top value; Calculate the first horizontal coordinate difference between the abscissa of the roadside boundary of the vehicle detection frame and the abscissa of the vertex of the long roadside boundary of the ground, calculate the second horizontal coordinate difference between the abscissa of the bottom point of the long roadside boundary of the ground and the abscissa of the vertex of the long roadside boundary of the ground, and divide the first horizontal coordinate difference by the second horizontal coordinate difference to obtain the left vehicle value; Calculate the third horizontal coordinate difference between the abscissa of the far-road boundary of the vehicle detection frame and the abscissa of the vertex of the long far-road boundary of the ground of the three-dimensional berth projection frame, calculate the fourth horizontal coordinate difference between the abscissa of the bottom point of the long far-road boundary of the ground and the abscissa of the vertex of the long far-road boundary of the ground, and divide the third horizontal coordinate difference by the fourth horizontal coordinate difference to obtain the right vehicle value; Calculate the boundary distances between the four boundaries of the vehicle detection frame and the four boundaries of the berth panoramic view respectively. If the boundary distance is less than the pixel threshold, the vehicle is a corner vehicle; If the vehicle detection frame has an intersection with the surrounding vehicle detection frames, and the ground boundary of the vehicle detection frame is above the ground boundary of the intersection vehicle detection frame, the vehicle is an occluded vehicle; Divide the height of the vehicle detection frame by the height of the circumscribed rectangle of the three-dimensional berth projection frame to obtain the vehicle-berth height ratio; If the intersection area ratio to the vehicle is less than the first vehicle ratio threshold, and the intersection area ratio to the berth is less than the first berth ratio threshold, the vehicle is not in the berth.

2. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, transform the berth calibration coordinates according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and calculate the parking state of each vehicle according to the three-dimensional berth projection frame and each vehicle detection frame. It also includes: If the left vehicle value is greater than the first lower left threshold and less than the first upper left threshold, and the intersection area ratio to the vehicle is greater than the second vehicle ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, the vehicle is in the berth; Or, if the left vehicle value is greater than the first lower left threshold and less than the first upper left threshold, and the intersection area ratio to the berth is greater than the second berth ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, the vehicle is in the berth; Or, if the right vehicle value is greater than the first lower right threshold and less than the first upper right threshold, and the intersection area ratio to the vehicle is greater than the second vehicle ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, the vehicle is in the berth; Or, if the right vehicle value is greater than the first lower right threshold and less than the first upper right threshold, and the intersection area ratio to the berth is greater than the second berth ratio threshold, and the roof value is greater than the first lower roof threshold and less than the first upper roof threshold, the vehicle is in the berth.

3. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, the calibration coordinates of the berth are transformed according to the three-dimensional projection algorithm to obtain a three-dimensional berth projection box corresponding to the berth number, and the docking state of each vehicle is calculated based on the three-dimensional berth projection box and each vehicle detection box. It further includes: If the left value of the vehicle is greater than the first upper left threshold and less than the second left threshold, and the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, and the ratio of the intersection area to the berth is greater than the second berth ratio threshold, and the vehicle is not the corner vehicle, then the vehicle is within the berth; Or, if the right value of the vehicle is greater than the first upper right threshold and less than the second right threshold, and the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, and the ratio of the intersection area to the berth is greater than the second berth ratio threshold, and the vehicle is not the corner vehicle, then the vehicle is within the berth.

4. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, the calibration coordinates of the berth are transformed according to the three-dimensional projection algorithm to obtain a three-dimensional berth projection box corresponding to the berth number, and the docking state of each vehicle is calculated based on the three-dimensional berth projection box and each vehicle detection box. It further includes: If the left value of the vehicle is greater than the third lower left threshold and less than the third upper left threshold, and the right value of the vehicle is greater than the third lower right threshold and less than the third upper right threshold, and the bottom value of the vehicle is greater than the second lower bottom threshold and less than the second upper bottom threshold, then the vehicle is within the berth; Or, the left value of the vehicle is greater than the third lower left threshold and less than the third upper left threshold, and the right value of the vehicle is greater than the third lower right threshold and less than the third upper right threshold, and the bottom value of the vehicle is greater than the third bottom threshold, and the vehicle is the occluded vehicle, then the vehicle is within the berth.

5. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, the calibration coordinates of the berth are transformed according to the three-dimensional projection algorithm to obtain a three-dimensional berth projection box corresponding to the berth number, and the docking state of each vehicle is calculated based on the three-dimensional berth projection box and each vehicle detection box. It further includes: If the ratio of the intersection area to the berth is greater than the third berth ratio threshold, and the vehicle-berth height ratio is greater than the first vehicle-berth height threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the berth is less than the fourth berth ratio threshold, and the vehicle-berth height ratio is greater than the first vehicle-berth height threshold, then the vehicle is within the berth.

6. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, the calibration coordinates of the berth are transformed according to the three-dimensional projection algorithm to obtain a three-dimensional berth projection box corresponding to the berth number, and the docking state of each vehicle is calculated based on the three-dimensional berth projection box and each vehicle detection box. It further includes: If the ratio of the intersection area to the vehicle is less than the third vehicle ratio threshold, and the ratio of the intersection area to the berth is greater than the fifth berth ratio threshold, and the ordinate of the outer far-ground boundary of the circumscribed rectangle of the three-dimensional berth projection frame is greater than the ordinate threshold of the far-ground boundary of the vehicle detection frame, and the vehicle-berth height ratio is greater than the second vehicle-berth height threshold, and the vehicle bottom value is greater than the fourth vehicle bottom threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the vehicle bottom value is greater than the fifth lower vehicle bottom threshold and less than the second upper vehicle bottom threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the vehicle top value is greater than the second lower vehicle top threshold and less than the second upper vehicle top threshold, and the ordinate of the far-ground boundary of the vehicle detection frame is greater than the ordinate threshold, and the vehicle bottom value is less than the sixth vehicle bottom threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the vehicle top value is greater than the second lower vehicle top threshold and less than the third vehicle top threshold, and the vehicle bottom value is greater than the seventh lower vehicle bottom threshold and less than the seventh upper vehicle bottom threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the vehicle top value is greater than the third vehicle top threshold and less than or equal to the fourth vehicle top threshold, and the vehicle bottom value is greater than the second upper vehicle bottom threshold and less than the eighth vehicle bottom threshold, and the absolute value of the slope of the ground road length boundary of the three-dimensional berth projection frame is greater than the first slope threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the absolute value of the slope of the ground road length boundary is less than the second slope threshold, and the vehicle top value is greater than the fifth vehicle top threshold and less than the second upper vehicle top threshold, and the vehicle bottom value is greater than the sixth vehicle bottom threshold and less than the second upper vehicle bottom threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, and the vehicle top value is greater than the second lower vehicle top threshold and less than the sixth vehicle top threshold, and the vehicle bottom value is greater than the sixth vehicle bottom threshold and less than the second upper vehicle bottom threshold, then the vehicle is within the berth; Or, if the ratio of the intersection area to the vehicle is greater than the fifth vehicle ratio threshold, and the ratio of the intersection area to the berth is greater than the seventh berth ratio threshold, and the vehicle bottom value is greater than the tenth vehicle bottom threshold and less than the fifth upper vehicle bottom threshold, then the vehicle is within the berth.

7. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, then the three-dimensional projection algorithm is used to transform the calibration coordinates of the berth to obtain the three-dimensional berth projection frame corresponding to the berth number, and based on the three-dimensional berth projection frame and each vehicle detection frame, the parking state of each vehicle is calculated. It further includes: If the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold, and the vehicle top value is greater than the seventh lower vehicle top threshold and less than the seventh upper vehicle top threshold, then the vehicle is within the berth; Alternatively, if the ratio of the intersection area to the berth is greater than the sixth berth ratio threshold, and the roof value is greater than the seventh lower roof threshold and less than the seventh upper roof threshold, then the vehicle is within the berth; Alternatively, if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold, and the bottom value is greater than the ninth lower bottom threshold and less than the ninth upper bottom threshold, then the vehicle is within the berth; Alternatively, if the ratio of the intersection area to the berth is greater than the sixth berth ratio threshold, and the bottom value is greater than the ninth upper bottom threshold and less than the ninth lower bottom threshold, then the vehicle is within the berth.

8. The berth status analysis method based on cloud-edge integration according to claim 1, characterized in that, If there is no corresponding historical berth cache information for the berth number, then the calibration coordinates of the berth are transformed according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and the docking state of each vehicle is calculated based on the three-dimensional berth projection frame and each vehicle detection frame. It further includes: If the left value of the vehicle is greater than the fourth left threshold and less than the third upper left threshold, and the right value of the vehicle is greater than the fourth right threshold and less than the third upper right threshold, and the roof value is greater than the eighth roof threshold and less than the second upper roof threshold, and the bottom value is greater than the third bottom threshold and less than the second upper bottom threshold, then the vehicle is within the berth.

9. A berth status analysis system based on cloud-edge integration, characterized in that, It includes: A camera data acquisition module for acquiring the vehicle detection frame of each vehicle corresponding to the panoramic view of the berth of each camera and the calibration coordinates of the berth number corresponding to each camera; A historical cache judgment module for judging whether there is corresponding historical berth cache information for the berth number; A calculation module for, if there is no corresponding historical berth cache information for the berth number, transforming the calibration coordinates of the berth according to the three-dimensional projection algorithm to obtain the three-dimensional berth projection frame corresponding to the berth number, and calculating the docking state of each vehicle based on the three-dimensional berth projection frame and each vehicle detection frame; A vehicle detection frame cache judgment module for, if there is corresponding historical berth cache information for the berth number, judging whether each vehicle detection frame exists in the historical berth cache information; A cache calling module for, if so, calling the docking state of the vehicle corresponding to each vehicle detection frame according to the historical berth cache information; A vehicle docking calculation module for, if not, calculating the docking state of the vehicle corresponding to the vehicle detection frame based on the three-dimensional berth projection frame and the vehicle detection frame; A vehicle-in-berth judgment module for judging whether there is a vehicle within the berth corresponding to the berth number according to the docking state of each vehicle; A vehicle-in-berth judgment result module one for, if so, indicating that there is a vehicle within the berth corresponding to the berth number; A vehicle-in-berth judgment result module two for, if not, indicating that there is no vehicle within the berth corresponding to the berth number; A cache module for forming new berth cache information from the vehicle docking calculation result corresponding to the berth number for the next berth state determination; The calculation module includes: The intersection area to vehicle ratio module is used to calculate the intersection area between the circumscribed rectangle of the three-dimensional berth projection frame and each vehicle detection frame, and obtain the intersection area to vehicle ratio based on the intersection area and the area of the vehicle detection frame; The intersection area to berth ratio module is used to obtain the intersection area to berth ratio based on the intersection area and the area of the circumscribed rectangle of the three-dimensional berth projection frame; The ground intersection point acquisition module is used to draw a parallel line to the lower ground width boundary of the three-dimensional berth projection frame according to the midpoint of the ground boundary of the vehicle detection frame, and obtain the ground intersection point of the parallel line and the ground long side boundary of the three-dimensional berth projection frame close to the road; The vehicle bottom value calculation module is used to calculate the ground intersection-top difference between the abscissa of the ground intersection point and the abscissa of the vertex of the ground long side boundary close to the road, and the ground bottom-top difference between the abscissa of the bottom point of the ground long side boundary close to the road and the abscissa of the vertex of the ground long side boundary close to the road, and divide the ground intersection-top difference by the ground bottom-top difference to obtain the vehicle bottom value; The air intersection point acquisition module is used to draw a parallel line to the upper air width boundary of the three-dimensional berth projection frame according to the midpoint of the far ground boundary of the vehicle detection frame, and obtain the air intersection point of the parallel line and the air long side boundary of the three-dimensional berth projection frame close to the road; The vehicle top value calculation module is used to calculate the air intersection-top difference between the abscissa of the air intersection point and the abscissa of the vertex of the air long side boundary close to the road, and the air bottom-top difference between the abscissa of the bottom point of the air long side boundary close to the road and the abscissa of the vertex of the air long side boundary close to the road, and divide the air intersection-top difference by the air bottom-top difference to obtain the vehicle top value; The vehicle left side value calculation module is used to calculate the first abscissa difference between the abscissa of the road-side boundary of the vehicle detection frame and the abscissa of the vertex of the ground long side boundary close to the road, calculate the second abscissa difference between the abscissa of the bottom point of the ground long side boundary close to the road and the abscissa of the vertex of the ground long side boundary close to the road, and divide the first abscissa difference by the second abscissa difference to obtain the vehicle left side value; The vehicle right side value calculation module is used to calculate the third abscissa difference between the abscissa of the far-road boundary of the vehicle detection frame and the abscissa of the vertex of the ground far-road long side boundary of the three-dimensional berth projection frame, calculate the fourth abscissa difference between the abscissa of the bottom point of the ground far-road long side boundary and the abscissa of the vertex of the ground far-road long side boundary, and divide the third abscissa difference by the fourth abscissa difference to obtain the vehicle right side value; The corner vehicle judgment module is used to calculate the boundary distances between the four boundaries of the vehicle detection frame and the four boundaries of the berth panoramic view respectively. If the boundary distance is less than the pixel threshold, the vehicle is a corner vehicle; The occluded vehicle judgment module is used to determine that the vehicle is an occluded vehicle if the vehicle detection frame has an intersection with the surrounding vehicle detection frames and the ground boundary of the vehicle detection frame is above the ground boundary of the intersection vehicle detection frames; The vehicle-berth height ratio calculation module is used to divide the height of the vehicle detection frame by the height of the circumscribed rectangle of the three-dimensional berth projection frame to obtain the vehicle-berth height ratio; The first judgment module is used to determine that the vehicle is not in the parking space if the ratio of the intersection area to the vehicle is less than the first vehicle ratio threshold and the ratio of the intersection area to the parking space is less than the first parking space ratio threshold.

10. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: The second judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the first lower left threshold and less than the first upper left threshold, the ratio of the intersection area to the vehicle is greater than the second vehicle ratio threshold, and the roof value of the vehicle is greater than the first lower roof threshold and less than the first upper roof threshold. The third judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the first lower left threshold and less than the first upper left threshold, the ratio of the intersection area to the parking space is greater than the second parking space ratio threshold, and the roof value of the vehicle is greater than the first lower roof threshold and less than the first upper roof threshold. The fourth judgment module is used to determine that the vehicle is in the parking space if the right side value of the vehicle is greater than the first lower right threshold and less than the first upper right threshold, the ratio of the intersection area to the vehicle is greater than the second vehicle ratio threshold, and the roof value of the vehicle is greater than the first lower roof threshold and less than the first upper roof threshold. The fifth judgment module is used to determine that the vehicle is in the parking space if the right side value of the vehicle is greater than the first lower right threshold and less than the first upper right threshold, the ratio of the intersection area to the parking space is greater than the second parking space ratio threshold, and the roof value of the vehicle is greater than the first lower roof threshold and less than the first upper roof threshold.

11. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: The sixth judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the first upper left threshold and less than the second left threshold, the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, the ratio of the intersection area to the parking space is greater than the second parking space ratio threshold, and the vehicle is not the corner vehicle. The seventh judgment module is used to determine that the vehicle is in the parking space if the right side value of the vehicle is greater than the first upper right threshold and less than the second right threshold, the bottom value of the vehicle is greater than the first lower bottom threshold and less than the first upper bottom threshold, the ratio of the intersection area to the parking space is greater than the second parking space ratio threshold, and the vehicle is not the corner vehicle.

12. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: The eighth judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the third lower left threshold and less than the third upper left threshold, the right side value of the vehicle is greater than the third lower right threshold and less than the third upper right threshold, and the bottom value of the vehicle is greater than the second lower bottom threshold and less than the second upper bottom threshold. The ninth judgment module is used to determine that the vehicle is in the parking space if the left side value of the vehicle is greater than the third lower left threshold and less than the third upper left threshold, the right side value of the vehicle is greater than the third lower right threshold and less than the third upper right threshold, the bottom value of the vehicle is greater than the third bottom threshold, and the vehicle is the occluded vehicle.

13. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: The tenth judgment module is used to determine that the vehicle is in the parking space if the ratio of the intersection area to the parking space is greater than the third parking space ratio threshold and the vehicle-parking height ratio is greater than the first vehicle-parking height threshold. An eleventh judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the berth is less than a fourth berth ratio threshold and the vehicle-berth height ratio is greater than the first vehicle-berth height threshold.

14. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: A twelfth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is less than a third vehicle ratio threshold, the ratio of the intersection area to the berth is greater than a fifth berth ratio threshold, the difference between the ordinate of the outer far-ground boundary of the circumscribed rectangle of the three-dimensional berth projection frame and the ordinate of the far-ground boundary of the vehicle detection frame is greater than the ordinate threshold, the vehicle-berth height ratio is greater than the second vehicle-berth height threshold, and the vehicle bottom value is greater than the fourth vehicle bottom threshold; A thirteenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold and the vehicle bottom value is greater than the fifth lower vehicle bottom threshold and less than the second upper vehicle bottom threshold; A fourteenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the vehicle top value is greater than the second lower vehicle top threshold and less than the second upper vehicle top threshold, the ordinate of the far-ground boundary of the vehicle detection frame is greater than the ordinate threshold, and the vehicle bottom value is less than the sixth vehicle bottom threshold; A fifteenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the vehicle top value is greater than the second lower vehicle top threshold and less than the third vehicle top threshold, and the vehicle bottom value is greater than the seventh lower vehicle bottom threshold and less than the seventh upper vehicle bottom threshold; A sixteenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the vehicle top value is greater than the third vehicle top threshold and less than or equal to the fourth vehicle top threshold, the vehicle bottom value is greater than the second upper vehicle bottom threshold and less than the eighth vehicle bottom threshold, and the absolute value of the slope of the ground road-length boundary of the three-dimensional berth projection frame is greater than the first slope threshold; A seventeenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the absolute value of the slope of the ground road-length boundary is less than the second slope threshold, the vehicle top value is greater than the fifth vehicle top threshold and less than the second upper vehicle top threshold, and the vehicle bottom value is greater than the sixth vehicle bottom threshold and less than the second upper vehicle bottom threshold; An eighteenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the third vehicle ratio threshold, the vehicle top value is greater than the second lower vehicle top threshold and less than the sixth vehicle top threshold, and the vehicle bottom value is greater than the sixth vehicle bottom threshold and less than the second upper vehicle bottom threshold; A nineteenth judgment module, configured to determine that the vehicle is within the berth if the ratio of the intersection area to the vehicle is greater than the fifth vehicle ratio threshold, the ratio of the intersection area to the berth is greater than the seventh berth ratio threshold, and the vehicle bottom value is greater than the tenth vehicle bottom threshold and less than the fifth upper vehicle bottom threshold.

15. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: The twentieth judgment module is used to determine that the vehicle is within the parking space if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold, and the roof value is greater than the seventh lower roof threshold and less than the seventh upper roof threshold; The twenty-first judgment module is used to determine that the vehicle is within the parking space if the ratio of the intersection area to the parking space is greater than the sixth parking space ratio threshold, and the roof value is greater than the seventh lower roof threshold and less than the seventh upper roof threshold; The twenty-second judgment module is used to determine that the vehicle is within the parking space if the ratio of the intersection area to the vehicle is greater than the fourth vehicle ratio threshold, and the vehicle bottom value is greater than the ninth lower vehicle bottom threshold and less than the ninth upper vehicle bottom threshold; The twenty-third judgment module is used to determine that the vehicle is within the parking space if the ratio of the intersection area to the parking space is greater than the sixth parking space ratio threshold, and the vehicle bottom value is greater than the ninth upper vehicle bottom threshold and less than the ninth lower vehicle bottom threshold.

16. The berth status analysis system based on cloud-edge integration according to claim 9, wherein, The calculation module further includes: The twenty-fourth judgment module is used to determine that the vehicle is within the parking space if the left side value of the vehicle is greater than the fourth left side threshold and less than the third upper left side threshold, the right side value of the vehicle is greater than the fourth right side threshold and less than the third upper right side threshold, the roof value is greater than the eighth roof threshold and less than the second upper roof threshold, and the vehicle bottom value is greater than the third vehicle bottom threshold and less than the second upper vehicle bottom threshold.

Images