Intelligent parking guide system based on machine vision

Abstract

The invention relates to the field of parking guidance, in particular to an intelligent parking guidance system based on machine vision, which dispatches real-time parking space data and monitoring data by responding to the situation that a target vehicle enters a parking lot; determining unoccupied parking spaces in the parking lot according to the parking space data, generating a plurality of corresponding parking paths, and determining an optimal path; identifying a plurality of passing areas of the optimal path, and analyzing the sensitive interference characterization values according to the operation characteristics to determine sensitive interference passing areas; obtaining a corresponding historical reciprocation path set, and carrying out matching fitting on the historical reciprocation path set and the optimal path so as to screen an intersection track region; determining the area of the intersection track region based on the monitoring data, and calculating the involved interference characterization parameters by combining the shortest distance between the target vehicle and the edge of the intersection track region and the vehicle speed, so as to judge whether to correct the optimal path or not; and responding to the judgment result to regenerate the parking path based on the position of the target vehicle, thereby efficiently and accurately providing guidance for parking and improving the parking efficiency.

Description

Technical Field

[0001] The present invention relates to the field of parking guidance, and in particular to an intelligent parking guidance system based on machine vision. Background Art

[0002] Early parking guidance systems mostly used ultrasonic, geomagnetic and other sensor technologies. Although they can detect the status of parking spaces to a certain extent, they have problems such as limited accuracy and susceptibility to environmental interference. For example, ultrasonic sensors may make misjudgments when continuously detecting multiple parking spaces, and geomagnetic sensors are easily affected by the uneven distribution of vehicle metal parts and changes in the external electromagnetic environment. As a result, it is difficult to achieve accurate global guidance for complex parking environments, such as multi-story parking lots and special-shaped parking lots. Moreover, they can usually only provide simple information on whether the parking space is vacant, and cannot monitor and analyze the specific location and driving trajectory of the vehicle in detail.

[0003] With the continuous advancement of computer technology, image processing technology and sensor technology, machine vision technology has gradually matured. It has the advantages of non-contact detection, comprehensive information acquisition and high precision. By analyzing the images taken by the camera, it can not only accurately determine the occupancy of parking spaces, but also identify vehicles more accurately, providing data support for more refined parking guidance and management. At the same time, the emergence of high-resolution, high-frame-rate cameras and powerful image processors makes it possible to collect and process image information in parking lots in real time.

[0004] Among them, for parking lot users, a fast and convenient parking experience is more important. The intelligent parking guidance system provides real-time parking space information and optimal driving route planning, so that drivers do not need to blindly search for parking spaces in the parking lot, saving time and energy and improving user satisfaction with parking lot services.

[0005] Chinese patent application publication number: CN114882724A, discloses a parking guidance system and a parking guidance method, which relate to the technical field of parking vehicle guidance, including step S1, selecting an empty parking space according to the entrance position information of the vehicle to be guided, and determining the guidance route according to the number of route turning points; step S2, obtaining the real-time guidance route carried out simultaneously in the parking lot, and calibrating the guidance route of the vehicle to be guided with the real-time guidance route carried out simultaneously to determine whether to guide the vehicle to be guided with the guidance route of the vehicle to be guided; step S3, obtaining the number of intersections between the guidance route of the vehicle to be guided and the real-time guidance route carried out simultaneously, and determining whether to reselect the guidance route of the vehicle to be guided or reselect the empty parking space of the vehicle to be guided according to the number of real-time guidance routes at each real-time intersection position. The invention reduces the driving time of the vehicle in the parking lot and the intersection of vehicles by controlling the parking route of the vehicle, thereby improving the parking efficiency and avoiding vehicle congestion.

[0006] However, there are still the following problems in the prior art: Parking guidance mainly focuses on providing fixed driving routes, and is unable to make advance intervention adjustments based on the real-time traffic conditions in the parking lot. Adjustment plans for driving routes are often given only when congestion has already occurred, resulting in reduced parking guidance efficiency. Summary of the invention

[0007] To this end, the present invention provides an intelligent parking guidance system based on machine vision to overcome the problem in the prior art that parking guidance mainly focuses on providing a fixed driving path and is unable to make advance intervention adjustments based on the real-time traffic conditions in the parking lot. Adjustment plans for the driving path are often given only when congestion has occurred, resulting in reduced parking guidance efficiency.

[0008] To achieve the above object, the present invention provides an intelligent parking guidance system based on machine vision, which includes: A response dispatching module is used to dispatch real-time parking space data and monitoring data in response to a target vehicle entering the parking lot; a path generation module, connected to the response scheduling module, for determining the available parking spaces in the parking lot according to the parking space data, generating a number of corresponding parking paths based on the available parking spaces, and determining an optimal path; An interference identification module, connected to the path generation module, for identifying a plurality of passing areas of the optimal path, and analyzing sensitive interference characterization values ​​according to operation characteristics of each passing area to determine sensitive interference passing areas; A matching and screening module, connected to the interference identification module, is used to obtain a set of historical paths of the sensitive interference passing area, and match and fit the set with the optimal path to screen the intersection trajectory area; An analysis and determination module connected to the matching and screening module is used to determine the area of ​​the intersection trajectory region based on the monitoring data, and calculate the interference characterization parameter in combination with the shortest distance between the target vehicle and the edge of the intersection trajectory region and the corresponding vehicle speed to determine whether to modify the optimal path; A path push module, used to respond to the determination result of the analysis and determination module, to regenerate a parking path based on the location of the target vehicle and push it to the user end; The operation characteristics include road width redundancy and vehicle speed deviation value.

[0009] Furthermore, the response scheduling module is also used to determine the target vehicle, including: To obtain a number of vehicles within the detection area; for identifying the steering direction of each of the vehicles and determining the horizontal angle between the steering direction and the barrier of the parking lot entrance; If there is any vehicle whose corresponding horizontal angle is 90°, the vehicle is determined as the target vehicle.

[0010] Furthermore, the path generation module is used to determine the optimal path, including: To determine the straight-line distance and number of bends of each parking path; The parking path with the shortest straight-line distance and the least number of curves among the parking paths is determined as the optimal path.

[0011] Furthermore, the interference identification module is used to analyze the sensitive interference characterization value according to the operation characteristics of each of the passing areas, including: The ratio of the road width redundancy threshold to the road width redundancy is used as a first sensitive interference feature; Using the ratio of the vehicle speed deviation value to the vehicle speed deviation threshold as a second sensitive interference feature; The sum of the first sensitive interference feature and the second sensitive interference feature is used as the sensitive interference characterization value.

[0012] Furthermore, the interference identification module is used to determine the sensitive interference path area, including: If the sensitive interference characterization value corresponding to the passing area is greater than or equal to the sensitive interference characterization threshold, the passing area is determined as the sensitive interference passing area.

[0013] Furthermore, the difference between the average speed of several vehicles passing through the area and the speed limit of the parking lot is used as the speed deviation value.

[0014] Furthermore, the matching screening module is used to screen the intersection trajectory area, including: To determine the overlap ratio between the optimal path segment in each passing area and several paths in the historical path set; Used to regard the passing areas with the overlap ratio greater than the predetermined ratio threshold as the shared areas; for screening the common areas meeting the intersection benchmark conditions as the intersection trajectory areas; The intersection reference condition is that the traffic volume in the common area in the current time period is greater than the traffic volume threshold and the driving time in the common area is greater than the driving time threshold.

[0015] Furthermore, the analysis and determination module is used to calculate the interference characterization parameters, including: The ratio of the intersection track area threshold to the intersection track area is used as the first interference feature; The sum of the ratio of the shortest distance threshold to the edge of the intersection trajectory area of ​​the target vehicle and the shortest distance and the ratio of the vehicle speed to the vehicle speed threshold is used as the second interference feature; The first interference feature and the second interference feature are weightedly summed to obtain the interference characterization parameter.

[0016] Furthermore, the analysis and determination module is used to determine whether to modify the optimal path, including: If the interference characterization parameter is less than the interference characterization parameter threshold, it is determined to modify the optimal path.

[0017] Furthermore, the path push module is used to respond to the determination result of the analysis and determination module, including: If the analysis and determination module determines that the optimal path should be corrected, a parking path is regenerated based on the location of the target vehicle and pushed to the user end.

[0018] Compared with the prior art, the present invention is provided with a response scheduling module for responding to a target vehicle entering a parking lot to schedule real-time parking space data and monitoring data; a path generation module for determining the available parking spaces in the parking lot according to the parking space data, generating a number of corresponding parking paths based on the available parking spaces, and determining the optimal path; an interference identification module for identifying a number of passing areas of the optimal path, analyzing the sensitive interference characterization value according to the operation characteristics of each passing area, so as to determine the sensitive interference passing area; a matching screening module for obtaining a historical path set of the sensitive interference passing area, matching and fitting with the optimal path, so as to screen the intersection trajectory area; an analysis and determination module for determining the area of ​​the intersection trajectory area based on the monitoring data, and calculating the interference characterization parameter in combination with the shortest distance of the target vehicle from the edge of the intersection trajectory area and the corresponding vehicle speed, so as to determine whether to modify the optimal path; a path push module for responding to the determination result of the analysis and determination module, so as to regenerate the parking path based on the location of the target vehicle. The present invention can provide guidance for parking efficiently and accurately, and improve the efficiency of parking.

[0019] In particular, the present invention sets an interference recognition module to analyze the road width redundancy and speed deviation value of the area where the target vehicle needs to pass through, and evaluates the sensitivity of the target vehicle to interference when passing through the area. In the actual parking process, the parking lot is in dynamic change in terms of vehicle flow. The actual operation of the area through which the vehicle passes can be reflected through the operation characteristics. Among them, the roads in some passing areas can accommodate multiple vehicles to travel in parallel, and the road width redundancy can reflect the abundance of road space and the smoothness of vehicle traffic. At the same time, the driving speed of the vehicle in the passing area can reflect the congestion of the area, and then, the speed deviation value is used to evaluate the degree of deviation of the vehicle driving speed in the passing area from the prescribed speed. Therefore, the present invention analyzes the sensitive interference characterization value through the operation characteristics of the passing area to characterize the degree of interference of the target vehicle in the passing area, and provides data support for the subsequent determination of the sensitive interference passing area. The present invention can provide guidance for parking efficiently and accurately, and improve the efficiency of parking.

[0020] In particular, the present invention sets up a matching and screening module to match and fit the optimal path with several historical paths of the sensitive interference passing area to determine the intersection trajectory area, consider the common situation of the paths traveled by the historical vehicles passing through the sensitive interference passing area and the optimal path, and at the same time take into account the traffic volume and driving time of the common area in the current time period, evaluate the traffic flow density in the intersection trajectory area, and conduct real-time analysis of the above situation to help discover traffic congestion areas in advance and correct the optimal path in time, thereby avoiding these areas that are prone to interference, generating a more efficient parking path, and ensuring smooth parking operations for the target vehicle. The present invention can provide guidance for parking efficiently and accurately, and improve parking efficiency.

[0021] In particular, the present invention sets an analysis and judgment module to consider the relevant parameters of the intersection trajectory area and the target vehicle itself, reflects the scale of the interference through the area of ​​the intersection trajectory area, and flexibly evaluates the degree of interference risk according to the dynamic changes of the target vehicle. The shortest distance between the target vehicle and the edge of the intersection trajectory area and the speed of the corresponding target vehicle can directly relate to the possibility of the target vehicle entering the intersection trajectory area without applying correction guidance and the degree to which the target vehicle is easily affected by the traffic conditions in the intersection trajectory area, thereby evaluating the possibility of continuing to use the optimal path causing the vehicle to slow down, wait or frequently adjust the driving direction due to the existence of the intersection trajectory area, thereby increasing the time consumption of the parking process of the target vehicle. Therefore, the present invention characterizes the interference of the target vehicle in the current time period through the interference characterization parameter, the degree of affecting smooth parking, and provides data support for determining whether to correct the optimal path. The present invention can provide guidance for parking efficiently and accurately, and improve parking efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A functional module diagram of an intelligent parking guidance system based on machine vision according to an embodiment of the invention; Figure 2 A logic decision diagram for determining a sensitive interference path area for an embodiment of the invention; Figure 3 A logical decision diagram for screening intersection trajectory regions for embodiments of the invention; Figure 4 This is a logical decision diagram for determining whether to modify the optimal path according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0024] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0025] It should be noted that, in the description of the present invention, the terms such as “inside” and the like indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0026] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0027] See also Figure 1 As shown, it is a functional module diagram of an intelligent parking guidance system based on machine vision according to an embodiment of the present invention. The intelligent parking guidance system based on machine vision according to an embodiment of the present invention includes: A response dispatching module is used to dispatch real-time parking space data and monitoring data in response to a target vehicle entering the parking lot; a path generation module, connected to the response scheduling module, for determining the available parking spaces in the parking lot according to the parking space data, generating a number of corresponding parking paths based on the available parking spaces, and determining an optimal path; An interference identification module, connected to the path generation module, for identifying a plurality of passing areas of the optimal path, and analyzing sensitive interference characterization values ​​according to operation characteristics of each passing area to determine sensitive interference passing areas; A matching and screening module, connected to the interference identification module, is used to obtain a set of historical paths of the sensitive interference passing area, and match and fit the set with the optimal path to screen the intersection trajectory area; An analysis and determination module connected to the matching and screening module is used to determine the area of ​​the intersection trajectory region based on the monitoring data, and calculate the interference characterization parameter in combination with the shortest distance between the target vehicle and the edge of the intersection trajectory region and the corresponding vehicle speed to determine whether to modify the optimal path; A path push module, used to respond to the determination result of the analysis and determination module, to regenerate a parking path based on the location of the target vehicle and push it to the user end; The operation characteristics include road width redundancy and vehicle speed deviation value.

[0028] Specifically, there is no specific limitation on the method of scheduling real-time parking space data and monitoring data. The data collected by the cameras installed in the parking lot can be stored in the corresponding database, and the real-time parking space data and monitoring data can be obtained by scheduling the database.

[0029] Specifically, there is no specific limitation on the method of generating a number of parking paths. A path planning algorithm can be used, for example, the A* algorithm, the Dijkstra algorithm, etc., to calculate the driving path from the location of the target vehicle to a number of vacant parking spaces based on pre-stored parking lot map data and relevant constraints such as road conditions. Of course, other methods can also be used to generate parking paths, which will not be repeated here.

[0030] Specifically, the parking lot contains several paths for vehicles to travel. Based on the planning and design of the parking lot, the width of the road is relatively flexible and varies in length and width. For example, the width of some path sections can allow two vehicles to travel in parallel, while the width of some path sections can only allow one vehicle to travel. Therefore, the situation in the area where the target vehicle passes through is determined based on the real-time monitoring data obtained, for example, whether there are other vehicles traveling on the path segment passing through the area, the length of the road width on which the vehicle is traveling, and then the remaining width of the path segment, that is, the road width redundancy, is determined, which will not be repeated here.

[0031] Specifically, there is no limitation on the specific structures of the response scheduling module, path generation module, interference identification module, matching screening module, analysis and judgment module and path push module, and they themselves or each unit therein can be composed of logic components or a combination of logic components, and the logic components include a field programmable processor, a computer or a microprocessor in a computer.

[0032] Specifically, the response scheduling module is also used to determine the target vehicle, including: To obtain a number of vehicles within the detection area; for identifying the steering direction of each of the vehicles and determining the horizontal angle between the steering direction and the barrier of the parking lot entrance; If there is any vehicle whose corresponding horizontal angle is 90°, the vehicle is determined as the target vehicle.

[0033] It is understandable that when a vehicle enters a parking lot, it needs to pass through the barrier at the entrance of the parking lot. The front of the vehicle is relatively parallel to the barrier. During this process, the vehicle body on the driver's side is in a vertical position relationship with the barrier. Therefore, the direction of the vehicle body on the driver's side is determined as the steering direction. The area that can be captured by the monitoring device arranged at the entrance of the parking lot is determined as the detection area.

[0034] Specifically, the path generation module is used to determine the optimal path, including: To determine the straight-line distance and number of bends of each parking path; The parking path with the shortest straight-line distance and the least number of curves among the parking paths is determined as the optimal path.

[0035] It is understandable that if the path that the vehicle needs to travel has more bends, it will take more time to travel compared to a straight path segment. Therefore, the time consumption of the parking path is comprehensively evaluated based on the straight-line distance of the parking path and the number of bends on the parking path to determine the optimal path. Among them, the straight-line distance from the target vehicle position point to the vacant parking space is determined as the straight-line distance of the parking path, and the position of the target vehicle when it is at the gate railing at the entrance of the parking lot is determined as the target vehicle position point.

[0036] Specifically, the interference identification module is used to analyze the sensitive interference characterization value according to the operation characteristics of each of the passing areas, including: The ratio of the road width redundancy threshold to the road width redundancy is used as a first sensitive interference feature; Using the ratio of the vehicle speed deviation value to the vehicle speed deviation threshold as a second sensitive interference feature; The sum of the first sensitive interference feature and the second sensitive interference feature is used as the sensitive interference characterization value.

[0037] Specifically, the difference between the average vehicle speed of several vehicles in the passing area and the speed limit of the parking lot is used as the vehicle speed deviation value.

[0038] In this embodiment, the purpose of setting the road width redundancy threshold and the vehicle speed deviation threshold is to characterize the situation that the congestion level in the target vehicle path area is relatively serious and the passage is relatively difficult. By obtaining historical monitoring data, calling the historical road width redundancy data and vehicle speed deviation value data in the same time period, solving the road width redundancy mean and the vehicle speed deviation mean, based on the purpose of setting the above two thresholds, the road width redundancy threshold is determined as the product of the road width redundancy mean and the road width redundancy coefficient, and the vehicle speed deviation threshold is determined as the product of the vehicle speed deviation mean and the vehicle speed deviation coefficient, wherein the road width redundancy coefficient is selected in the interval [0.85,0.9], and the vehicle speed deviation coefficient is selected in the interval [1.02,1.06].

[0039] Specifically, the present invention sets an interference recognition module to analyze the road width redundancy and speed deviation value of the area where the target vehicle needs to pass through, so as to evaluate the sensitivity of the target vehicle to interference when passing through the area. In the actual parking process, the parking lot is in dynamic change in terms of vehicle flow, and the actual operation conditions of the area through which the vehicle passes can be reflected through the operation characteristics. Among them, the roads in some passing areas can accommodate multiple vehicles to travel in parallel, and the road width redundancy can reflect the abundance of road space and the smoothness of vehicle traffic. For example, a wider road width can allow vehicles to pass more smoothly, reduce mutual interference between vehicles, and reduce the risk of collision or scratching, while a narrower road width may cause difficulty in vehicle passage, especially in the event of oncoming vehicles, which requires driving. The personnel need to operate the vehicle more carefully, and even one of the vehicles may need to perform evasive maneuvers, which increases the time and difficulty of vehicle driving and affects the traffic efficiency. At the same time, the driving speed of vehicles in the passing area can reflect the congestion situation in the area. Then, the deviation degree of the driving speed of vehicles in the passing area relative to the prescribed speed is evaluated by the speed deviation value. For example, a large speed deviation value indicates that the overall driving speed of vehicles in the area may be too fast or too slow, which may affect the smoothness of traffic. Therefore, the present invention analyzes the sensitive interference characterization value through the operation characteristics of the passing area to characterize the affected interference degree of the target vehicle in the passing area, and provides data support for the subsequent determination of the sensitive interference passing area. The present invention can provide guidance for parking efficiently and accurately, and improve the efficiency of parking.

[0040] Specifically, see Figure 2 As shown, it is a logic decision diagram for determining a sensitive interference path area according to an embodiment of the present invention. The interference identification module is used to determine the sensitive interference path area, including: If the sensitive interference characterization value corresponding to the passing area is greater than or equal to the sensitive interference characterization threshold, the passing area is determined as the sensitive interference passing area; If the sensitive interference characterization value corresponding to the passing area is less than the sensitive interference characterization threshold, the passing area is determined as a non-sensitive interference passing area.

[0041] The sensitive interference characterization threshold is selected in the interval [2.15,2.23].

[0042] It is understandable that parking lots usually limit the speed of vehicles passing through. The speed of vehicles in the passing area can reflect the current traffic congestion level in the passing area. Furthermore, the speed deviation value can further evaluate the severity of the congestion.

[0043] Specifically, see Figure 3 As shown, it is a logical decision diagram for screening the intersection trajectory area according to an embodiment of the present invention. The matching screening module is used to screen the intersection trajectory area, including: To determine the overlap ratio between the optimal path segment in each passing area and a number of paths in the historical path set, where the overlap ratio is the ratio of the number of paths in the passing area that overlap with the optimal path segment to the total number of paths; Used to regard the passing areas with the overlap ratio greater than the predetermined ratio threshold as the shared areas; for screening the common areas meeting the intersection benchmark conditions as the intersection trajectory areas; The intersection reference condition is that the traffic volume in the common area in the current time period is greater than the traffic volume threshold and the driving time in the common area is greater than the driving time threshold.

[0044] In this embodiment, the purpose of setting the predetermined ratio threshold is to characterize the situation where the optimal path segment has a high degree of overlap with the traveling paths. By acquiring historical monitoring data, calling each traveling path in the same time period, determining the ratio of overlap between each traveling path and the remaining traveling paths in the passing area, solving the ratio mean, and based on the purpose of setting the predetermined ratio threshold, determining the ratio mean as the predetermined ratio threshold; The purpose of setting the traffic flow threshold and the driving duration threshold is to characterize the situation where the traffic conditions in the common area are relatively congested. By obtaining historical monitoring data, calling the historical traffic flow data and the historical driving duration data in the same time period, solving the mean traffic flow and the mean driving duration, based on the purpose of setting the above two thresholds, the traffic flow threshold is determined as the product of the mean traffic flow and the traffic deviation coefficient, and the driving duration threshold is determined as the product of the mean driving duration and the duration deviation coefficient, wherein the traffic deviation coefficient is selected in the interval [1.2,1.4], and the duration deviation coefficient is selected in the interval [1.05,1.1].

[0045] Specifically, the historical travel path set is a set consisting of a number of historical travel paths pre-stored in a relevant database, wherein the travel paths are collected through monitoring data, which will not be described in detail.

[0046] It is understandable that the traffic in and out of the parking lot varies at different times of the day. For example, on work commuting days, the traffic in and out of the parking lot will be relatively large during the morning and evening rush hours. Therefore, by dividing the time periods to analyze the traffic flow and driving duration in each time period accordingly, the vehicle condition can be analyzed more accurately. In this embodiment, the duration corresponding to the current time period is set to 30 minutes.

[0047] Specifically, the present invention sets up a matching and screening module to match and fit the optimal path with several historical paths of the sensitive interference passing area to determine the intersection trajectory area, consider the common situation of the paths of historical vehicles traveling in the sensitive interference passing area and the optimal path, and take into account the traffic volume and driving time of the shared area in the current time period to evaluate the traffic flow density in the intersection trajectory area. If the sensitive interference passing area is not only an area shared by the optimal path, but also has a large traffic pressure in the current time period in the area, which is easy to involve and interfere with the driving of the target vehicle, real-time analysis of the above situation is helpful to discover traffic congestion areas in advance and correct the optimal path in time, thereby avoiding these areas that are prone to interference, generating a more efficient parking path, and ensuring smooth parking operations for the target vehicle. The present invention can provide guidance for parking efficiently and accurately, and improve parking efficiency.

[0048] Specifically, the analysis and determination module is used to calculate the interference characterization parameters, including: The ratio of the intersection track area threshold to the intersection track area is used as the first interference feature; The sum of the ratio of the shortest distance threshold to the edge of the intersection trajectory area of ​​the target vehicle and the shortest distance and the ratio of the vehicle speed to the vehicle speed threshold is used as the second interference feature; The first interference feature and the second interference feature are weightedly summed to obtain the interference characterization parameter.

[0049] In actual situations, the relevant parameters of the vehicles in the parking lot can more directly reflect the possibility of the vehicles being involved in the congested traffic and the degree of interference. Therefore, in the implementation, the shortest distance between the target vehicle and the edge of the intersection trajectory area and the corresponding vehicle speed are given priority. Therefore, a slightly higher weight is given to the second interference feature calculated based on the above two features. Therefore, when performing weighted summation, the weight of the first interference feature is set to 0.4, and the weight of the second interference feature is set to 0.6; The purpose of setting the intersection trajectory area threshold, the shortest distance threshold from the target vehicle to the edge of the intersection trajectory area, and the speed threshold of the target vehicle is to characterize the situation where the regional vehicle conditions in the current time period have a high degree of interference with parking. By obtaining historical related data of the same time period, calling the intersection trajectory area historical data, the shortest distance historical data from the target vehicle to the edge of the intersection trajectory area, and the speed historical data of the target vehicle, the mean intersection trajectory area, the mean shortest distance from the target vehicle to the edge of the intersection trajectory area, and the mean speed of the target vehicle are solved. Based on the purpose of setting the above three thresholds The area threshold of the intersection trajectory area is determined as the product of the mean area of ​​the intersection trajectory area and the area deviation coefficient, the shortest distance threshold of the target vehicle from the edge of the intersection trajectory area is determined as the product of the mean shortest distance of the target vehicle from the edge of the intersection trajectory area and the distance deviation coefficient, and the speed threshold of the target vehicle is determined as the product of the mean speed of the target vehicle and the speed deviation coefficient, wherein the area deviation coefficient is selected in the interval [0.8, 0.9], the distance deviation coefficient is selected in the interval [0.85, 0.9], and the speed deviation coefficient is selected in the interval [1.05, 1.1].

[0050] Specifically, see Figure 4 As shown, it is a logic decision diagram for determining whether to modify the optimal path according to an embodiment of the present invention. The analysis and determination module is used to determine whether to modify the optimal path, including: If the interference characterization parameter is less than the interference characterization parameter threshold, it is determined to modify the optimal path; If the interference characterization parameter is greater than or equal to the interference characterization parameter threshold, it is determined that there is no need to modify the optimal path.

[0051] The threshold of the interference characterization parameter is selected in the interval [1.68,1.74].

[0052] Specifically, the present invention sets up an analysis and judgment module, considers the relevant parameters of the intersection trajectory area and the target vehicle itself, reflects the scale of the interference through the area of ​​the intersection trajectory area, and flexibly evaluates the degree of interference risk according to the dynamic changes of the target vehicle. The shortest distance between the target vehicle and the edge of the intersection trajectory area and the speed of the corresponding target vehicle can directly relate to the possibility of the target vehicle entering the intersection trajectory area without applying correction guidance and the degree to which the target vehicle is easily affected by the traffic conditions in the intersection trajectory area, and then evaluates the possibility that continuing to use the optimal path will cause the vehicle to slow down, wait or frequently adjust the driving direction due to the existence of the intersection trajectory area, thereby increasing the time consumption of the parking process of the target vehicle. Therefore, the present invention characterizes the interference of the target vehicle in the current time period through the interference characterization parameter, the degree of affecting smooth parking, and provides data support for determining whether to correct the optimal path. The present invention can efficiently and accurately provide guidance for parking and improve parking efficiency.

[0053] Specifically, the path push module is used to respond to the determination result of the analysis and determination module, including: If the analysis and determination module determines that the optimal path should be corrected, a parking path is regenerated based on the location of the target vehicle and pushed to the user terminal; If the analysis and determination module determines that the optimal path should be corrected, the parking operation is completed according to the optimal path.

[0054] Specifically, there is no specific limitation on the method of pushing the parking path. The generated path can be pushed to the user terminal pre-installed in the mobile phone by the driver of the vehicle, such as a parking lot APP, which will not be elaborated here.

[0055] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An intelligent parking guidance system based on machine vision, characterized in that: include: A response dispatching module is used to dispatch real-time parking space data and monitoring data in response to a target vehicle entering the parking lot; a path generation module, connected to the response scheduling module, for determining the available parking spaces in the parking lot according to the parking space data, generating a number of corresponding parking paths based on the available parking spaces, and determining an optimal path; An interference identification module, connected to the path generation module, for identifying a plurality of passing areas of the optimal path, and analyzing sensitive interference characterization values ​​according to operation characteristics of each passing area to determine sensitive interference passing areas; A matching and screening module, connected to the interference identification module, is used to obtain a set of historical paths of the sensitive interference passing area, and match and fit the set with the optimal path to screen the intersection trajectory area; An analysis and determination module connected to the matching and screening module is used to determine the area of ​​the intersection trajectory region based on the monitoring data, and calculate the interference characterization parameter in combination with the shortest distance between the target vehicle and the edge of the intersection trajectory region and the corresponding vehicle speed to determine whether to modify the optimal path; A path push module, used to respond to the determination result of the analysis and determination module, to regenerate a parking path based on the location of the target vehicle and push it to the user end; The operation characteristics include road width redundancy and vehicle speed deviation value.

2. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The response scheduling module is also used to determine the target vehicle, including: To obtain a number of vehicles within the detection area; for identifying the steering direction of each of the vehicles and determining the horizontal angle between the steering direction and the barrier of the parking lot entrance; If there is any vehicle whose corresponding horizontal angle is 90°, the vehicle is determined as the target vehicle.

3. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The path generation module is used to determine the optimal path, including: To determine the straight-line distance and number of bends of each parking path; The parking path with the shortest straight-line distance and the least number of curves among the parking paths is determined as the optimal path.

4. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The interference identification module is used to analyze the sensitive interference characterization value according to the operation characteristics of each of the passing areas, including: The ratio of the road width redundancy threshold to the road width redundancy is used as a first sensitive interference feature; Using the ratio of the vehicle speed deviation value to the vehicle speed deviation threshold as a second sensitive interference feature; The sum of the first sensitive interference feature and the second sensitive interference feature is used as the sensitive interference characterization value.

5. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The interference identification module is used to determine the sensitive interference path area. include, If the sensitive interference characterization value corresponding to the passing area is greater than or equal to the sensitive interference characterization threshold, the passing area is determined as the sensitive interference passing area.

6. The intelligent parking guidance system based on machine vision according to claim 4, characterized in that: The difference between the average speed of several vehicles in the passing area and the speed limit of the parking lot is used as the speed deviation value.

7. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The matching screening module is used to screen the intersection trajectory area, including: To determine the overlap ratio between the optimal path segment in each passing area and several paths in the historical path set; Used to regard the passing areas with the overlap ratio greater than the predetermined ratio threshold as the shared areas; for screening the common areas meeting the intersection benchmark conditions as the intersection trajectory areas; The intersection reference condition is that the traffic volume in the common area in the current time period is greater than the traffic volume threshold and the driving time in the common area is greater than the driving time threshold.

8. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The analysis and determination module is used to calculate the interference characterization parameters, including: The ratio of the intersection track area threshold to the intersection track area is used as the first interference feature; The sum of the ratio of the shortest distance threshold to the edge of the intersection trajectory area of ​​the target vehicle and the shortest distance and the ratio of the vehicle speed to the vehicle speed threshold is used as the second interference feature; The first interference feature and the second interference feature are weightedly summed to obtain the interference characterization parameter.

9. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The analysis and determination module is used to determine whether to modify the optimal path, including: If the interference characterization parameter is less than the interference characterization parameter threshold, it is determined to modify the optimal path.

10. The intelligent parking guidance system based on machine vision according to claim 1, characterized in that: The path push module is used to respond to the determination result of the analysis and determination module. include, If the analysis and determination module determines that the optimal path should be corrected, a parking path is regenerated based on the location of the target vehicle and pushed to the user end.

Images