A method, device, electronic device and storage medium for planning a vehicle route

Abstract

The embodiments of the present application provide a method, device, electronic device and storage medium for planning a car-finding route. Based on the real-time recording of parking events, the target parking events of the target object are analyzed, thereby associating the characteristics of the target object with the target area where the target parking event occurs. Compared with the prior art, the method for planning a car-finding route provided by the embodiments of the present application can determine the accurate target area based on the recorded parking events without being affected by environmental factors. Even if the light in the shooting environment is dim or the shooting angle of the camera device is not good, the accurate target area can be determined according to the above method, thereby improving the accuracy of planning the car-finding route.

Description

Technical Field

[0001] The present application relates to the field of intelligent analysis technology, and in particular to a method, device, electronic device and storage medium for planning a vehicle route. Background Art

[0002] In large indoor parking lots, the parking space arrangement rules and building structures are usually repetitive, and there is usually a lack of obvious guidance information in the parking lot. Therefore, after the target object parks the target vehicle in the parking space and leaves, when the target object needs to find the target vehicle, the design of the above parking lot will cause great inconvenience to the target object; therefore, in large indoor parking lots, the service system often provides the target object with corresponding vehicle search guidance services to assist the target object in finding the target vehicle.

[0003] Under the relevant technology, before providing the car-finding guidance service, the service system needs to use the video equipment (such as cameras, etc.) deployed in the parking lot to identify the facial information of the target object and the license plate information of the target vehicle when detecting that the target object gets off the vehicle, and determine the coordinate information of the target vehicle in the electronic map corresponding to the large indoor parking lot; in this way, when the service system receives the car-finding instruction provided by the target object, it can plan the corresponding car-finding route for the target object based on the facial information of the target object + the license plate information and coordinate information of the target vehicle, taking the current position of the target object as the starting point, that is, provide the car-finding guidance service.

[0004] However, the above method of providing car-finding guidance services places high demands on the shooting angle and image clarity of the camera equipment, and is greatly affected by environmental factors. When environmental conditions are poor, the accuracy of car-finding route planning will be affected.

[0005] For example, in actual scenarios, when the light in the shooting environment is dim, the shooting picture will be blurry, resulting in some feature data missing in the facial information of the target object and the license plate information of the target vehicle identified from the shooting picture, thereby affecting the accuracy of the planning of the corresponding vehicle search route when the service system plans the corresponding vehicle search route.

[0006] For example, in actual applications, when the shooting angle of the camera equipment is not good, the complete image of the target object or target vehicle will not appear in the shooting picture, resulting in the facial information of the target object and the license plate information of the target vehicle identified from the shooting picture. There is a phenomenon of missing some feature data, thereby affecting the accuracy of vehicle search route planning. Summary of the invention

[0007] The embodiments of the present invention provide a method, device, electronic device and storage medium for planning a vehicle-finding route, which are used to improve the planning accuracy of the vehicle-finding route.

[0008] In a first aspect, an embodiment of the present application provides a method for planning a vehicle route, comprising:

[0009] An image to be detected captured at a target moment is acquired, and at least one video frame to be detected captured within a first time range from the target moment forward is acquired, and based on the at least one video frame to be detected, parking status information of each parking area in a set of parking areas included in the image to be detected within the first time range is obtained, wherein the image to be detected includes a target object in the parking lot and the corresponding surrounding environment.

[0010] Perform image processing on the image to be detected to obtain the facial feature information of the target object and the position information of the target object in the image to be detected.

[0011] Based on the obtained position information and parking status information of the target object, a target parking area corresponding to the target object is determined in the parking area set, and a target area identifier of the target parking area is obtained.

[0012] The acquired facial feature information and target area identification are sent to the server, so that when the server receives the vehicle search instruction sent by the target object, it plans a corresponding vehicle search route for the target object based on the acquired facial feature information and target area identification, taking the current position of the target object as the starting point.

[0013] In a second aspect, an embodiment of the present application further provides a vehicle route planning device, including:

[0014] The first acquisition module is used to acquire the image to be detected captured at the target moment, and acquire at least one video frame to be detected captured within a first time range from the target moment forward, and obtain parking status information of each parking area in the first time range in a set of parking areas included in the image to be detected based on the at least one video frame to be detected, wherein the image to be detected includes a target object in the parking lot and the corresponding surrounding environment.

[0015] The second acquisition module is used to perform image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected.

[0016] The determination module is used to determine the target parking area corresponding to the target object in the parking area set based on the location information and the parking state information, and obtain the target area identifier of the target parking area.

[0017] The communication module is used to send facial feature information and target area identification to the server, so that when the server receives the car search instruction sent by the target object, it plans a corresponding car search route for the target object based on the facial feature information and the target area identification, taking the current position of the target object as the starting point.

[0018] In an optional embodiment, when obtaining parking status information of each parking area in a set of parking areas included in the image to be detected based on the obtained at least one video frame to be detected, the first acquisition module is specifically used to:

[0019] For at least one obtained video frame to be detected, the following operations are performed respectively: based on the parking conditions of vehicles in each parking space contained in a video frame to be detected, the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected are determined.

[0020] Based on the obtained occupied parking space quantity information and vacant parking space quantity information corresponding to at least one video frame to be detected, the local parking event closest to the target time and the corresponding local parking time of each parking space within the first time range are determined respectively.

[0021] The number of parking spaces included in each parking area in the parking area set is determined respectively, and based on the local parking events and the corresponding local parking times corresponding to each parking space, the candidate parking events and the corresponding candidate parking times of each parking area within the first time range are determined respectively.

[0022] Based on the obtained candidate parking events and the corresponding candidate parking times, parking status information corresponding to each parking area is generated respectively.

[0023] In an optional embodiment, when determining the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected based on the parking conditions of vehicles in each parking space contained in a video frame to be detected, the first acquisition module is specifically used to:

[0024] Based on a preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in a video frame to be detected is obtained respectively, and based on a preset offset model, the vehicle coordinate information of each vehicle is offset processed to obtain the vehicle offset coordinate information corresponding to each vehicle.

[0025] Obtain the parking space area coordinate information of each parking space in a video frame to be detected.

[0026] According to the vehicle offset coordinate information corresponding to each vehicle and the parking area coordinate information of each parking space, the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected are determined.

[0027] In an optional embodiment, when respectively acquiring vehicle coordinate information of each vehicle in a video frame to be detected based on a preset vehicle detection algorithm, the first acquisition module is specifically used to:

[0028] According to a preset segmentation rule, a video frame to be detected is segmented to obtain at least one video frame sub-image corresponding to the video frame to be detected.

[0029] For at least one video frame sub-image, the following operations are performed respectively: determining the number of vehicles contained in a sub-image to be detected, and obtaining local vehicle coordinate information of each vehicle in a video frame sub-image based on a preset vehicle detection algorithm.

[0030] Based on the local vehicle coordinate information of each vehicle in at least one video frame sub-image, a preset segmentation and restoration algorithm is used to obtain the vehicle coordinate information of each vehicle in a video frame to be detected.

[0031] In an optional embodiment, when determining candidate parking events and corresponding candidate parking times for each parking area within the first time range based on the local parking events and corresponding local parking times corresponding to each parking space, the first acquisition module is specifically configured to:

[0032] For each parking area, perform the following operations:

[0033] Based on the local parking events and corresponding local parking times corresponding to each parking space contained in a parking area, the local parking time closest to the target time is taken as the candidate parking time corresponding to the parking area, and the local parking event corresponding to the candidate parking time is taken as the candidate parking event corresponding to the parking area.

[0034] In an optional embodiment, when performing image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected, the second acquisition module is specifically used to:

[0035] Based on the preset face detection algorithm, the image to be detected is processed to obtain the facial feature information of the target object.

[0036] Based on the preset portrait detection algorithm, the image to be detected is processed to obtain the position information of the target object in the image to be detected.

[0037] In an optional embodiment, when determining the target parking area corresponding to the target object in the parking area set based on the location information and the parking state information, the determination module is specifically used to:

[0038] Based on the parking status information, at least one candidate parking area, each candidate parking event and the corresponding candidate parking time within the first time range, a candidate parking event closest to the target time is selected as the latest parking event, and the candidate parking time corresponding to the latest parking event is selected as the latest parking time.

[0039] It is determined whether the latest parking time is within a preset second time range. If so, the candidate parking area corresponding to the latest parking event is used as the target parking area; otherwise, the coupling area composed of the parking areas in the parking area set is used as the target parking area.

[0040] In a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the computer program is executed by the processor, the processor implements any one of the vehicle route planning methods in the first aspect above.

[0041] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for planning a vehicle-finding route of the first aspect is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 An architecture diagram of a vehicle-finding route planning system provided in an embodiment of the present application;

[0043] Figure 2 A schematic diagram of an image acquisition device provided in an embodiment of the present application;

[0044] Figure 3 A schematic diagram of a process flow of a method for planning a vehicle search route provided in an embodiment of the present application;

[0045] Figure 4 An example diagram of each video frame to be detected provided in the embodiment of the present application;

[0046] Figure 5 A schematic diagram of a process for obtaining information on the number of occupied parking spaces provided in an embodiment of the present application;

[0047] Figure 6 An example diagram of a video x to be detected provided in an embodiment of the present application;

[0048] Figure 7 A schematic diagram of a process for determining the number of occupied parking spaces provided in an embodiment of the present application;

[0049] Figure 8a , Figure 8b An example diagram of vehicle coordinate information offset provided in an embodiment of the present application;

[0050] Fig. 9 A schematic diagram of a process for obtaining coordinate information provided in an embodiment of the present application;

[0051] Fig.10 A schematic diagram of a process for obtaining a video frame sub-image provided in an embodiment of the present application;

[0052] Fig.11 An example diagram of obtaining vehicle coordinate information in a video frame sub-image provided in an embodiment of the present application;

[0053] Fig.12 A schematic diagram of the image processing process provided in the embodiment of the present application;

[0054] Fig.13 A schematic diagram of a process for obtaining a target parking area provided in an embodiment of the present application;

[0055] Fig.14 An example diagram of a target parking area after coupling provided in an embodiment of the present application;

[0056] Fig.15 An example diagram of a car-finding route provided in an embodiment of the present application;

[0057] Fig.16 A schematic diagram of a vehicle-finding route planning device provided in an embodiment of the present application;

[0058] Fig.17 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to improve the accuracy of the car-finding route planning, in the embodiment of the present application, based on the real-time recording of parking events, the target parking event of the target object is analyzed, so as to associate the characteristics of the target object with the target area where the target parking event occurs. Compared with the prior art, the car-finding route planning method provided in the embodiment of the present application can determine the accurate target area based on the recorded parking events without being affected by environmental factors. Even if the light in the shooting environment is dim or the shooting angle of the camera device is not good, the accurate target area can be determined according to the above method, thereby improving the accuracy of the car-finding route planning.

[0060] The preferred implementation modes of the present application are further described in detail below in conjunction with the accompanying drawings.

[0061] See also Figure 1 As shown, in the embodiment of the present application, a large number of imaging devices, such as cameras, cameras, etc., are installed inside the indoor parking lot. Optionally, the imaging device can use a fisheye lens to achieve the maximum photographic angle. Through the fisheye lens, the imaging device can collect the latest images inside the indoor parking lot in real time, such as:

[0062] When the target object walks into the parking lot from the entrance, the latest image of the target object at the moment of entering the parking lot can be collected;

[0063] When the target object gets off the vehicle, the latest image of the target object leaving the target vehicle can be collected.

[0064] Each image obtained by the camera device can be used as an image to be detected.

[0065] See also Figure 2 As shown, taking any image captured by a camera device as an example, one image includes various parking areas in the parking lot, each parking area contains N parking spaces, and N is a preset natural number.

[0066] In the embodiment of the present application, for ease of implementation, the environmental space in the parking lot is planned. Specifically, three parking spaces are used as a parking area. This is because a camera device using a fisheye lens can capture a panoramic image of up to 12 parking spaces, and a panoramic image captured by a camera device using a fisheye lens is usually divided into four parking areas.

[0067] Based on the above parking lot environment planning method, refer to Figure 3 As shown, in the embodiment of the present application, the specific manner in which the management device performs vehicle search route planning is as follows:

[0068] Step 31: Acquire an image to be detected captured at the target moment, and acquire at least one video frame to be detected captured within a first time range from the target moment forward, and based on the acquired at least one video frame to be detected, obtain parking status information of each parking area in a set of parking areas included in the image to be detected within the first time range, wherein the image to be detected includes a target object in the parking lot and the corresponding surrounding environment.

[0069] In actual scenarios, the target object often drives the target vehicle into an indoor parking lot, parks the target vehicle, gets off the target vehicle, and leaves the area where the target vehicle is located.

[0070] Since the target object on the target vehicle cannot be detected, the acquired image to be detected including the target object and the corresponding surrounding environment usually corresponds to the time range after the target object gets off the target vehicle, that is, the target parking event corresponding to the target object has ended at this time; therefore, in order to accurately calculate the time when the target parking event occurs for the target object, it is necessary to set a first time range forward based on the target moment when the image to be detected is acquired. Within the first time range, it is determined that the target parking event may have occurred for the target object within the set of parking areas contained in the image to be detected.

[0071] Optionally, a corresponding first time range may be set according to the traffic volume in the parking lot.

[0072] For example, if the number of vehicles entering or leaving a parking lot increases significantly over a period of time, it is considered that the traffic volume in the parking lot has increased during this period. In this case, the target object may leave the target vehicle faster than usual. Based on this situation, a shorter first time range can be set, such as 3 minutes ahead of the target time.

[0073] For another example, if the number of vehicles entering or leaving a parking lot entrance or exit decreases significantly over a period of time, it is considered that the traffic volume in the parking lot decreases over this period of time. In this case, the target object may leave the target vehicle after parking for a period of time. Based on this situation, a longer first time range may be set, such as 10 minutes ahead of the target time.

[0074] After setting the first time range, further, in the process of executing step 30, at least one video frame to be detected can be obtained according to the image to be detected and the first time range.

[0075] Specifically, at least one video frame to be detected containing an image to be detected can be obtained in the following manner: all panoramic images captured by a camera device that captures the image to be detected within a first time range are obtained, the above panoramic images are arranged in chronological order, and are encoded respectively to obtain a series of panoramic image streams arranged in chronological order, and then at least one corresponding encoded panoramic image is extracted from the above panoramic image streams in sequence according to a preset frame extraction frequency as the video frame to be detected.

[0076] For example, assuming that the first time range is set to 10 minutes, the camera device captures all panoramic images within 10 minutes after the target moment, encodes all images, and extracts 500 video frames to be detected containing the image a to be detected based on the set frame extraction frequency F.

[0077] See also Figure 4 As shown, taking 3 of the 500 video frames to be detected as an example, video frame 1 to be detected, video frame 2 to be detected and video frame 3 to be detected all contain a parking area set (i.e., parking areas A, B, C, and D), and in video frame 3 to be detected, a target object appears, indicating that a target parking event has occurred.

[0078] It can be seen that through the above 500 video frames to be detected, the parking status information of each parking area in the past 10 minutes can be obtained, and each parking status information represents the occupancy status of each parking space in the corresponding parking area. For example, the parking status information of parking area A in the past 10 minutes is: in the past 10 minutes, parking space 1 and parking space 3 have been occupied.

[0079] See also Figure 5 As shown, after obtaining at least one video frame to be detected within the first time range in the above manner, further, in the process of executing step 30, when obtaining the parking state information of each parking area in the first time range in the parking area set included in the detection image based on the obtained at least one video frame to be detected, the following operations may also be performed but are not limited to:

[0080] Step 311: for at least one video frame to be detected, perform the following operations respectively: based on the parking conditions of vehicles in each parking space contained in a video frame to be detected, determine the number of occupied parking spaces and the number of vacant parking spaces corresponding to the video frame to be detected.

[0081] For ease of description, in an embodiment of the present application, any one of the at least one video frame to be detected (hereinafter referred to as the video frame x to be detected) is taken as an example for relevant explanation, and the processing process of other video frames to be detected adopts the same method and will not be repeated.

[0082] Specifically, when executing step 311, it is possible to determine which parking spaces are occupied and which parking spaces are vacant based on the parking conditions of the respective parking spaces presented in the video frame to be detected, and obtain corresponding information on the number of occupied parking spaces and the number of vacant parking spaces.

[0083] For example, see Figure 6 As shown, in the video frame x to be detected, parking space 1, parking space 3, parking space 4, parking space 6, parking space 8, and parking space 11 are all occupied, and parking space 2, parking space 5, parking space 7, parking space 9, parking space 10, and parking space 12 are still idle. The management device counts the parking conditions of each parking space and determines that in the video frame x to be detected, the number of occupied parking spaces is 6, and the number of idle parking spaces is 6.

[0084] By adopting the above method, the parking conditions of vehicles corresponding to each video frame to be detected can be obtained. Assuming that there are 5 video frames to be detected, the parking conditions of vehicles corresponding to each of them are shown in Table 1:

[0085] Table 1

[0086]

[0087]

[0088] Further, in an optional embodiment, still taking the video frame x to be detected as an example, see Figure 7 As shown, when executing step 311, the management device may also adopt the following manner:

[0089] Step 3111: Based on a preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in the video frame x to be detected is obtained respectively, and based on a preset offset model, the vehicle coordinate information of each vehicle obtained is offset processed to obtain the vehicle offset coordinate information corresponding to each vehicle.

[0090] In specific implementations, since some vehicles are larger and their bodies exceed the specified parking range, a preset vehicle detection algorithm can be used to obtain the vehicle coordinate information of each vehicle (such as the coordinates of the vehicle's center point), and determine whether the obtained vehicle coordinate information is within the preset parking range to determine the parking situation of each vehicle, so as to avoid false detection problems caused by the body exceeding the parking range.

[0091] In addition, since the closer the vehicle is to the camera device, the larger the image in the video frame x to be detected, the vehicle coordinate information corresponding to the vehicle obtained by the vehicle detection algorithm will easily fall within the parking range of the adjacent parking space. To address the above problem, in an embodiment of the present application, the management device can further offset the obtained coordinate information, and then judge the parking situation of each vehicle through the obtained vehicle offset coordinate information.

[0092] Specifically, the management device performs offset processing on the obtained vehicle coordinate information, and the following situations exist:

[0093] Case 1: The management device performs vertical offset processing on the acquired coordinate information.

[0094] For example, see Figure 8a As shown, assuming that the resolution of the video frame x to be detected is: 1080*960 pixels, and the upper left corner of the video frame x to be detected is taken as the origin (i.e., the origin O), a coordinate system is established;

[0095] Based on the preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in the video frame to be detected is obtained respectively, as shown in Table 2:

[0096] Table 2

[0097]

[0098] Then, assuming that according to the preset offset model, the vehicle coordinate information of each vehicle is offset by vertically offsetting the above coordinate information by 25%, the vehicle offset coordinate information of each vehicle finally obtained is specifically shown in Table 3:

[0099] Table 3

[0100]

[0101] Case 2: The management device performs horizontal offset processing on the acquired coordinate information.

[0102] In an alternative embodiment, see Figure 8b As shown, assuming that the resolution of the video frame x to be detected is: 1080*960 pixels, and the upper left corner of the video frame x to be detected is taken as the origin (i.e., the origin O), a coordinate system is established;

[0103] Based on the preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in the video frame to be detected is obtained respectively, as shown in Table 4 below:

[0104] Table 4

[0105]

[0106] Then, assuming that according to the preset offset model, the vehicle coordinate information of each vehicle is offset by horizontally offsetting the above coordinate information by 25%, and considering the image symmetry, the vehicle coordinate information of the vehicles located in parking area A and parking area C needs to be offset to the right, and the vehicle coordinate information corresponding to the vehicles located in parking area B and parking area D needs to be offset to the left, then the vehicle offset coordinate information of each vehicle finally obtained is specifically shown in Table 5:

[0107] Table 5

[0108]

[0109] See also Fig. 9 As shown, in an optional embodiment, during the execution of step 3111, when respectively acquiring the vehicle coordinate information of each vehicle in the to-be-detected video frame x based on a preset vehicle detection algorithm, the following method may be specifically adopted:

[0110] Step 31111: Segment the video frame x to be detected according to a preset segmentation rule to obtain at least one video frame sub-image corresponding to the video frame x to be detected.

[0111] Specifically, the video frame to be detected may be divided into at least one corresponding video frame sub-image according to the boundaries of each parking area in the video frame to be detected x, so that each video frame sub-image corresponds to a parking area.

[0112] For example, still taking the video frame x to be detected as an example, refer to Fig.10 As shown, in the video frame x to be detected, according to the respective boundaries of parking area A, parking area B, parking area C, and parking area D, the video frame to be detected is divided into four corresponding video frame sub-images, each corresponding to the above four parking areas.

[0113] Step 31112: for at least one video frame sub-image, perform the following operations respectively: determine the number of vehicles contained in a sub-image to be detected, and based on a preset vehicle detection algorithm, obtain the local vehicle coordinate information of each vehicle in a video frame sub-image respectively.

[0114] Specifically, in an embodiment of the present application, when executing step 31112, the management device can respectively determine the number of vehicles corresponding to each video frame sub-image, and obtain the local vehicle coordinate information of each vehicle corresponding to each video frame sub-image according to the vehicle detection algorithm; optionally, the management device can detect the center point coordinates of each vehicle through the vehicle detection algorithm, and use the center point coordinates of each vehicle as the local vehicle coordinate information of the corresponding vehicle.

[0115] See also Fig.11 As shown, taking the video frame sub-image b as an example, assuming that the number of vehicles contained in the video frame sub-image b is 2, namely: vehicle 3 parked in parking space 4, and vehicle 4 parked in parking space 6; then, the management device can obtain the local vehicle coordinate information of vehicle 3 and vehicle 4 based on the preset vehicle detection algorithm. Assuming that the center point coordinates of the detected vehicles 3 and 4 are respectively used as the local vehicle coordinate information of vehicles 3 and 4, the local vehicle coordinate information of vehicle 3 obtained is: (190, 245); the local vehicle coordinate information of vehicle 4 is: (384, 304).

[0116] The same method is used to obtain the local vehicle coordinate information of the corresponding vehicles in video frame sub-image a, video frame sub-image c, and video frame sub-image d respectively. The above process only takes the local vehicle coordinate information of each vehicle corresponding to video frame sub-image b as an example, and the detailed process will not be repeated one by one.

[0117] Step 31113: Based on the local vehicle coordinate information of each vehicle in at least one video frame sub-image obtained, a preset segmentation and restoration algorithm is used to obtain the vehicle coordinate information of each vehicle in the video frame x to be detected.

[0118] Specifically, in an embodiment of the present application, when executing step 31113, the management device can use a preset segmentation and restoration algorithm to splice at least one obtained video frame sub-image and restore it into a complete video frame x to be detected. During this process, the segmentation and restoration algorithm will convert the local vehicle coordinate information corresponding to each vehicle obtained into the original coordinate system in the video frame x to be detected, so as to obtain the vehicle coordinate information of all vehicles contained in the video frame x to be detected.

[0119] Further, in an optional embodiment, Fig.10Taking the four video frame sub-images shown as an example, the management device uses a segmentation and restoration algorithm to restore the four video frame sub-images obtained above according to the original sequence to obtain the local vehicle coordinate information of each vehicle in the video frame x to be detected.

[0120] Assuming that the resolution of each video frame sub-image is: 540*480 pixels, the local vehicle coordinate information of each vehicle is shown in Table 6 below:

[0121] Table 6

[0122]

[0123] The segmentation and restoration algorithm is used to map the coordinate information of each local vehicle to the coordinate system of the video frame x to be detected. Assuming that the resolution of the video frame x to be detected is 1080*960 pixels, the vehicle coordinate information of each vehicle is shown in Table 7 below:

[0124] Table 7

[0125]

[0126] Step 3112: Obtain the parking space area coordinate information of each parking space in the video frame x to be detected.

[0127] Specifically, in an embodiment of the present application, after obtaining the vehicle offset coordinate information of each vehicle in the video frame x to be detected, when executing step 3112, the management device also needs to obtain the parking area coordinate information of each parking space to determine the parking situation of each vehicle, wherein the parking area coordinate information corresponding to each parking space is pre-configured in the server and obtained from the server by the management device through the data transmission channel connected between the server and the management device when executing step 3012.

[0128] For example, still Figure 8a Taking the video frame x to be detected as an example, assuming that the video frame x to be detected is 1080*960 pixels, the parking space area coordinate information corresponding to parking space 1 to parking space 12 in the video frame to be detected is specifically shown in Table 8 below:

[0129] Table 8

[0130] Parking space 1 Parking Space 2 Parking Space 3 Parking space 4 Parking Space 5 Parking Space 6 (172，311) (258，240) (356，175) (728，175) (827，219) (921，273) Parking Space 7 Parking Space 8 Parking Space 9 Parking space 10 Parking Space 11 Parking Space 12 (206，774) (309，703) (399，649) (715，649) (835，665) (934，747)

[0131] Step 3113: Determine the number of occupied parking spaces and the number of vacant parking spaces corresponding to the video frame x to be detected based on the obtained vehicle offset coordinate information corresponding to each vehicle and the parking area coordinate information of each parking space.

[0132] In the embodiment of the present application, taking the vehicle offset coordinate information (hereinafter referred to as vehicle offset coordinate information x) of any vehicle (hereinafter referred to as vehicle x) and the parking space area coordinate information (hereinafter referred to as parking space area coordinate information y) corresponding to any parking space (hereinafter referred to as parking space y) as an example, the execution process of step 3113 is introduced as follows:

[0133] Specifically, the management device may set a parking space coordinate range according to the parking space area coordinate information y. If the vehicle offset coordinate information x is within the parking space coordinate range, the management device determines that the parking space corresponding to the parking space area coordinate information y is occupied.

[0134] In an optional embodiment, still taking the video frame x to be detected as an example, assuming that the vehicle offset coordinate information x of vehicle x is: (730, 306), and the parking area coordinate information y of parking space y is: (728, 175); the management device sets a parking space coordinate range according to the parking space area coordinate information y (728, 175), including a horizontal coordinate range and a vertical coordinate range, wherein the horizontal coordinate range is: [630, 797], and the vertical coordinate range is: [163, 441]; at this time, the vehicle offset coordinate information x is within the parking space coordinate range, and the management device determines that the parking space y corresponding to the parking space area coordinate information y is occupied.

[0135] Through the above method, the management device can confirm the parking situation of each parking space in the video frame x to be detected, thereby obtaining the number of occupied parking spaces and the number of vacant parking spaces corresponding to the video frame x to be detected, which will not be repeated here.

[0136] Step 312: Based on the obtained occupied parking space quantity information and vacant parking space quantity information corresponding to at least one video frame to be detected, respectively determine the local parking event closest to the target time and the corresponding local parking time for each parking space within the first time range.

[0137] Specifically, for each parking space, the parking situation of vehicles in different video frames to be detected may be different. If the parking space is continuously occupied in consecutive video frames to be detected, it means that a parking event has occurred. The latest parking space being continuously occupied corresponds to the latest parking event, which can be regarded as a local parking event for the corresponding parking space, and the occurrence time of the local parking event can also be used as the corresponding local parking time.

[0138] Still taking the above five video frames x to be detected as an example, assuming that the above five video frames to be detected occur consecutively, the management device will determine that a parking event has occurred in a parking space only when it determines that a parking space is occupied in at least two consecutive video frames to be detected.

[0139] For example, based on the content recorded in Table 1, the management device can determine that parking space 1 is occupied in all five video frames to be detected, then the management device determines that a parking event has occurred in parking space 1 during this period of time, and because parking space 1 has been occupied, the management device determines that the next parking event has not occurred in parking space 1 during this period of time, then the parking event is the local parking event corresponding to parking space 1 that is closest to the target time.

[0140] In addition, Table 1 also records that parking space 8 is occupied in both the continuous video frame 1 to be detected and the video frame 2 to be detected, so the management device determines that a parking event has occurred in parking space 8 during this period; but then, Table 1 also records that parking space 8 is not occupied in the video frame 3 to be detected, so the management device determines that the last parking event corresponding to parking space 8 has ended; further, Table 1 continues to record that parking space 8 is occupied again in the continuous video frame 4 to be detected and the video frame 5 to be detected, so the management device determines that a parking event has occurred again in parking space 8 during this period. Assuming that the time corresponding to the video frame 4 to be detected and the video frame 5 to be detected is closer to the target time, the management device determines that this parking event closer to the target time is the local parking event corresponding to parking space 8.

[0141] Based on the above method, the partial parking events and corresponding partial parking times of the 12 parking spaces can be obtained, which will not be elaborated here.

[0142] Step 313: Determine the number of parking spaces contained in each parking area in the obtained parking area set, and determine the candidate parking events and the corresponding candidate parking times for each parking area within the first time range based on the local parking events and the corresponding local parking times corresponding to each parking space.

[0143] In the embodiment of the present application, each parking area corresponds to a different parking space, and each parking space has a determined local parking event and a corresponding local parking time. For each single parking area, the management device can determine that the local parking event corresponding to a parking space closest to the target time is a candidate parking event, and the corresponding local parking time is a candidate parking time according to the local parking times of the corresponding parking spaces.

[0144] Still taking the partial parking events of each parking space obtained from the above five video frames x to be detected as an example. Figure 5 It can be seen that the parking spaces corresponding to parking area A are parking space 1, parking space 2, and parking space 3;

[0145] Assume that the local parking event of parking space 1 is: a parking event that occurs 1 minute before the target time;

[0146] Assume that the local parking event of parking space 2 is: a parking event that occurs 3 minutes before the target time;

[0147] Assume that the local parking event of parking space 3 is: a parking event that occurs 5 minutes before the target time;

[0148] Since the local parking time corresponding to the local parking event corresponding to parking space 1 is closest to the target time, the local parking event corresponding to parking space 1 is determined as the candidate parking event of parking area A, and the corresponding local parking time is the candidate parking time.

[0149] Based on the above method, candidate parking events and corresponding candidate parking times of parking area B, parking area C, and parking area D may also be obtained, which will not be elaborated here.

[0150] Step 314: Based on the obtained candidate parking events and the corresponding candidate parking times, parking status information corresponding to each parking area is generated respectively.

[0151] For example, in an alternative embodiment,

[0152] Assume that the candidate parking event corresponding to parking area A is: a parking event that occurred in parking space 1 1 minute before the target time;

[0153] Assume that the candidate parking event for parking area B is: a parking event that occurred in parking space 5 3 minutes before the target time;

[0154] Assume that the candidate parking events for parking area C are: a parking event that occurred in parking space 10 5 minutes before the target time;

[0155] Assume that the candidate parking event for parking area D is: a parking event that occurs in parking space 12 2 minutes before the target time.

[0156] The management device generates parking status information corresponding to parking area A, parking area B, parking area C and parking area D respectively according to the candidate parking events and corresponding candidate parking times of the four parking areas.

[0157] According to the parking status information corresponding to each parking area, the management device can quickly confirm in which parking space in each parking area the parking event closest to the target time occurred, and at the same time, the corresponding parking time can also be determined.

[0158] Step 32: Perform image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected.

[0159] In the embodiment of the present application, the facial feature information of the target object is obtained in order to confirm the identity of the target object. Furthermore, the location information of the target object is obtained in order to determine the end point of the target vehicle search route planning.

[0160] For details, see Fig.12 As shown, in an optional embodiment, when executing step 32, the following operations may be performed but are not limited to:

[0161] Step 321: Based on a preset face detection algorithm, perform image processing on the image to be detected to obtain facial feature information of the target object.

[0162] Specifically, in the embodiment of the present application, the management device can detect the target object in the image to be detected based on the face detection algorithm, and obtain the facial feature information of the target object.

[0163] For example, a multi-task cascaded convolutional neural network algorithm (MTCNN) is used to obtain the facial feature information L = [L0, L1, ..., L 128 ].

[0164] Step 322: Based on a preset portrait detection algorithm, perform image processing on the image to be detected to obtain position information of the target object in the image to be detected.

[0165] Specifically, in an embodiment of the present application, the management device can obtain the object coordinate information of the center point of the target object in the image to be detected based on a preset portrait detection algorithm, and use the object coordinate information of the above center point as the position information of the target object in the image to be detected.

[0166] For example, assuming that based on a preset portrait detection algorithm, the object coordinate information of the center point of the target object in the image to be detected is obtained as: (472, 388), then the management device uses (472, 388) as the position information of the target object in the image to be detected.

[0167] Step 33: Based on the obtained location information and parking status information of the target object, determine the target parking area corresponding to the target object in the parking area set, and obtain the target area identifier of the target parking area.

[0168] Specifically, in an embodiment of the present application, when executing step 33, based on the object position information of the target object, the event position information of the target object that may perform the target parking event within the first time range can be determined, and based on the parking status information, the event time information of the target object that may perform the target parking event can be further determined. Then, based on the obtained event position information and event time information, the candidate parking events and candidate parking times corresponding to each parking area in the parking area set can be matched, so that the successfully matched parking area is used as the target parking area.

[0169] For example, the management device determines, based on the object position information of the target object in the image to be detected, that is, (472, 388), that the event position information of the target object that may perform the target parking event within the first time range is: (377, 321);

[0170] And assume that the parking area set contains the following parking areas:

[0171] Parking area A, parking area B, parking area C, parking area D, and the candidate parking events and candidate parking times corresponding to each parking area are as follows:

[0172] Candidate parking events corresponding to parking area A: a parking event that occurred in parking space 1 1 minute before the target time; and Candidate parking times corresponding to parking area A: 1 minute before the target time;

[0173] Candidate parking events corresponding to parking area B: a parking event that occurred in parking space 5 3 minutes before the target time; and Candidate parking times corresponding to parking area B: 3 minutes before the target time;

[0174] The candidate parking event corresponding to parking area C is: a parking event that occurred in parking space 10 5 minutes before the target time; and the candidate parking time corresponding to parking area C is: 5 minutes before the target time;

[0175] The candidate parking event corresponding to parking area D is: a parking event that occurred in parking space 12 2 minutes before the target time; and the candidate parking time corresponding to parking area C is: 2 minutes before the target time.

[0176] Based on the above parking status information, the event time information of the target object that may perform the target parking event can be further determined, which is: 1 minute, 3 minutes, 5 minutes, and 2 minutes.

[0177] Then, after detection, the management device determines that the target parking area corresponding to the target object in the parking area set contained in the image to be detected is parking area A, and obtains the target area identification of parking area A, which is 01-020-1 (floor-area-parking area A).

[0178] Further, see Fig.13 As shown, in an optional embodiment, during the execution of step 33, when determining the target parking area corresponding to the target object in the parking area set based on the obtained location information and parking status information of the target object, the management device may specifically perform but is not limited to the following operations:

[0179] Step 331: Based on the parking status information obtained, at least one candidate parking area, each candidate parking event and the corresponding candidate parking time within the first time range, select a candidate parking event closest to the target time as the latest parking event, and the candidate parking time corresponding to the latest parking event as the latest parking time.

[0180] In an optional embodiment,

[0181] Assume that the candidate parking events for parking area A are: a parking event that occurred in parking space 1 1 minute before the target time;

[0182] Assume that the candidate parking event for parking area B is: a parking event that occurred in parking space 5 3 minutes before the target time;

[0183] Assume that the candidate parking events for parking area C are: a parking event that occurred in parking space 10 5 minutes before the target time;

[0184] Assume that the candidate parking event for parking area D is: a parking event that occurs in parking space 12 2 minutes before the target time.

[0185] The candidate parking event of parking area A closest to the target time is selected as the latest parking event, and the candidate parking time corresponding to the latest parking event, that is, 1 minute before the target time, is selected as the latest parking time.

[0186] Step 332: Determine whether the latest parking time is within a preset second time range. If so, execute step 333; otherwise, execute step 324.

[0187] In actual scenarios, the target object may not get off the target vehicle immediately after parking it. Considering the possible time difference between the target moment and the latest parking time, it is optional to set a second time range forward based on the target moment when the image to be detected is acquired. The management device will determine whether the latest parking event is the target parking event based on whether the latest parking time is within the preset second time range.

[0188] Step 333: taking the candidate parking area corresponding to the latest parking event as the target parking area.

[0189] Specifically, if the management device determines that the latest parking time is within the preset second time range, it means that the target parking event is most likely to occur at the latest parking time, and the management device can determine the latest parking event as the target parking event.

[0190] For example, assuming that the management device sets the second time range to: 5 minutes before the target time; the latest parking event obtained is: a parking event occurring in parking space 1 in parking area A 1 minute before the target time, then the corresponding latest parking time is: 1 minute before the target time.

[0191] Since the latest parking time is within the preset second time range, the management device determines that the parking area A corresponding to the latest parking event is the target parking area.

[0192] Step 334: The coupling area formed by the parking areas in the parking area set is used as the target parking area.

[0193] Specifically, if the management device determines that the latest parking time is not within the preset second time range, it means that the target parking event is less likely to occur at the latest parking time, and the management device needs to aggregate the parking areas and couple the parking areas.

[0194] For example, assuming that the management device sets the second time range to: 2 minutes before the target time; the latest parking event obtained is: a parking event occurring in parking space 1 in parking area A 5 minutes before the target time, then the corresponding latest parking time is: 5 minutes before the target time.

[0195] Since the latest parking time is not within the preset second time range, the management device needs to group the parking areas, couple the parking areas, and use the coupled areas as target parking areas.

[0196] See also Fig.14 As shown, the management device will couple parking area A, parking area B, parking area C, and parking area D based on the boundaries of each parking area, obtain the coupled area M, and use area M as the target parking area.

[0197] Step 34: Send the acquired facial feature information and target area identification to the server, so that when the server receives the vehicle search instruction sent by the target object, it plans a corresponding vehicle search route for the target object based on the acquired facial feature information and target area identification, taking the current position of the target object as the starting point.

[0198] Specifically, after the management device sends the acquired facial feature information and target area identifier to the server, the server will associate and store the facial feature information and target area identifier of the target object.

[0199] Then, when the server receives the car search instruction sent by the target object, it will obtain the facial collection data of the target object based on the car search instruction, and the current location information of the target object. Then, the server will match the facial collection data with the pre-stored facial feature information. After the match is successful, the target area identifier corresponding to the matched facial feature information is obtained. Then, the location information of the target parking area is determined based on the target area identifier. Finally, based on the current object location information of the target object and the location information of the target parking area, a planned car search route is generated.

[0200] For example, assuming that the facial data is successfully matched with the pre-stored facial feature information, based on the target area identifier 01-020-1 corresponding to the successfully matched facial feature information, the location information of the target parking area is determined to be (732, 295), and at the same time, the current location information of the target object is obtained as (2000, 136), then the corresponding car search route is planned for the target object, such as Fig.15 shown.

[0201] It is worth noting that the embodiment of the present application only takes a special case based on a fisheye lens as an example. In more general cases, the camera equipment can also use other lenses to collect different types of images to be detected, and the management device can still use the above-mentioned vehicle route planning method for different types of images to be detected to plan the target vehicle route for the target object. Therefore, these situations should also be within the scope of protection of this application.

[0202] See also Fig.16 As shown, a vehicle route planning device provided in an embodiment of the present application includes: a first acquisition module 1601, a second acquisition module 1602, a determination module 1603, and a communication module 1604, wherein:

[0203] The first acquisition module 1601 is used to acquire the image to be detected captured at the target moment, and acquire at least one video frame to be detected captured within a first time range from the target moment forward, and obtain parking status information of each parking area in the first time range in a set of parking areas included in the image to be detected based on the at least one video frame to be detected, wherein the image to be detected includes a target object in the parking lot and the corresponding surrounding environment.

[0204] The second acquisition module 1602 is used to perform image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected.

[0205] The determination module 1603 is used to determine the target parking area corresponding to the target object in the parking area set based on the location information and the parking state information, and obtain the target area identifier of the target parking area.

[0206] Communication module 1604 is used to send facial feature information and target area identification to the server, so that when the server receives the car search instruction sent by the target object, it plans a corresponding car search route for the target object based on the facial feature information and target area identification, taking the current position of the target object as the starting point.

[0207] In an optional embodiment, when obtaining parking status information of each parking area in a set of parking areas included in the image to be detected based on the obtained at least one video frame to be detected, the first acquisition module 1601 is specifically used to:

[0208] For at least one obtained video frame to be detected, the following operations are performed respectively: based on the parking conditions of vehicles in each parking space contained in a video frame to be detected, the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected are determined.

[0209] Based on the obtained occupied parking space quantity information and vacant parking space quantity information corresponding to at least one video frame to be detected, the local parking event closest to the target time and the corresponding local parking time of each parking space within the first time range are determined respectively.

[0210] The number of parking spaces included in each parking area in the parking area set is determined respectively, and based on the local parking events and the corresponding local parking times corresponding to each parking space, the candidate parking events and the corresponding candidate parking times of each parking area within the first time range are determined respectively.

[0211] Based on the obtained candidate parking events and the corresponding candidate parking times, parking status information corresponding to each parking area is generated respectively.

[0212] In an optional embodiment, when determining the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected based on the parking conditions of vehicles in each parking space contained in a video frame to be detected, the first acquisition module 1601 is specifically used to:

[0213] Based on a preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in a video frame to be detected is obtained respectively, and based on a preset offset model, the vehicle coordinate information of each vehicle is offset processed to obtain the vehicle offset coordinate information corresponding to each vehicle.

[0214] Obtain the parking space area coordinate information of each parking space in a video frame to be detected.

[0215] According to the vehicle offset coordinate information corresponding to each vehicle and the parking area coordinate information of each parking space, the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected are determined.

[0216] In an optional embodiment, when respectively acquiring vehicle coordinate information of each vehicle in a video frame to be detected based on a preset vehicle detection algorithm, the first acquisition module 1601 is specifically used to:

[0217] According to a preset segmentation rule, a video frame to be detected is segmented to obtain at least one video frame sub-image corresponding to the video frame to be detected.

[0218] For at least one video frame sub-image, the following operations are performed respectively: determining the number of vehicles contained in a sub-image to be detected, and obtaining local vehicle coordinate information of each vehicle in a video frame sub-image based on a preset vehicle detection algorithm.

[0219] Based on the local vehicle coordinate information of each vehicle in at least one video frame sub-image, a preset segmentation and restoration algorithm is used to obtain the vehicle coordinate information of each vehicle in a video frame to be detected.

[0220] In an optional embodiment, when determining the candidate parking events and the corresponding candidate parking times of each parking area within the first time range based on the local parking events and the corresponding local parking times corresponding to each parking space, the first acquisition module 1601 is specifically used to:

[0221] For each parking area, perform the following operations:

[0222] Based on the local parking events and corresponding local parking times corresponding to each parking space contained in a parking area, the local parking time closest to the target time is taken as the candidate parking time corresponding to the parking area, and the local parking event corresponding to the candidate parking time is taken as the candidate parking event corresponding to the parking area.

[0223] In an optional embodiment, when performing image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected, the second acquisition module 1602 is specifically used to:

[0224] Based on the preset face detection algorithm, the image to be detected is processed to obtain the facial feature information of the target object.

[0225] Based on the preset portrait detection algorithm, the image to be detected is processed to obtain the position information of the target object in the image to be detected.

[0226] In an optional embodiment, when determining the target parking area corresponding to the target object in the parking area set based on the location information and the parking state information, the determination module 1603 is specifically used to:

[0227] Based on the parking status information, at least one candidate parking area, each candidate parking event and the corresponding candidate parking time within the first time range, a candidate parking event closest to the target time is selected as the latest parking event, and the candidate parking time corresponding to the latest parking event is selected as the latest parking time.

[0228] It is determined whether the latest parking time is within a preset second time range. If so, the candidate parking area corresponding to the latest parking event is used as the target parking area; otherwise, the coupling area composed of the parking areas in the parking area set is used as the target parking area.

[0229] Based on the same inventive concept as the above-mentioned application embodiment, the present application embodiment also provides an electronic device, which can be used for planning a car-finding route. In one embodiment, the electronic device can be a server, or a terminal device or other electronic device. In this embodiment, the structure of the electronic device can be as follows: Fig.17 As shown, it includes a memory 1701 , a communication interface 1703 and one or more processors 1702 .

[0230] The memory 1701 is used to store computer programs executed by the processor 1702. The memory 1701 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and programs required for running the instant messaging function, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.

[0231] The memory 1701 may be a volatile memory, such as a random-access memory (RAM); the memory 1701 may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), or the memory 1701 is any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 1701 may be a combination of the above memories.

[0232] The processor 1702 may include one or more central processing units (CPU) or a digital processing unit, etc. The processor 1702 is used to implement the above-mentioned vehicle route planning method when calling the computer program stored in the memory 1701.

[0233] The communication interface 1703 is used to communicate with terminal devices and other servers.

[0234] The specific connection medium between the memory 1701, the communication interface 1703 and the processor 1702 is not limited in the embodiment of the present application. Fig.17 In the embodiment, the memory 1701 and the processor 1702 are connected via a bus 1704. The bus 1704 is connected to the processor 1702 via a bus 1704. Fig.17 The connections between other components are shown in bold lines, which are only for illustration and are not intended to be limiting. Bus 1704 can be divided into address bus, data bus, control bus, etc. For ease of illustration, Fig.17 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0235] According to one aspect of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device executes any one of the above-mentioned embodiments of the method for planning a route for finding a vehicle. The program product can use any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0236] In the embodiment of the present application, the facial feature information of the target object and the position information of the target object in the image to be detected are obtained, and the target parking area corresponding to the target object is determined based on the above position information and the latest parking events that occurred in each parking area in the parking area set, and the target area identification of the target parking area is obtained, and then the corresponding car search route is planned for the target object based on the facial feature information of the target object and the target area identification of the target parking area. Based on this method, the facial feature information of the identified target object is associated with the target area identification of the corresponding target parking area. Compared with the prior art, this method is not affected by environmental factors, avoids the problem of missing some feature data due to environmental factors, ensures the credibility of the target parking area corresponding to the target object, and improves the planning accuracy of the car search route.

[0237] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0238] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0239] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0240] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0241] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for planning a vehicle route, characterized in that: include: Acquire an image to be detected captured at a target time, and acquire at least one video frame to be detected captured within a first time range from the target time forward, and obtain parking state information of each parking area in a set of parking areas included in the image to be detected within the first time range based on the at least one video frame to be detected, wherein the image to be detected includes a target object in a parking lot and a corresponding surrounding environment; Performing image processing on the image to be detected to obtain facial feature information of a target object and position information of the target object in the image to be detected; Based on the location information and the parking state information, determining a target parking area corresponding to the target object in the parking area set, and acquiring a target area identifier of the target parking area; The facial feature information and the target area identifier are sent to a server, so that when the server receives a car search instruction sent by the target object, it plans a corresponding car search route for the target object based on the facial feature information and the target area identifier, taking the current position of the target object as the starting point.

2. The method according to claim 1, characterized in that The obtaining, based on the at least one video frame to be detected, parking status information of each parking area in the parking area set included in the image to be detected within the first time range includes: For the at least one video frame to be detected, the following operations are respectively performed: based on the parking conditions of vehicles in each parking space contained in the video frame to be detected, the number of occupied parking spaces and the number of vacant parking spaces corresponding to the video frame to be detected are determined; Based on the obtained information on the number of occupied parking spaces and the number of vacant parking spaces corresponding to each of the at least one video frame to be detected, respectively determine the local parking events closest to the target time and the corresponding local parking times for each of the parking spaces within the first time range; Determine the number of parking spaces included in each parking area in the parking area set, and determine candidate parking events and corresponding candidate parking times for each parking area within the first time range based on the local parking events and corresponding local parking times corresponding to each parking space; Based on the obtained candidate parking events and the corresponding candidate parking times, the parking state information corresponding to each parking area is generated respectively.

3. The method according to claim 2, characterized in that The method of determining the number of occupied parking spaces and the number of vacant parking spaces corresponding to a video frame to be detected based on the parking conditions of the vehicles in each parking space included in the video frame to be detected includes: Based on a preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in the one to-be-detected video frame is respectively obtained, and based on a preset offset model, the vehicle coordinate information of each vehicle is offset processed to obtain the vehicle offset coordinate information corresponding to each vehicle; Obtaining the parking space area coordinate information of each parking space in the one video frame to be detected; According to the vehicle offset coordinate information corresponding to each of the vehicles and the parking area coordinate information of each of the parking spaces, the number information of occupied parking spaces and the number information of vacant parking spaces corresponding to the one video frame to be detected are determined.

4. The method according to claim 3, characterized in that Based on a preset vehicle detection algorithm, the vehicle coordinate information of each vehicle in the one video frame to be detected is obtained respectively, including: According to a preset segmentation rule, the one to-be-detected video frame is segmented to obtain at least one video frame sub-image corresponding to the one to-be-detected video frame; For the at least one video frame sub-image, the following operations are respectively performed: determining the number of vehicles contained in a sub-image to be detected, and obtaining local vehicle coordinate information of each vehicle in the sub-image of the video frame based on a preset vehicle detection algorithm; Based on the local vehicle coordinate information of each vehicle in the at least one video frame sub-image, a preset segmentation and restoration algorithm is used to obtain the vehicle coordinate information of each vehicle in the one video frame to be detected.

5. The method according to claim 2, characterized in that The determining, based on the local parking events and the local parking times corresponding to the respective parking spaces, candidate parking events and the corresponding candidate parking times for the respective parking areas within the first time range respectively includes: For each parking area, perform the following operations respectively: Based on the local parking events and corresponding local parking times corresponding to each parking space included in a parking area, the local parking time closest to the target time is used as the candidate parking time corresponding to the parking area, and the local parking event corresponding to the candidate parking time is used as the candidate parking event corresponding to the parking area.

6. The method according to any one of claims 1 to 5, characterized in that The performing image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected includes: Based on a preset face detection algorithm, performing image processing on the image to be detected to obtain facial feature information of the target object; Based on a preset portrait detection algorithm, image processing is performed on the image to be detected to obtain position information of the target object in the image to be detected.

7. The method according to any one of claims 2 to 5, characterized in that: The determining, based on the position information and the parking state information, a target parking area corresponding to the target object in the parking area set includes: Based on the parking state information, the at least one candidate parking area, the candidate parking events and the corresponding candidate parking times within the first time range, selecting a candidate parking event closest to the target time as the latest parking event, and the candidate parking time corresponding to the latest parking event as the latest parking time; Determine whether the latest parking time is within a preset second time range, If yes, taking the candidate parking area corresponding to the latest parking event as the target parking area; Otherwise, a coupling area formed by each parking area in the parking area set is used as the target parking area.

8. A vehicle route planning device, characterized in that: include: A first acquisition module is used to acquire an image to be detected captured at a target time, and to acquire at least one video frame to be detected captured within a first time range from the target time forward, and based on the at least one video frame to be detected, obtain parking status information of each parking area in a set of parking areas included in the image to be detected within the first time range, wherein the image to be detected includes a target object in a parking lot and a corresponding surrounding environment; A second acquisition module is used to perform image processing on the image to be detected to obtain facial feature information of the target object and position information of the target object in the image to be detected; a determination module, configured to determine, based on the location information and the parking state information, a target parking area corresponding to the target object in the parking area set, and obtain a target area identifier of the target parking area; The communication module is used to send the facial feature information and the target area identifier to the server, so that when the server receives the car search instruction sent by the target object, it plans a corresponding car search route for the target object based on the facial feature information and the target area identifier and takes the current position of the target object as the starting point.

9. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for planning a vehicle-finding route according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images