License plate recognition method and device for geographic information aided decision-making

Abstract

The invention relates to the technical field of license plate recognition, and discloses a geographic information aided decision-making license plate recognition method and device, and the method comprises the steps: carrying out the image and data synchronous coding operation of a vehicle image of a target vehicle collected by flight equipment and target metadata, and obtaining a coded vehicle image of the target vehicle; carrying out initial license plate recognition operation on the coded vehicle image to obtain an initial license plate recognition result of the target vehicle, and judging whether geographic information auxiliary decision-making operation needs to be carried out on the license plate of the target vehicle or not; and if yes, performing geographic information auxiliary decision-making operation on the license plate of the target vehicle to obtain license plate geographic address information of the target vehicle so as to determine a target license plate number identification result of the target vehicle. Visibly, when the license plate image recognition effect is not ideal, the image recognition result is combined with geographic analysis, accurate geographic address locking is carried out on the recognized target, and the recognition reliability and accuracy of the vehicle license plate are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of license plate recognition, and in particular, to a license plate recognition method and device assisted by geographic information for decision-making. Background Art

[0002] In the field of license plate recognition, drones have become an important tool for obtaining vehicle information due to their unique high-altitude perspective and flexible maneuverability. By analyzing the video images transmitted by drones, vehicle license plates can be effectively recognized, providing new technical means for fields such as traffic management and security monitoring.

[0003] However, drones still face many challenges in the field of license plate recognition. First, the image quality collected by drones highly depends on specific environmental conditions and is easily interfered by external factors such as changes in lighting, adverse weather (such as rain, snow, haze), etc. These factors can cause problems such as blurred images, decreased contrast, and loss of key information, thereby affecting the detection accuracy of artificial intelligence algorithms. Second, unstable factors such as jitter and tilt may occur during the flight of drones, which will also have an adverse impact on the image quality. In addition, complex and changeable actual application scenarios (such as urban dense areas, highways, etc.) further increase the difficulty of license plate detection, resulting in the license plate recognition accuracy being difficult to meet the actual needs. Therefore, there is an urgent need to develop a technical solution that can improve the license plate recognition accuracy. Summary of the Invention

[0004] The present invention provides a license plate recognition method and device assisted by geographic information for decision-making, which can combine the image recognition result with geographic analysis to accurately lock the geographical address of the recognition target when the license plate image recognition effect is not ideal, effectively improving the recognition reliability and accuracy of vehicle license plates.

[0005] To solve the above technical problems, in the first aspect of the present invention, a license plate recognition method assisted by geographic information for decision-making is disclosed. The method includes:

[0006] Obtain a vehicle image of a target vehicle collected by a flight device, and determine target metadata corresponding to the vehicle image;

[0007] According to the vehicle image and the target metadata, perform a graph-data synchronization encoding operation on the vehicle image and the target metadata to obtain an encoded vehicle image of the target vehicle;

[0008] Perform an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle, and determine whether a geographic information assisted decision-making operation needs to be performed on the license plate of the target vehicle according to the initial license plate recognition result;

[0009] When the judgment result is yes, perform a geographic information assisted decision-making operation on the license plate of the target vehicle to obtain the license plate geographic address information of the target vehicle, and determine the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographic address information.

[0010] As an optional implementation manner, in the first aspect of the present invention, the target metadata corresponding to the vehicle image includes the longitude and latitude parameters, altitude parameters, attitude angle parameters of the flying device, and the relative attitude angle parameters and field of view angle parameters of the pod of the flying device when the flying device captures the vehicle image;

[0011] Among them, the performing an initial license plate recognition operation on the encoded vehicle image to obtain the initial license plate recognition result of the target vehicle includes:

[0012] Perform a decoding operation on the encoded vehicle image through a preset decoder to obtain the decoded vehicle image and decoded metadata of the target vehicle;

[0013] Determine the vehicle rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image according to the decoded vehicle image and the decoded metadata;

[0014] Determine the license plate rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image according to the vehicle rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image;

[0015] Perform an initial license plate recognition operation on the decoded vehicle image according to the license plate rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image to obtain the initial license plate recognition result of the target vehicle.

[0016] As an optional implementation manner, in the first aspect of the present invention, the determining whether it is necessary to perform a geographic information assisted decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result includes:

[0017] Determine the clarity value of the initial license plate recognition result, and determine whether the clarity value is less than a preset clarity threshold;

[0018] When it is determined that the clarity value is less than the clarity threshold, determine that it is necessary to perform a geographic information assisted decision-making operation on the license plate of the target vehicle; or,

[0019] Determine the information loss situation of the initial license plate recognition result, and determine whether the information loss situation is a preset information loss situation;

[0020] When it is determined that the information loss situation is the preset information loss situation, it is determined that a geographic information assisted decision-making operation needs to be performed on the license plate of the target vehicle.

[0021] As an optional implementation manner, in the first aspect of the present invention, the performing a geographic information assisted decision-making operation on the license plate of the target vehicle to obtain the license plate geographic address information of the target vehicle includes:

[0022] Obtain the target parameters of the camera of the flying device; the target parameters of the camera include the internal parameters of the camera, the distortion parameters of the camera, and the physical size parameters of the camera sensor;

[0023] According to the field of view angle parameters of the pod and the physical size parameters of the camera sensor, calculate the focal length parameters of the camera, and perform a geodetic coordinate system conversion operation on the license plate rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image through the internal parameters of the camera, the distortion parameters of the camera, the focal length parameters of the camera, and the target metadata, to obtain the geodetic coordinate values corresponding to the license plate rectangular frame coordinate parameters;

[0024] Query the license plate geographic address information of the target vehicle according to the geodetic coordinate values corresponding to the license plate rectangular frame coordinate parameters.

[0025] As an optional implementation manner, in the first aspect of the present invention, the determining the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographic address information includes:

[0026] Determine the license plate administrative region address information of the target vehicle according to the license plate geographic address information;

[0027] Determine the license plate administrative region information to be compared of the target vehicle according to the license plate administrative region address information of the target vehicle;

[0028] Determine the administrative region associated image content corresponding to the initial license plate recognition result according to the initial license plate recognition result, and determine whether the license plate administrative region information to be compared matches the administrative region associated image content according to the administrative region associated image content and the license plate administrative region information to be compared;

[0029] When it is determined that the license plate administrative region information to be compared matches the administrative region associated image content, determine the target license plate number recognition result of the target vehicle according to the license plate administrative region information to be compared.

[0030] As an optional implementation manner, in the first aspect of the present invention, the method further includes:

[0031] When it is determined that the administrative region information of the license plate to be compared does not match the content of the administrative region associated image, determine, according to the administrative region information of the license plate to be compared, the adjacent license plate administrative region information that satisfies a preset geographical position relationship with the administrative region information of the license plate to be compared;

[0032] According to the content of the administrative region associated image and the adjacent license plate administrative region information, determine whether the adjacent license plate administrative region information matches the content of the administrative region associated image;

[0033] When it is determined that the adjacent license plate administrative region information matches the content of the administrative region associated image, determine, according to the adjacent license plate administrative region information, the recognition result of the target license plate number of the target vehicle;

[0034] When it is determined that the adjacent license plate administrative region information does not match the content of the administrative region associated image, update the adjacent license plate administrative region information according to the administrative region information of the license plate to be compared, and trigger the execution of the operation of determining whether the adjacent license plate administrative region information matches the content of the administrative region associated image according to the content of the administrative region associated image and the adjacent license plate administrative region information.

[0035] As an optional implementation manner, in the first aspect of the present invention, the focal length parameter of the camera is:

[0036]

[0037] wherein, w is the physical size parameter of the sensor of the camera, and AOV is the field of view angle parameter of the pod.

[0038] The second aspect of the present invention discloses a license plate recognition device for geographical information assisted decision-making, the device includes:

[0039] An acquisition module, configured to acquire a vehicle image of a target vehicle collected by a flight device;

[0040] A first determination module, configured to determine target metadata corresponding to the vehicle image;

[0041] An encoding module, configured to perform a map-data synchronous encoding operation on the vehicle image and the target metadata according to the vehicle image and the target metadata, to obtain an encoded vehicle image of the target vehicle;

[0042] A recognition module, configured to perform an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle;

[0043] A judgment module, configured to determine whether it is necessary to perform a geographical information assisted decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result;

[0044] An auxiliary decision-making module, configured to perform a geographic information auxiliary decision-making operation on the license plate of the target vehicle when the judgment result of the judgment module is yes, so as to obtain the license plate geographic address information of the target vehicle;

[0045] A second determination module, configured to determine the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographic address information.

[0046] As an optional implementation manner, in the second aspect of the present invention, the target metadata corresponding to the vehicle image includes the longitude and latitude parameters, altitude parameters, attitude angle parameters of the flying device, and the relative attitude angle parameters and field of view angle parameters of the pod of the flying device when the flying device captures the vehicle image;

[0047] Wherein, the manner in which the recognition module performs an initial license plate recognition operation on the encoded vehicle image to obtain the initial license plate recognition result of the target vehicle specifically includes:

[0048] Perform a decoding operation on the encoded vehicle image through a preset decoder to obtain the decoded vehicle image and decoded metadata of the target vehicle;

[0049] Determine the vehicle rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image according to the decoded vehicle image and the decoded metadata;

[0050] Determine the license plate rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image according to the vehicle rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image;

[0051] Perform an initial license plate recognition operation on the decoded vehicle image according to the license plate rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image to obtain the initial license plate recognition result of the target vehicle.

[0052] As an optional implementation manner, in the second aspect of the present invention, the manner in which the judgment module determines whether to perform a geographic information auxiliary decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result specifically includes:

[0053] Determine the clarity value of the initial license plate recognition result, and judge whether the clarity value is less than a preset clarity threshold;

[0054] When it is judged that the clarity value is less than the clarity threshold, it is determined that a geographic information auxiliary decision-making operation needs to be performed on the license plate of the target vehicle; or,

[0055] Determine the information loss situation of the initial license plate recognition result, and determine whether the information loss situation is a preset information loss situation;

[0056] When it is determined that the information loss situation is the preset information loss situation, determine that a geographic information assisted decision operation needs to be performed on the license plate of the target vehicle.

[0057] As an alternative implementation manner, in the second aspect of the present invention, the manner in which the assisted decision module performs a geographic information assisted decision operation on the license plate of the target vehicle to obtain the license plate geographic address information of the target vehicle specifically includes:

[0058] Obtain the target parameters of the camera of the flying device; the target parameters of the camera include the internal parameters of the camera, the distortion parameters of the camera, and the physical size parameters of the camera sensor;

[0059] According to the field of view angle parameter of the pod and the physical size parameter of the camera sensor, calculate the focal length parameter of the camera, and perform a geodetic coordinate system conversion operation on the license plate rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image through the internal parameters of the camera, the distortion parameters of the camera, the focal length parameter of the camera, and the target metadata, to obtain the geodetic coordinate values corresponding to the license plate rectangular frame coordinate parameters;

[0060] Query the license plate geographic address information of the target vehicle according to the geodetic coordinate values corresponding to the license plate rectangular frame coordinate parameters.

[0061] As an alternative implementation manner, in the second aspect of the present invention, the manner in which the second determination module determines the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographic address information specifically includes:

[0062] Determine the license plate administrative region address information of the target vehicle according to the license plate geographic address information;

[0063] Determine the license plate administrative region information to be compared of the target vehicle according to the license plate administrative region address information of the target vehicle;

[0064] Determine the administrative region associated image content corresponding to the initial license plate recognition result according to the initial license plate recognition result, and determine whether the license plate administrative region information to be compared matches the administrative region associated image content according to the administrative region associated image content and the license plate administrative region information to be compared;

[0065] When it is determined that the license plate administrative region information to be compared matches the administrative region associated image content, determine the target license plate number recognition result of the target vehicle according to the license plate administrative region information to be compared.

[0066] As an alternative implementation, in the second aspect of the present invention, the manner in which the second determination module determines the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographical address information specifically further includes:

[0067] When it is determined that the license plate administrative region information to be compared does not match the administrative region associated image content, according to the license plate administrative region information to be compared, determine the adjacent license plate administrative region information that satisfies a preset geographical position relationship with the license plate administrative region information to be compared;

[0068] According to the administrative region associated image content and the adjacent license plate administrative region information, determine whether the adjacent license plate administrative region information matches the administrative region associated image content;

[0069] When it is determined that the adjacent license plate administrative region information matches the administrative region associated image content, according to the adjacent license plate administrative region information, determine the target license plate number recognition result of the target vehicle;

[0070] When it is determined that the adjacent license plate administrative region information does not match the administrative region associated image content, update the adjacent license plate administrative region information according to the license plate administrative region information to be compared, and trigger the execution of the operation of determining whether the adjacent license plate administrative region information matches the administrative region associated image content according to the administrative region associated image content and the adjacent license plate administrative region information.

[0071] As an alternative implementation, in the second aspect of the present invention, the focal length parameter of the camera is:

[0072]

[0073] Wherein, w is the physical size parameter of the camera sensor, and AOV is the field of view angle parameter of the pod.

[0074] The third aspect of the present invention discloses another license plate recognition device for geographical information assisted decision-making, and the device includes:

[0075] A memory storing executable program code;

[0076] A processor coupled to the memory;

[0077] The processor calls the executable program code stored in the memory and executes the license plate recognition method for geographical information assisted decision-making disclosed in the first aspect of the present invention.

[0078] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute the license plate recognition method for geographic information assisted decision-making disclosed in the first aspect of the present invention.

[0079] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0080] In the embodiments of the present invention, an operation of synchronizing image and data encoding is performed on the vehicle image and target metadata of the target vehicle collected by the flying device to obtain an encoded vehicle image of the target vehicle; an initial license plate recognition operation is performed on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle, and it is determined whether it is necessary to perform a geographic information assisted decision-making operation on the license plate of the target vehicle; if so, a geographic information assisted decision-making operation is performed on the license plate of the target vehicle to obtain the geographic address information of the license plate of the target vehicle, so as to determine the target license plate number recognition result of the target vehicle. It can be seen that implementing the present invention can combine the image recognition result with geographic analysis to accurately lock the geographic address of the recognition target when the license plate image recognition effect is not ideal, effectively improving the recognition reliability and accuracy of vehicle license plates. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0082] Figure 1 is a schematic flowchart of a license plate recognition method for geographic information assisted decision-making disclosed in an embodiment of the present invention;

[0083] Figure 2 is a schematic flowchart of another license plate recognition method for geographic information assisted decision-making disclosed in an embodiment of the present invention;

[0084] Figure 3 is a schematic structural diagram of a license plate recognition device for geographic information assisted decision-making disclosed in an embodiment of the present invention;

[0085] Figure 4 is a schematic structural diagram of another license plate recognition device for geographic information assisted decision-making disclosed in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0086] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solution in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work belong to the scope of protection of the present invention.

[0087] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product or terminal comprising a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or terminals.

[0088] Referring to "embodiment" herein means that a specific feature, structure or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0089] The present invention discloses a license plate recognition method and device assisted by geographic information for decision-making, which can combine the image recognition result with geographic analysis to accurately lock the geographic address of the recognition target when the recognition effect of the license plate image is not ideal, effectively improving the recognition reliability and accuracy of vehicle license plates.

[0090] Embodiment 1

[0091] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a license plate recognition method assisted by geographic information for decision-making disclosed in the embodiments of the present invention. Optionally, this method can be implemented by a license plate recognition device, which can be integrated in a license plate recognition device (such as an airborne edge computing device, a smart computer, a tablet, a smart phone, etc.), or can be a local server or a cloud server for processing the prediction model training process, etc. The embodiments of the present invention do not make limitations. As Figure 1 shown, the license plate recognition method assisted by geographic information for decision-making may include the following operations:

[0092] 101. Obtain the vehicle image of the target vehicle collected by the flight device and determine the target metadata corresponding to the vehicle image.

[0093] In an embodiment of the present invention, further, the target metadata corresponding to the vehicle image includes the longitude and latitude parameters, altitude parameters, attitude angle parameters of the flight device (such as the pitch angle, yaw angle, and roll angle parameters of a drone) when the vehicle image is collected by the flight device, as well as the relative attitude angle parameters of the pod of the flight device (such as the pitch angle, yaw angle, and roll angle parameters of the pod relative to the drone), and the field of view angle parameters. Optionally, the flight device may include a drone, an airplane, a helicopter, etc., and the vehicle image of the target vehicle may be determined from the real-time video stream, offline video, or single picture of the target vehicle collected by the flight device.

[0094] 102. Perform a map-data synchronization encoding operation on the vehicle image and the target metadata to obtain an encoded vehicle image of the target vehicle.

[0095] In an embodiment of the present invention, the map-data synchronization encoding operation can ensure the synchronization of the vehicle image data and the target metadata. After obtaining the encoded vehicle image of the target vehicle, based on the transmission protocol, the encoded vehicle image can be transmitted to the AI recognition and detection module in the license plate recognition device, so that the AI recognition and detection module performs an initial license plate recognition on the encoded vehicle image.

[0096] 103. Perform an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle, and determine whether it is necessary to perform a geographical information assisted decision operation on the license plate of the target vehicle according to the initial license plate recognition result.

[0097] In an embodiment of the present invention, that is, according to the image features, perform an initial license plate recognition operation on the encoded vehicle image to obtain the coordinate parameters of the vehicle and the license plate rectangular frame of the target vehicle. Further, determining whether it is necessary to perform a geographical information assisted decision operation on the license plate of the target vehicle according to the initial license plate recognition result can be understood as when the initial license plate recognition effect is not ideal, such as when the encoded vehicle image is too blurred, the recognition error rate is large, or the first two digits of the license plate number in the encoded vehicle image are blocked and the specific content cannot be recognized, then it is determined that it is necessary to perform a geographical information assisted decision operation on the license plate of the target vehicle.

[0098] 104. When the judgment result is yes, perform a geographical information assisted decision operation on the license plate of the target vehicle to obtain the geographical address information of the license plate of the target vehicle, and determine the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the geographical address information of the license plate.

[0099] In the embodiments of the present invention, it can be understood that first, the geographical coordinates of the target vehicle are determined through the initial license plate recognition result, and then the geographical coding service is queried according to the geographical coordinates to obtain the license plate geographical address information of the target vehicle. Thus, without the training and inference of the neural network, the AI recognition auxiliary decision can be made by combining the license plate geographical address information, and the final target recognition result, that is, the target license plate number recognition result of the target vehicle, can be output.

[0100] It can be seen that implementing the embodiments of the present invention can perform map-data synchronous encoding on the vehicle image and target metadata of the target vehicle collected by the flight device, and then perform an initial license plate recognition operation on the encoded vehicle image of the target vehicle to obtain the initial license plate recognition result of the target vehicle. Thus, in the case where the initial license plate recognition effect is not ideal, the initial license plate recognition result and the determined license plate geographical address information are further combined to determine the target license plate number recognition result of the target vehicle. In this way, by combining the image recognition result with geographical analysis, the geographical address of the recognition target can be accurately locked, effectively improving the recognition reliability and accuracy of the vehicle license plate, and thus meeting the actual needs of fields such as traffic management and security monitoring.

[0101] Embodiment 2

[0102] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of another license plate recognition method assisted by geographical information disclosed in the embodiments of the present invention. Optionally, this method can be implemented by a license plate recognition device, which can be integrated in a license plate recognition device (such as an airborne edge computing device, a smart computer, a tablet, a smart phone, etc.), or can be a local server or a cloud server for processing the prediction model training process, etc. The embodiments of the present invention do not make limitations. As Figure 2 shown, the license plate recognition method assisted by geographical information can include the following operations:

[0103] 201. Obtain the vehicle image of the target vehicle collected by the flight device, and determine the target metadata corresponding to the vehicle image.

[0104] 202. According to the vehicle image and the target metadata, perform a map-data synchronous encoding operation on the vehicle image and the target metadata to obtain the encoded vehicle image of the target vehicle.

[0105] 203. Through a preset decoder, perform a decoding operation on the encoded vehicle image to obtain the decoded vehicle image and decoded metadata of the target vehicle.

[0106] In an embodiment of the present invention, the AI recognition and detection module in the license plate recognition device (this module can be deployed on an airborne edge computing device, such as NVIDIA TX2 or NVIDIA NX, etc., or can also be deployed on a cloud server with an NVIDIA graphics card. Among them, if it is deployed on an airborne edge computing device, the image data can be directly obtained from the image acquisition module in the flight device; if it is deployed on a cloud server, the image acquisition module can transmit the image data to the cloud server through network streaming) can perform a decoding operation on the encoded vehicle image through an FMV decoder.

[0107] 204. Determine the vehicle rectangle frame coordinate parameters of the target vehicle in the decoded vehicle image according to the decoded vehicle image and the decoded metadata.

[0108] In an embodiment of the present invention, the vehicle rectangle frame coordinate parameters of the target vehicle in the decoded vehicle image can be understood as using a deep neural network model and selecting YOLO as the detection framework to output one or more of the upper left corner coordinate parameters, upper right corner coordinate parameters, lower left corner coordinate parameters, and lower right corner coordinate parameters of the vehicle rectangle frame containing the target vehicle.

[0109] 205. Determine the license plate rectangle frame coordinate parameters of the target vehicle in the decoded vehicle image according to the vehicle rectangle frame coordinate parameters of the target vehicle in the decoded vehicle image.

[0110] In an embodiment of the present invention, the vehicle rectangle frame coordinate parameters of the target vehicle in the decoded vehicle image can be understood as using the RetinaNet detection framework or other lightweight single-object detection frameworks to output one or more of the upper left corner coordinate parameters, upper right corner coordinate parameters, lower left corner coordinate parameters, and lower right corner coordinate parameters of the license plate rectangle frame containing the license plate of the target vehicle.

[0111] 206. Perform an initial license plate recognition operation on the decoded vehicle image according to the license plate rectangle frame coordinate parameters of the target vehicle in the decoded vehicle image to obtain the initial license plate recognition result of the target vehicle, and judge whether it is necessary to perform a geographical information assisted decision operation on the license plate of the target vehicle according to the initial license plate recognition result.

[0112] In an embodiment of the present invention, among them, the initial license plate recognition result of the target vehicle may include the Chinese, English, numbers, etc. recognized in the decoded vehicle image.

[0113] 207. When the judgment result is yes, perform a geographical information assisted decision operation on the license plate of the target vehicle to obtain the geographical address information of the license plate of the target vehicle, and determine the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the geographical address information of the license plate.

[0114] In the embodiments of the present invention, for other descriptions of steps 205 and 206, please refer to the detailed descriptions of steps 102 and 103 in Embodiment 1, which will not be repeated in the embodiments of the present invention.

[0115] It can be seen that implementing the embodiments of the present invention can decode the decoded vehicle image of the target vehicle through a preset decoder, and then determine the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image. Thus, according to the coordinate parameters of the license plate rectangular frame, an initial license plate recognition operation is performed on the decoded vehicle image to evaluate the initial license plate recognition effect of the target vehicle. In this way, through the geographic information assisted decision-making operation, combining the initial license plate recognition result and the license plate geographic address information, the license plate number of the target vehicle can be determined more accurately, improving the scene applicability of the license plate recognition technology, such as traffic law enforcement scenarios, vehicle tracking scenarios, etc., thereby significantly enhancing the practical value of the license plate recognition technology. At the same time, the accuracy of AI recognition can be improved through geographic algorithms without the need for training and inference of neural networks, effectively enhancing the license plate recognition efficiency, and thus enhancing the overall user experience.

[0116] In an alternative embodiment, the above-mentioned step 206 of determining whether to perform a geographic information assisted decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result includes:

[0117] Determine the clarity value of the initial license plate recognition result, and determine whether the clarity value is less than a preset clarity threshold;

[0118] When it is determined that the clarity value is less than the clarity threshold, it is determined that a geographic information assisted decision-making operation needs to be performed on the license plate of the target vehicle; or,

[0119] Determine the information loss situation of the initial license plate recognition result, and determine whether the information loss situation is a preset information loss situation;

[0120] When it is determined that the information loss situation is a preset information loss situation, it is determined that a geographic information assisted decision-making operation needs to be performed on the license plate of the target vehicle.

[0121] In this alternative embodiment, the preset information loss situation can be understood as the loss of one or two characters in the first two characters of the initial license plate recognition result (such as the loss of "Yue J", "Yue", or "J" in the license plate image "Yue J11111"). Further, if it is determined that the clarity value is greater than or equal to the clarity threshold, or the information loss situation is not the preset information loss situation, the initial license plate recognition result can be directly determined as the target license plate number recognition result.

[0122] It can be seen that when the clarity value of the initial license plate recognition result is too low or the preset information is lost, the optional embodiment can automatically perform a geographic information assisted decision on the license plate of the target vehicle to obtain the license plate geographic address information of the target vehicle. In this way, it helps to quickly identify potential problems in the initial license plate recognition result, trigger the geographic information assisted decision operation, improve the intelligent level and wide applicability of license plate recognition, reduce unnecessary calculations under high-quality images or complete recognition results, and optimize the utilization efficiency of computing resources.

[0123] In another optional embodiment, the operation of performing a geographic information assisted decision on the license plate of the target vehicle in step 207 to obtain the license plate geographic address information of the target vehicle includes:

[0124] Obtain the target parameters of the camera of the flying device; the target parameters of the camera include the internal parameters of the camera, the distortion parameters of the camera, and the physical size parameters of the camera sensor;

[0125] According to the field of view angle parameters of the pod and the physical size parameters of the camera sensor, calculate the focal length parameters of the camera, and perform a geodetic coordinate system conversion operation on the coordinate parameters of the license plate rectangle of the target vehicle in the decoded vehicle image through the internal parameters of the camera, the distortion parameters of the camera, the focal length parameters of the camera, and the target metadata to obtain the geodetic coordinate values corresponding to the license plate rectangle coordinate parameters;

[0126] Query the license plate geographic address information of the target vehicle according to the geodetic coordinate values corresponding to the license plate rectangle coordinate parameters.

[0127] In this optional embodiment, further, the focal length parameter of the camera is:

[0128]

[0129] where w is the physical size parameter of the camera sensor and AOV is the field of view angle parameter of the pod.

[0130] For example, first obtain the original image data of the input target vehicle, parse the POS data (latitude, longitude, altitude, and attitude angle of the aircraft) of the current image frame corresponding to the drone, the attitude angle of the pod relative to the drone, and the field of view angle as input parameters. Then obtain the internal parameters, distortion parameters, and physical size parameters of the camera, calculate the current focal length parameter of the camera, and based on the internal parameters, distortion parameters, and current focal length parameter of the camera, as well as the POS data of the drone, the attitude angle of the pod relative to the drone, and the field of view angle, and through the digital surface model DSM and the collinearity equation, convert the coordinate parameters of the license plate rectangle to obtain its geodetic coordinate values, so as to determine the license plate geographic address information of the target vehicle through geocoding services and GIS topological analysis processing.

[0131] It can be seen that this optional embodiment can calculate the focal length parameter of the camera according to the field of view angle parameter of the pod and the physical size parameter of the camera sensor, and then perform a geodetic coordinate system conversion operation on the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image through the internal parameters of the camera, the distortion parameters of the camera, the focal length parameter of the camera, and the target metadata, so as to query the license plate geographical address information of the target vehicle according to the obtained geodetic coordinate values. In this way, the reliability and accuracy of determining the geographical location information of the target vehicle can be improved, providing a reliable data basis for subsequent geographical information-assisted decision-making; in addition, through the seamless integration of geocoding services and GIS, manual intervention is reduced, the data processing efficiency in the license plate recognition process is improved, and the functions and application scope of the license plate recognition technology are expanded.

[0132] In another optional embodiment, determining the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographical address information in step 207 includes:

[0133] Determine the license plate administrative region address information of the target vehicle according to the license plate geographical address information;

[0134] Determine the license plate administrative region information to be compared of the target vehicle according to the license plate administrative region address information of the target vehicle;

[0135] Determine the administrative region-associated image content corresponding to the initial license plate recognition result according to the initial license plate recognition result, and judge whether the license plate administrative region information to be compared matches the administrative region-associated image content according to the administrative region-associated image content and the license plate administrative region information to be compared;

[0136] When it is determined that the license plate administrative region information to be compared matches the administrative region-associated image content, determine the target license plate number recognition result of the target vehicle according to the license plate administrative region information to be compared.

[0137] In this optional embodiment, further, the method further includes:

[0138] When it is determined that the license plate administrative region information to be compared does not match the administrative region-associated image content, determine the adjacent license plate administrative region information that satisfies the preset geographical location relationship with the license plate administrative region information to be compared according to the license plate administrative region information to be compared;

[0139] Judge whether the adjacent license plate administrative region information matches the administrative region-associated image content according to the administrative region-associated image content and the adjacent license plate administrative region information;

[0140] When it is determined that the adjacent license plate administrative region information matches the administrative region-associated image content, determine the target license plate number recognition result of the target vehicle according to the adjacent license plate administrative region information;

[0141] When it is determined that the administrative district information of the adjacent license plate does not match the administrative district associated image content, the administrative district information of the adjacent license plate is updated according to the administrative district information of the license plate to be compared, and the operation of determining whether the administrative district information of the adjacent license plate matches the administrative district associated image content and the administrative district information of the adjacent license plate is triggered.

[0142] For example, if the license plate geographical address information is used to determine that the target vehicle's license plate belongs to Guangdong Province, that is, the first character of its license plate is "粤" (that is, the administrative region information of the target vehicle's license plate to be compared), and then combined with the initial license plate recognition result, it is determined that the administrative region information of the license plate to be compared matches the content of the administrative region-related image (such as the initial license plate recognition result actually contains a relatively vague "粤" character), then the first character of the license plate of the target vehicle can be determined to be "粤", thereby determining the complete license plate number; and if combined with the initial license plate recognition result, it is determined that the administrative region information of the license plate to be compared does not match the content of the administrative region-related image (such as the initial license plate recognition result actually contains a relatively vague "湘" character), then the abbreviation of the administrative region bordering Guangdong Province can be determined first, and then compared with the initial license plate recognition result again, until the polling of the administrative region information of the adjacent license plate matches the content of the administrative region-related image. Similarly, the same is true for the comparison process of administrative region information such as prefecture-level cities and autonomous regions of license plates.

[0143] It can be seen that this optional embodiment is able to compare the administrative district information of adjacent license plates with the administrative district-associated image content, and when the two do not match, further compare the administrative district information of adjacent license plates with the administrative district-associated image content until the matching administrative district information is polled. In this way, the granularity of administrative district information matching can be refined, and the intelligence level of the license plate recognition system can be improved, thereby improving the reliability and accuracy of license plate recognition of target vehicles; at the same time, it can be flexibly applied to administrative district information matching in different regions and at different levels, and thus can adapt to a variety of application scenarios (such as cross-provincial vehicle tracking, urban traffic management, etc.), thereby significantly improving the user experience.

[0144] Embodiment 3

[0145] See also Figure 3 , Figure 3 Schematic diagram of the structure of a license plate recognition device for geographic information-assisted decision-making disclosed in an embodiment of the present invention. Figure 3 As shown, the license plate recognition device for geographic information-assisted decision-making may include:

[0146] An acquisition module 301 is used to acquire a vehicle image of a target vehicle collected by a flight device;

[0147] The first determination module 302 is configured to determine target metadata corresponding to the vehicle image;

[0148] The encoding module 303 is configured to perform a map-data synchronous encoding operation on the vehicle image and the target metadata according to the vehicle image and the target metadata, so as to obtain an encoded vehicle image of the target vehicle;

[0149] The recognition module 304 is configured to perform an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle;

[0150] The judgment module 305 is configured to judge whether a geographic information assisted decision operation needs to be performed on the license plate of the target vehicle according to the initial license plate recognition result;

[0151] The assisted decision module 306 is configured to perform a geographic information assisted decision operation on the license plate of the target vehicle when the judgment result of the judgment module 305 is yes, so as to obtain the license plate geographic address information of the target vehicle;

[0152] The second determination module 307 is configured to determine a target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographic address information.

[0153] In an embodiment of the present invention, the target metadata corresponding to the vehicle image includes the longitude and latitude parameters, altitude parameters, attitude angle parameters of the flight device, the relative attitude angle parameters and the field of view angle parameters of the pod of the flight device when the flight device captures the vehicle image.

[0154] It can be seen that implementing Figure 3 The described license plate recognition device with geographic information assisted decision can perform map-data synchronous encoding on the vehicle image and the target metadata of the target vehicle captured by the flight device, and then perform an initial license plate recognition operation on the encoded vehicle image of the target vehicle to obtain an initial license plate recognition result of the target vehicle. Thus, in the case where the initial license plate recognition effect is not ideal, the initial license plate recognition result is further combined with the determined license plate geographic address information to determine the target license plate number recognition result of the target vehicle. In this way, by combining the image recognition result with geographic analysis, the recognition target can be accurately locked by geographic address, effectively improving the recognition reliability and accuracy of the vehicle license plate, so as to meet the actual needs of traffic management and security monitoring and other fields.

[0155] In an optional embodiment, the manner in which the recognition module 304 performs an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle specifically includes:

[0156] Through a preset decoder, perform a decoding operation on the encoded vehicle image to obtain a decoded vehicle image and decoded metadata of the target vehicle;

[0157] Based on the decoded vehicle image and the decoded metadata, determine the coordinate parameters of the vehicle rectangular frame of the target vehicle in the decoded vehicle image;

[0158] Based on the coordinate parameters of the vehicle rectangular frame of the target vehicle in the decoded vehicle image, determine the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image;

[0159] Based on the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image, perform an initial license plate recognition operation on the decoded vehicle image to obtain the initial license plate recognition result of the target vehicle.

[0160] It can be seen that implementing Figure 3 The license plate recognition device for geographic information-assisted decision-making described above can decode the decoded vehicle image of the target vehicle through a preset decoder, and then determine the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image. Thus, based on the coordinate parameters of the license plate rectangular frame, an initial license plate recognition operation is performed on the decoded vehicle image to evaluate the initial license plate recognition effect of the target vehicle. In this way, through geographic information-assisted decision-making operations, combined with the initial license plate recognition result and the license plate geographic address information, the license plate number of the target vehicle can be determined more accurately, improving the scene applicability of the license plate recognition technology, such as traffic law enforcement scenarios, vehicle tracking scenarios, etc., thereby significantly enhancing the practical value of the license plate recognition technology; at the same time, the accuracy of AI recognition can be improved through geographic algorithms without the need for neural network training and inference, effectively improving the license plate recognition efficiency, and thus enhancing the overall user experience.

[0161] In another alternative embodiment, the manner in which the determination module 305 determines whether to perform a geographic information-assisted decision-making operation on the license plate of the target vehicle based on the initial license plate recognition result specifically includes:

[0162] Determine the clarity value of the initial license plate recognition result, and determine whether the clarity value is less than a preset clarity threshold;

[0163] When it is determined that the clarity value is less than the clarity threshold, it is determined that a geographic information-assisted decision-making operation needs to be performed on the license plate of the target vehicle; or,

[0164] Determine the information loss situation of the initial license plate recognition result, and determine whether the information loss situation is a preset information loss situation;

[0165] When it is determined that the information loss situation is a preset information loss situation, it is determined that a geographic information-assisted decision-making operation needs to be performed on the license plate of the target vehicle.

[0166] It can be seen that implementing Figure 3The described license plate recognition device assisted by geographic information can automatically perform geographic information assisted decision-making on the license plate of the target vehicle when the clarity value of the initial license plate recognition result is too low or there is a situation of preset information loss, so as to obtain the license plate geographic address information of the target vehicle. In this way, it helps to quickly identify potential problems in the initial license plate recognition result and trigger the geographic information assisted decision-making operation, improving the intelligent level and wide applicability of license plate recognition. It also reduces unnecessary calculations under high-quality images or complete recognition results and optimizes the utilization efficiency of calculation data.

[0167] In another optional embodiment, the manner in which the auxiliary decision-making module 306 performs geographic information assisted decision-making on the license plate of the target vehicle to obtain the license plate geographic address information of the target vehicle specifically includes:

[0168] Obtain the target parameters of the camera of the flying device; the target parameters of the camera include the internal parameters of the camera, the distortion parameters of the camera, and the physical size parameters of the camera sensor;

[0169] According to the field of view angle parameter of the pod and the physical size parameter of the camera sensor, calculate the focal length parameter of the camera, and perform a geodetic coordinate system conversion operation on the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image through the internal parameters of the camera, the distortion parameters of the camera, the focal length parameter of the camera, and the target metadata, to obtain the geodetic coordinate value corresponding to the license plate rectangular frame coordinate parameters;

[0170] Query the license plate geographic address information of the target vehicle according to the geodetic coordinate value corresponding to the license plate rectangular frame coordinate parameters.

[0171] In this optional embodiment, further, the focal length parameter of the camera is:

[0172]

[0173] where w is the physical size parameter of the camera sensor and AOV is the field of view angle parameter of the pod.

[0174] It can be seen that the implementation Figure 3The described license plate recognition device assisted by geographic information can calculate the focal length parameter of the camera based on the field of view angle parameter of the pod and the sensor physical size parameter of the camera, and then perform a geodetic coordinate system conversion operation on the license plate rectangular frame coordinate parameter of the target vehicle in the decoded vehicle image through the internal parameter of the camera, the distortion parameter of the camera, the focal length parameter of the camera, and the target metadata, so as to query the license plate geographic address information of the target vehicle according to the obtained geodetic coordinate value. In this way, the reliability and accuracy of determining the geographic location information of the target vehicle can be improved, providing a reliable data basis for subsequent geographic information-assisted decision-making; in addition, through the seamless integration of geocoding services and GIS, manual intervention is reduced, the data processing efficiency in the license plate recognition process is improved, and the functions and application scope of the license plate recognition technology are expanded.

[0175] In yet another optional embodiment, the specific manner in which the second determination module 307 determines the target license plate number recognition result of the target vehicle based on the initial license plate recognition result and the license plate geographic address information includes:

[0176] Determine the license plate administrative region address information of the target vehicle according to the license plate geographic address information;

[0177] Determine the license plate administrative region information to be compared of the target vehicle according to the license plate administrative region address information of the target vehicle;

[0178] Determine the administrative region associated image content corresponding to the initial license plate recognition result according to the initial license plate recognition result, and determine whether the license plate administrative region information to be compared matches the administrative region associated image content according to the administrative region associated image content and the license plate administrative region information to be compared;

[0179] When it is determined that the license plate administrative region information to be compared matches the administrative region associated image content, determine the target license plate number recognition result of the target vehicle according to the license plate administrative region information to be compared.

[0180] In this optional embodiment, further, the specific manner in which the second determination module 307 determines the target license plate number recognition result of the target vehicle based on the initial license plate recognition result and the license plate geographic address information further includes:

[0181] When it is determined that the license plate administrative region information to be compared does not match the administrative region associated image content, determine the adjacent license plate administrative region information that satisfies the preset geographic location relationship with the license plate administrative region information to be compared according to the license plate administrative region information to be compared;

[0182] Determine whether the adjacent license plate administrative region information matches the administrative region associated image content according to the administrative region associated image content and the adjacent license plate administrative region information;

[0183] When it is determined that the adjacent license plate administrative region information matches the administrative region associated image content, determine the target license plate number recognition result of the target vehicle according to the adjacent license plate administrative region information;

[0184] When it is determined that the adjacent license plate administrative region information does not match the administrative region associated image content, update the adjacent license plate administrative region information according to the license plate administrative region information to be compared, and trigger the operation of determining whether the adjacent license plate administrative region information matches the administrative region associated image content according to the administrative region associated image content and the adjacent license plate administrative region information.

[0185] It can be seen that implementing Figure 3 The described license plate recognition device for geographic information assisted decision-making can compare the adjacent license plate administrative region information with the administrative region associated image content, and when the two do not match, further compare the adjacent license plate administrative region information with the administrative region associated image content until the matching administrative region information is polled. In this way, the granularity of administrative region information matching can be refined, the intelligent level of the license plate recognition system can be improved, thereby improving the reliability and accuracy of license plate recognition for the target vehicle; at the same time, it can be flexibly applied to administrative region information matching in different regions and different levels, and then can adapt to diverse application scenarios (such as cross-province / city vehicle tracking, urban traffic management, etc.), thereby significantly enhancing the user experience.

[0186] Embodiment Four

[0187] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of another license plate recognition device for geographic information assisted decision-making disclosed in an embodiment of the present invention. As Figure 4 shown, the license plate recognition device for geographic information assisted decision-making may include:

[0188] A memory 401 storing executable program code;

[0189] A processor 402 coupled to the memory 401;

[0190] The processor 402 calls the executable program code stored in the memory 401 and executes the steps in the license plate recognition method for geographic information assisted decision-making described in Embodiment One or Embodiment Two of the present invention.

[0191] Embodiment Five

[0192] An embodiment of the present invention discloses a computer storage medium, which stores computer instructions that, when called, are used to execute the steps in the license plate recognition method for geographic information assisted decision-making described in Embodiment One or Embodiment Two of the present invention.

[0193] Embodiment Six

[0194] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute the steps in the license plate recognition method for geographic information assisted decision-making described in Embodiment 1 or Embodiment 2.

[0195] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0196] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation manner can be realized by means of software plus a necessary general hardware platform, and of course, it can also be realized by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disk memories, tape memories, or any other computer-readable medium capable of carrying or storing data.

[0197] Finally, it should be noted that: What is disclosed by a license plate recognition method and device for geographic information-assisted decision-making disclosed in the embodiments of the present invention is only the preferred embodiments of the present invention, which are only used to illustrate the technical solutions of the present invention, rather than to limit it; Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: They can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A license plate recognition method for geographic information-assisted decision-making, characterized in that: The method comprises: Acquire a vehicle image of a target vehicle captured by the flight equipment, and determine target metadata corresponding to the vehicle image; According to the vehicle image and the target metadata, performing an image-data synchronization encoding operation on the vehicle image and the target metadata to obtain an encoded vehicle image of the target vehicle; Performing an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle, and judging whether it is necessary to perform a geographic information-assisted decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result; When the judgment result is yes, a geographic information assisted decision-making operation is performed on the license plate of the target vehicle to obtain the geographic address information of the license plate of the target vehicle, and the target license plate number recognition result of the target vehicle is determined based on the initial license plate recognition result and the license plate geographic address information.

2. The license plate recognition method for geographic information-assisted decision-making according to claim 1 is characterized in that: The target metadata corresponding to the vehicle image includes the latitude and longitude parameters, altitude parameters, attitude angle parameters of the flight device, and the relative attitude angle parameters and field of view angle parameters of the pod of the flight device when the flight device collects the vehicle image; The performing of an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle includes: Decoding the encoded vehicle image through a preset decoder to obtain a decoded vehicle image and decoded metadata of the target vehicle; Determining vehicle rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image according to the decoded vehicle image and the decoded metadata; Determining the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image according to the vehicle rectangular frame coordinate parameters of the target vehicle in the decoded vehicle image; According to the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image, an initial license plate recognition operation is performed on the decoded vehicle image to obtain an initial license plate recognition result of the target vehicle.

3. The license plate recognition method for geographic information-assisted decision-making according to claim 2 is characterized in that: The step of judging whether it is necessary to perform a geographic information-assisted decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result includes: Determining a clarity value of the initial license plate recognition result, and judging whether the clarity value is less than a preset clarity threshold; When it is determined that the clarity value is less than the clarity threshold, it is determined that a geographic information-assisted decision-making operation needs to be performed on the license plate of the target vehicle; or, Determining the information loss situation of the initial license plate recognition result, and judging whether the information loss situation is a preset information loss situation; When it is determined that the information loss situation is the preset information loss situation, it is determined that a geographic information assisted decision-making operation needs to be performed on the license plate of the target vehicle.

4. The license plate recognition method for geographic information-assisted decision-making according to claim 2 or 3, characterized in that: The performing of a geographic information assisted decision-making operation on the license plate of the target vehicle to obtain the geographic address information of the license plate of the target vehicle includes: Acquire target parameters of the camera of the flying device; the target parameters of the camera include intrinsic parameters of the camera, distortion parameters of the camera, and physical size parameters of the sensor of the camera; Calculating the focal length parameters of the camera according to the field of view angle parameters of the pod and the physical size parameters of the sensor of the camera, and performing a geodetic coordinate system conversion operation on the coordinate parameters of the license plate rectangular frame of the target vehicle in the decoded vehicle image through the intrinsic parameters of the camera, the distortion parameters of the camera, the focal length parameters of the camera and the target metadata, to obtain the geodetic coordinate values ​​corresponding to the coordinate parameters of the license plate rectangular frame; The geographic address information of the license plate of the target vehicle is queried according to the geodetic coordinate values ​​corresponding to the coordinate parameters of the license plate rectangular frame.

5. The license plate recognition method for geographic information-assisted decision-making according to claim 4 is characterized in that: Determining the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographic address information includes: Determining the license plate administrative district address information of the target vehicle according to the license plate geographical address information; Determine the license plate administrative area information to be compared of the target vehicle according to the license plate administrative area address information of the target vehicle; Determine, according to the initial license plate recognition result, the administrative district associated image content corresponding to the initial license plate recognition result, and determine, according to the administrative district associated image content and the administrative district information of the license plate to be compared, whether the administrative district information of the license plate to be compared matches the administrative district associated image content; When it is determined that the administrative area information of the license plate to be compared matches the content of the administrative area-associated image, a target license plate number recognition result of the target vehicle is determined based on the administrative area information of the license plate to be compared.

6. The license plate recognition method for geographic information-assisted decision-making according to claim 5 is characterized in that: The method further comprises: When it is determined that the license plate administrative district information to be compared does not match the administrative district associated image content, determining, based on the license plate administrative district information to be compared, adjacent license plate administrative district information that satisfies a preset geographical location relationship with the license plate administrative district information to be compared; According to the administrative district-associated image content and the adjacent license plate administrative district information, determining whether the adjacent license plate administrative district information matches the administrative district-associated image content; When it is determined that the adjacent license plate administrative district information matches the administrative district associated image content, determining a target license plate number recognition result of the target vehicle according to the adjacent license plate administrative district information; When it is determined that the adjacent license plate administrative district information does not match the administrative district associated image content, the adjacent license plate administrative district information is updated according to the administrative district information of the license plate to be compared, and the operation of determining whether the adjacent license plate administrative district information matches the administrative district associated image content and the adjacent license plate administrative district information is triggered.

7. The license plate recognition method for geographic information-assisted decision-making according to claim 4 is characterized in that: The focal length parameters of the camera are: Wherein, w is the physical size parameter of the camera sensor, and AOV is the field of view angle parameter of the pod.

8. A license plate recognition device for geographic information-assisted decision-making, characterized in that: The device comprises: An acquisition module, used to acquire a vehicle image of a target vehicle collected by the flight equipment; A first determination module, used to determine target metadata corresponding to the vehicle image; An encoding module, configured to perform an image-data synchronization encoding operation on the vehicle image and the target metadata according to the vehicle image and the target metadata, so as to obtain an encoded vehicle image of the target vehicle; A recognition module, used to perform an initial license plate recognition operation on the encoded vehicle image to obtain an initial license plate recognition result of the target vehicle; A judgment module, used to judge whether it is necessary to perform a geographic information-assisted decision-making operation on the license plate of the target vehicle according to the initial license plate recognition result; An auxiliary decision-making module, used for performing a geographic information auxiliary decision-making operation on the license plate of the target vehicle to obtain the geographic address information of the license plate of the target vehicle when the judgment result of the judgment module is yes; The second determination module is used to determine the target license plate number recognition result of the target vehicle according to the initial license plate recognition result and the license plate geographical address information.

9. A license plate recognition device for geographic information-assisted decision-making, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the license plate recognition method for geographic information assisted decision-making as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer instructions, which, when called, are used to execute the license plate recognition method for geographic information-assisted decision-making as described in any one of claims 1-7.

Images