Positioning method, storage medium, program product, cloud server and electronic device

Abstract

The present disclosure relates to a positioning method, a storage medium, a program product, a cloud server and an electronic device, and relates to the technical field of positioning. The method can comprise: receiving video data sent by a second device and reference positioning information of the second device; extracting a plurality of key frame images in the video data; obtaining a plurality of position feature information according to the reference positioning information; and determining target positioning information corresponding to the second device according to the plurality of key frame images and the plurality of position feature information. Through multi-link matching confirmation, more accurate positioning information can be obtained, the accuracy of positioning the second device is improved, and in subsequent application scenarios, more accurate target positioning information can be used to provide corresponding services for passengers, thereby improving the use experience of the passengers.

Description

Technical Field

[0001] This disclosure relates to the field of positioning technology, and more specifically, to a positioning method, storage medium, program product, cloud server, and electronic device. Background Technology

[0002] Mobile ride-hailing services play a vital role in modern transportation, with an increasing number of people using them for travel. Currently, determining a user's pick-up point in mobile ride-hailing services is primarily achieved through the phone's GPS (Global Positioning System) or by having the user manually input their location.

[0003] However, on the one hand, GPS positioning is not accurate enough and may have large errors in certain environments, affecting the user's ride-hailing experience; on the other hand, the user's manual input of the location is highly dependent on the user's familiarity with the current environment. When the user is not familiar with the environment, they will not be able to correctly determine the current pick-up point. Summary of the Invention

[0004] To address the aforementioned technical problems, this disclosure provides a positioning method, storage medium, program product, cloud server, and electronic device.

[0005] In a first aspect, this disclosure provides a positioning method applied to a first device, the method comprising:

[0006] Receive video data sent by the second device and the reference positioning information of the second device;

[0007] Extract multiple keyframe images from the video data;

[0008] Based on the reference positioning information, multiple location feature information is obtained;

[0009] Based on the multiple keyframe images and the multiple location feature information, the target positioning information corresponding to the second device is determined.

[0010] Optionally, the location feature information includes a target positioning image; obtaining multiple location feature information based on the reference positioning information includes:

[0011] Based on the reference positioning information, multiple target positioning images are determined from a preset positioning database; wherein, the preset positioning database includes multiple positioning images.

[0012] Optionally, the preset positioning database further includes image positioning information corresponding to each positioning image, and determining multiple target positioning images from the preset positioning database based on the reference positioning information includes:

[0013] The positioning image within the first positioning range of the image positioning information is taken as the target positioning image; wherein, the first positioning range is the positioning range determined with the position corresponding to the reference positioning information as the center.

[0014] Optionally, determining the target positioning information corresponding to the second device based on the plurality of keyframe images and the plurality of location feature information includes:

[0015] Each keyframe image is matched with the plurality of target positioning images, and the successfully matched keyframe image and target positioning image are taken as an image pair;

[0016] The target location information is determined based on the image pair.

[0017] Optionally, matching each of the keyframe images with the plurality of target localization images includes:

[0018] Identify one or more key feature points in each of the keyframe images;

[0019] For each keyframe image, the keyframe image is matched with the plurality of target localization images based on one or more key feature points in the keyframe image.

[0020] Optionally, matching each keyframe image with the plurality of target localization images based on one or more key feature points in the keyframe image includes:

[0021] For each keyframe image, if the target positioning image contains key feature points from the keyframe image, then the target positioning image and the keyframe image are determined to be a successful match.

[0022] Optionally, determining the target location information based on the image pair includes:

[0023] For each image pair, determine the relative position information of the keyframe image in the image pair relative to the corresponding target positioning image; and determine the undetermined position corresponding to the keyframe image based on the relative position information and the image positioning information corresponding to the target positioning image.

[0024] The target positioning information is determined based on the undetermined position corresponding to the keyframe image in each image pair.

[0025] Optionally, determining the relative position information of the keyframe image in the image pair relative to the corresponding target positioning image includes:

[0026] The keyframe image is mapped onto the target plane where the corresponding target positioning image is located through affine transformation;

[0027] Based on the target plane, determine the distance difference between the center point of the keyframe image and the center point of the target positioning image;

[0028] Based on the distance difference, the relative position information of the keyframe image in the image pair with respect to the corresponding target positioning image is determined.

[0029] Optionally, when multiple image pairs are included, determining the target localization information based on the undetermined position corresponding to the keyframe image in each image pair includes:

[0030] Determine the preset weight corresponding to the target localization image in each image pair;

[0031] Based on the preset weights, a weighted average of multiple undetermined locations is determined to obtain the target positioning information.

[0032] Optionally, the location feature information includes target map feature information; obtaining multiple location feature information based on the reference positioning information includes:

[0033] Based on the reference positioning information, multiple target map feature information is determined from a preset high-precision map database; wherein, the preset high-precision map database includes multiple map feature information.

[0034] Optionally, the preset high-precision map database also includes map positioning information corresponding to each of the map feature information, and determining multiple target map feature information from the preset high-precision map database based on the reference positioning information includes:

[0035] The map feature information within the second positioning range of the map positioning information is used as the target map feature information; wherein, the second positioning range is the positioning range determined with the location corresponding to the reference positioning information as the center.

[0036] Optionally, determining the target positioning information corresponding to the second device based on the plurality of keyframe images and the plurality of location feature information includes:

[0037] Feature extraction is performed on each of the keyframe images to obtain the image feature information corresponding to the keyframe image;

[0038] Each of the image feature information is matched with the plurality of target map feature information, and the successfully matched image feature information and target map feature information are taken as a feature pair;

[0039] The target location information is determined based on the feature pairs.

[0040] Optionally, matching each of the image feature information with the plurality of target map feature information includes:

[0041] For each of the image feature information, determine the feature similarity between the image feature information and each of the target map feature information;

[0042] If the feature similarity is greater than or equal to a preset similarity threshold, it is determined that the image feature information and the target map feature information are successfully matched.

[0043] Optionally, the step of combining the successfully matched image feature information and the target map feature information as a feature pair includes:

[0044] The successfully matched image feature information and the target map feature information are treated as a pair of undetermined features;

[0045] Based on pre-defined road element features, the feature pair is determined from a plurality of undetermined feature pairs.

[0046] Optionally, the step of combining the successfully matched image feature information and the target map feature information as a feature pair includes:

[0047] The step of matching each image feature with the plurality of target map feature information is performed iteratively multiple times.

[0048] If the number of iterations reaches a preset number, the successfully matched image feature information and the target map feature information are treated as a feature pair.

[0049] Optionally, determining the target location information based on the feature pair includes:

[0050] Determine the map positioning information corresponding to the target map feature information in the feature pair;

[0051] The target location information is determined based on the map location information.

[0052] Optionally, extracting multiple keyframe images from the video data includes:

[0053] The multiple video frame images included in the video data are subjected to preset processing to obtain the multiple key frame images;

[0054] The preset processing includes one or more of noise reduction processing, jitter reduction processing, and deduplication processing.

[0055] Optionally, the method further includes:

[0056] The target location information is sent to a third device.

[0057] Secondly, this disclosure provides a positioning method applied to a second device, the method comprising:

[0058] Collect video data within the target area of ​​the environment where the second device is located;

[0059] Obtain the reference positioning information of the second device when it collects the video data;

[0060] The video data and the reference positioning information are sent to the first device so that the first device can extract multiple key frame images from the video data, obtain multiple location feature information based on the reference positioning information, and determine the target positioning information corresponding to the second device based on the multiple key frame images and the multiple location feature information.

[0061] Thirdly, this disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the positioning method provided in the first aspect of this disclosure, or implements the steps of the positioning method provided in the second aspect of this disclosure.

[0062] Fourthly, this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the positioning method provided in the first aspect of this disclosure, or the steps of the positioning method provided in the second aspect of this disclosure.

[0063] Fifthly, this disclosure provides a cloud server, comprising: a first memory storing a computer program thereon; and a first processor for executing the computer program in the memory to implement the steps of the positioning method provided in the first aspect of this disclosure.

[0064] In a sixth aspect, this disclosure provides an electronic device, comprising: a second memory storing a computer program thereon; and a second processor for executing the computer program in the memory to implement the steps of the positioning method provided in the second aspect of this disclosure.

[0065] This disclosure provides a positioning method. First, it receives video data and reference positioning information of the second device sent by a second device. Second, it extracts multiple keyframe images from the video data. Then, it obtains multiple location feature information based on the reference positioning information. Finally, it determines the target positioning information corresponding to the second device based on the multiple keyframe images and the multiple location feature information. Through this method, the second device sends video data and reference positioning information to a first device. The first device can obtain multiple location feature information based on the reference positioning information and, based on the multiple keyframe images and location feature information in the video data, obtain the target positioning information corresponding to the second device. By verifying through multiple steps, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0066] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0067] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings:

[0068] Figure 1 This is a flowchart illustrating a positioning method according to an exemplary embodiment.

[0069] Figure 2 This is a flowchart illustrating a positioning method according to an exemplary embodiment.

[0070] Figure 3 This is a flowchart illustrating a positioning method according to an exemplary embodiment.

[0071] Figure 4 This is a flowchart illustrating a positioning method according to an exemplary embodiment.

[0072] Figure 5 This is a flowchart illustrating a positioning method according to an exemplary embodiment.

[0073] Figure 6 This is a block diagram illustrating a positioning device according to an exemplary embodiment.

[0074] Figure 7 This is a block diagram illustrating a positioning device according to an exemplary embodiment.

[0075] Figure 8 This is a block diagram illustrating an electronic device according to an exemplary embodiment.

[0076] Figure 9 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0077] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0078] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily construed as referring to a specific order or sequence. Furthermore, in the description with reference to the accompanying drawings, the same reference numerals in different drawings denote the same elements.

[0079] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0080] In the description of this disclosure, unless otherwise stated, "multiple" means two or more, and other quantifiers are similar; "at least one," "one or more," or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one 'a' can represent any number of 'a's; as another example, one or more of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple; "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. The character " / " indicates that the preceding and following related objects are in an "or" relationship.

[0081] Although operations or steps are described in a specific order in the accompanying drawings in the embodiments of this disclosure, it should not be construed as requiring these operations or steps to be performed in the specific order or serial order shown, or requiring all of the shown operations or steps to be performed to obtain the desired result. In the embodiments of this disclosure, these operations or steps may be performed serially; they may be performed in parallel; or a portion of these operations or steps may be performed.

[0082] Before introducing the positioning method, storage medium, program product, cloud server, and electronic device provided in this disclosure, the application background of the various embodiments of this disclosure is first introduced. This disclosure is applied in the scenario of device positioning, in which device positioning is required to determine the user's pick-up point. Currently, this is mainly achieved through mobile phone GPS positioning or through the user manually inputting the location. This method depends on the user's current environmental location and the user's familiarity with the location. In addition, for the elderly and other people who are not familiar with mobile phone operating systems, this method requires more time to confirm the accuracy of the location, resulting in the following problems:

[0083] 1. Positioning error: GPS positioning may have significant errors in certain environments (such as urban centers with high-rise buildings, underground passages, etc.), which will affect the user's ride-hailing experience.

[0084] 2. POI (Point of Interest) Dependence: Ride-hailing apps usually automatically match nearby points of interest (POIs), but in some areas without obvious POIs, the positioning accuracy is low.

[0085] 3. User confusion: Users have difficulty determining their location on two-way roads or in complex environments, especially those who are unfamiliar with the location.

[0086] To address the aforementioned technical problems, this invention provides a positioning method, storage medium, program product, cloud server, and electronic device. A second device sends video data and reference positioning information to a first device. The first device can obtain multiple location feature information based on the reference positioning information, and obtain the target positioning information corresponding to the second device based on multiple keyframe images and location feature information in the video data. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0087] Secondly, this disclosure is further described in one application scenario: ride-hailing. In this scenario, passengers need to send their pick-up point to the driver so that the driver can pick them up. In this disclosure, passengers can collect video data using a second device and upload the video data and the corresponding reference location information to a first device. The first device determines the target location information of the second device based on the video data and the reference location information, and can send this target location information to a third device. The third device can then determine the passenger's pick-up point. In this case, the passenger does not need to upload their location themselves, thus providing a more convenient ride-hailing experience. The first device can be a cloud server or a vehicle, the second device can be an electronic device held by the passenger (passenger device), and the third device can be an electronic device held by the driver (driver device).

[0088] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0089] Figure 1 This is a flowchart illustrating a positioning method according to an exemplary embodiment, such as... Figure 1 As shown, applied to a first device, which may include a cloud server or a vehicle, the method may include the following steps:

[0090] In step S101, video data sent by the second device and the reference positioning information of the second device are received.

[0091] In this step, the second device refers to the electronic device held by the passenger, i.e., the passenger device. In a taxi-hailing scenario, the passenger can use the second device to capture video footage of their surroundings to obtain video data. The reference positioning information is the location information of the passenger when capturing the video data, obtained by the positioning device in the second device, such as GPS positioning information, BeiDou satellite positioning information, etc.

[0092] Understandably, both GPS and BeiDou satellite positioning information have relatively low positioning accuracy, especially in complex urban roads or underground passages, making it impossible to accurately determine the passenger's location. Therefore, in this embodiment, the location information of the second device is used as reference positioning information to assist in further determining the location of the second device.

[0093] In step S102, multiple keyframe images are extracted from the video data.

[0094] For example, multiple video frame images included in the video data can be subjected to preset processing to obtain multiple keyframe images. This preset processing includes one or more of denoising, jitter reduction, and deduplication.

[0095] Specifically, the deduplication process for multiple video frame images can be performed by determining the repetition rate between every two video frame images, and if the repetition rate exceeds the preset repetition rate, any one of the video frame images can be used as the key frame image.

[0096] In step S103, multiple location feature information is obtained based on the reference positioning information.

[0097] In one possible implementation, the location feature information may include target positioning images. Specifically, multiple target positioning images can be determined from a preset positioning database based on the reference positioning information; wherein the preset positioning database includes multiple positioning images.

[0098] In another possible implementation, the location feature information may include target map feature information. Specifically, multiple target map feature information can be determined from a pre-set high-precision map database based on the reference positioning information; wherein the pre-set high-precision map database includes multiple map feature information.

[0099] In step S104, the target positioning information corresponding to the second device is determined based on the multiple keyframe images and the multiple location feature information.

[0100] In some embodiments, by matching multiple keyframe images with multiple location feature information, one or more matching pairs are obtained, and the target positioning information corresponding to the second device is determined based on the one or more matching pairs.

[0101] For example, if the location feature information includes a target location image, the matching pair may include an image pair; if the location feature information includes target map feature information, the matching pair may include a feature pair.

[0102] Furthermore, after obtaining the target location information, the first device can dispatch an order to the passenger and send the target location information to the third device of the driver accepting the order. This allows the third device to determine the passenger's pick-up point based on the received target location information, thereby providing the passenger with a ride service. The third device refers to the electronic device held by the driver, i.e., the driver's device.

[0103] With the technical solution provided in this disclosure, in a ride-hailing scenario, passengers only need to capture video data and send the video data and reference positioning information to the first device, which then determines the target positioning information of the second device. Passengers only need to confirm the drop-off point and do not need to determine the pick-up point themselves.

[0104] Meanwhile, considering the elderly and those who have difficulty operating the second device, a quick ride-hailing service can also be provided, along with a large-font display and voice broadcast function. Passengers only need to upload video data, and the first device can send the determined target location information of the second device to the third device based on the video data and the reference location information of the second device. Passengers can then provide the drop-off point to the driver offline.

[0105] Using the above method, the second device sends video data and reference positioning information to the first device. The first device can then obtain multiple location feature information based on the reference positioning information, and, based on multiple keyframe images and location feature information in the video data, obtain the target positioning information corresponding to the second device. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0106] Figure 2 This is a flowchart illustrating a positioning method according to an exemplary embodiment, taking a target positioning image as an example, where the location feature information includes the target positioning image. Figure 2 As shown, the method may include the following steps:

[0107] In step S201, video data sent by the second device and the reference positioning information of the second device are received.

[0108] In step S202, multiple keyframe images are extracted from the video data.

[0109] For example, multiple video frame images included in the video data can be subjected to preset processing to obtain multiple keyframe images. This preset processing includes one or more of denoising, jitter reduction, and deduplication.

[0110] In step S203, multiple target positioning images are determined from a preset positioning database based on the reference positioning information.

[0111] The preset positioning database includes multiple positioning images and image positioning information corresponding to each positioning image.

[0112] For example, a positioning image within a first positioning range can be used as the target positioning image. Here, the first positioning range is the positioning range centered on the location corresponding to the reference positioning information.

[0113] In other words, in this step, the target positioning image that is near the reference positioning information can be selected from multiple positioning images based on the reference positioning information, so as to facilitate the subsequent determination of the target positioning information of the second device based on the target positioning image.

[0114] In step S204, each keyframe image is matched with the plurality of target positioning images, and the successfully matched keyframe image and target positioning image are taken as an image pair.

[0115] In some embodiments, firstly, one or more key feature points in each keyframe image may be identified.

[0116] The key feature points may include, for example, the location points of environmental elements contained in the keyframe image, such as road signs, buildings, slogans, road elements (such as lane lines, pedestrian crossings), etc.

[0117] Secondly, for each keyframe image, based on one or more key feature points in the keyframe image, the keyframe image can be matched with the multiple target localization images.

[0118] For example, for each keyframe image, if the target positioning image contains the key feature points in the keyframe image, then the target positioning image and the keyframe image are determined to be a successful match, and the successfully matched keyframe image and the target positioning image are taken as an image pair.

[0119] Preferably, if the proportion of key feature points contained in the target positioning image to the total number of feature points in the target positioning image is greater than or equal to a preset proportion, it can be determined that the target positioning image and the key frame image are successfully matched.

[0120] In step S205, the target location information is determined based on the image pair.

[0121] In some embodiments, firstly, for each image pair, the relative position information of the keyframe image in the image pair relative to the corresponding target positioning image can be determined.

[0122] Specifically, an affine transformation can be used to map the keyframe image onto the target plane containing the corresponding target localization image. For example, an affine transformation matrix can be used to map the keyframe image onto the target plane of the target localization image. This affine transformation matrix can be, for example, H = findHomography(A, B), where A represents the keyframe image and B represents the target localization image. Then, based on the target plane, the distance difference between the center point of the keyframe image and the center point of the target localization image can be determined. Subsequently, based on this distance difference, the relative position information of the keyframe image in the image pair relative to the corresponding target localization image can be determined.

[0123] In this way, after obtaining the relative position information, we can know the direction and degree of deviation between the keyframe image and the target positioning image on the same plane, which makes it easier to determine the position corresponding to the keyframe image in the subsequent process.

[0124] Secondly, based on the relative position information and the image positioning information corresponding to the target positioning image, the undetermined position corresponding to the keyframe image is determined.

[0125] In this step, the undetermined position corresponding to the keyframe image can be obtained based on the relative position information and the image positioning information corresponding to the target positioning image. When multiple image pairs are included, the undetermined position corresponding to the keyframe image in each image pair can be determined sequentially.

[0126] Finally, the target location information is determined based on the undetermined position corresponding to the keyframe image in each image pair.

[0127] In the case of an image pair, the undetermined position corresponding to the keyframe image can be directly used as the target location information.

[0128] In the case of multiple image pairs, a first average position of the undetermined position corresponding to the keyframe image in the multiple image pairs can be determined, and this first average position can be used as the target localization information.

[0129] Furthermore, considering that the positioning images stored in the preset positioning database may vary in clarity and reliability, a weight can be preset for each positioning image. Accordingly, in this step, when multiple image pairs are included, a preset weight can be determined for the target positioning image in each image pair. Based on the preset weight, a weighted average of multiple locations to be determined is obtained to acquire the target positioning information.

[0130] For example, target location information can be determined using the following formula:

[0131]

[0132] Where P represents the target location information. A preset weight is assigned to each target localization image, where N is the number of image pairs and Pi is the undetermined position corresponding to the keyframe image.

[0133] Using the above method, the second device sends video data and reference positioning information to the first device. The first device can then obtain multiple location feature information based on the reference positioning information, and, based on multiple keyframe images and location feature information in the video data, obtain the target positioning information corresponding to the second device. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0134] Figure 3 This is a flowchart illustrating a positioning method according to an exemplary embodiment, taking as an example that the location feature information includes target map feature information, such as... Figure 3 As shown, the method may include the following steps:

[0135] In step S301, video data sent by the second device and the reference positioning information of the second device are received.

[0136] In step S302, multiple keyframe images are extracted from the video data.

[0137] For example, multiple video frame images included in the video data can be subjected to preset processing to obtain multiple keyframe images. This preset processing includes one or more of denoising, jitter reduction, and deduplication.

[0138] In step S303, based on the reference positioning information, multiple target map feature information are determined from a preset high-precision map database.

[0139] The preset high-precision map database includes multiple map feature information and corresponding map positioning information for each feature. Currently, high-precision maps often contain a large amount of geographic location information (i.e., map positioning information) and map environment feature information (i.e., map features). These map feature information may include, for example, features of buildings, roads, signs, etc. Therefore, in this embodiment, the high-precision map can be used to further determine the target positioning information of the second device, thereby improving positioning accuracy.

[0140] For example, map feature information within a second positioning range can be used as the target map feature information. Here, the second positioning range is the positioning range centered on the location corresponding to the reference positioning information.

[0141] In other words, in this step, target map features near the reference location information can be selected from multiple map features based on the reference location information, which facilitates further determination of the target location information of the second device based on the target map features.

[0142] In step S304, feature extraction is performed on each keyframe image to obtain the image feature information corresponding to the keyframe image.

[0143] For example, features can be extracted from keyframe images using a convolutional neural network to obtain the image feature information corresponding to the keyframe images. This convolutional neural network can be represented as F=CNN(I), where I is the keyframe image.

[0144] In step S305, each image feature information is matched with the plurality of target map feature information, and the successfully matched image feature information and target map feature information are taken as a feature pair.

[0145] In some embodiments, firstly, for each image feature information, the feature similarity between the image feature information and each target map feature information can be determined.

[0146] For example, the similarity function similarity() can be used to calculate the feature similarity between image feature information and each feature information of the target map. Here, the feature information can be a feature vector.

[0147] Then, if the feature similarity is greater than or equal to a preset similarity threshold, it is determined that the image feature information and the target map feature information are successfully matched, and the successfully matched image feature information and the target map feature information are taken as a feature pair.

[0148] In other embodiments, to improve the accuracy of feature pair determination, in one possible implementation, the successfully matched image feature information and the target map feature information can be considered as a pending feature pair. The feature pair is then determined from multiple pending feature pairs based on pre-defined road element features.

[0149] The road element features are used to characterize the geometric shape and / or topological relationships of road elements. Image feature information in undetermined feature pairs can be obtained through road element features. For example, if a road extends in the same direction and conforms to the road alignment, there are no lane lines at intersections, so there should be no continuous line-shaped image feature information. The undetermined feature pairs containing such image feature information can be discarded to improve the accuracy and reliability of the feature pairs.

[0150] In another possible implementation, the step of matching each image feature with the multiple target map features can be performed iteratively multiple times. When the number of iterations reaches a preset number, the successfully matched image feature and the target map feature are considered as a single feature pair.

[0151] In this way, some error feature information can also be filtered out through multiple iterations. When the number of iterations reaches the preset number, the image feature information and the target map feature information of each successfully matched image after the last matching are taken as a feature pair.

[0152] In step S306, the target location information is determined based on the feature pair.

[0153] In some embodiments, map positioning information corresponding to the target map feature information in the feature pair can be determined. Then, the target positioning information is determined based on the map positioning information.

[0154] If a feature pair is included, the map location information can be directly used as the target location information.

[0155] In the case of multiple feature pairs, the second average position corresponding to the map positioning information in the multiple feature pairs can be determined, and the second average position can be used as the target positioning information.

[0156] For example, target location information can be determined using the following formula:

[0157]

[0158] Where P represents the target localization information, and M represents the number of feature pairs. Location information for the map.

[0159] In addition, after obtaining the target location information, in order to further expand the data in the preset location database, the successfully matched keyframe image can be used as a new location image, and the target location information corresponding to the keyframe image can be stored in the preset location database as the image location information corresponding to the new location image.

[0160] Simultaneously, the target map feature information in the preset high-precision map database can be updated based on the successfully matched keyframe images, the changed map data can be identified, and the map data in the successfully matched keyframe images can be used to update the preset high-precision map database to further improve the preset high-precision map database.

[0161] Thus, while improving the convenience and experience of passenger ride-hailing services, the embodiments of this disclosure also enable more user-uploaded video data to participate in the construction of high-precision maps as part of the crowdsourced data, achieving a closed loop of personal crowdsourced data. Under this ride-hailing solution, rapid updates of high-precision maps can be further supported.

[0162] Using the above method, the second device sends video data and reference positioning information to the first device. The first device can then obtain multiple location feature information based on the reference positioning information, and, based on multiple keyframe images and location feature information in the video data, obtain the target positioning information corresponding to the second device. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0163] Figure 4 This is a flowchart illustrating a positioning method according to an exemplary embodiment, applied to a second device, such as... Figure 4 As shown, the method may include the following steps:

[0164] In step S401, video data within the target range of the environment where the second device is located is collected.

[0165] In practical applications, this can be packaged as a standalone app or embedded in a general ride-hailing app. Passengers can access the camera by opening the app and simply use the second device to capture video footage (approximately 1-2 seconds) within a certain range (e.g., 180°) of their current location. For example, a passenger can start at a 90° angle to the left of the road they are facing and shift to the right, capturing video footage for approximately 1-2 seconds. Preferably, when capturing video footage, passengers should try to capture areas containing clear landmarks (e.g., signs, traffic lights, etc.). If the landmarks in the passenger's area are not obvious, video footage containing lane lines and other road-related elements can also be captured. Afterward, the captured video footage (i.e., video data) and the reference positioning information of the second device when capturing the video footage can be uploaded to the first device.

[0166] In step S402, the reference positioning information of the second device when it collects the video data is obtained.

[0167] The reference positioning information is the location information of the passenger when the video data was taken, obtained by the positioning device in the second device. For example, it may be GPS positioning information, Beidou satellite positioning information, etc.

[0168] In step S403, the video data and the reference positioning information are sent to the first device so that the first device can extract multiple key frame images from the video data, obtain multiple location feature information based on the reference positioning information, and determine the target positioning information corresponding to the second device based on the multiple key frame images and the multiple location feature information.

[0169] To improve the data analysis efficiency of the first device and reduce the amount of data transmission, in this embodiment, the second device can also perform redundancy judgment on multiple key frame images contained in the video data, delete key frame images with less feature information, and send the filtered key frame images to the first device, so that the first device can determine the target positioning information corresponding to the second device based on the filtered multiple key frame images and the multiple location feature information.

[0170] Using the above method, the second device sends video data and reference positioning information to the first device. The first device can then obtain multiple location feature information based on the reference positioning information, and, based on multiple keyframe images and location feature information in the video data, obtain the target positioning information corresponding to the second device. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0171] Considering that in practical applications, the number of images contained in a pre-defined positioning database is often limited and cannot fully reflect the images of all roads, while a pre-defined high-precision map database contains a large amount of map feature information for all roads, ensuring data integrity. However, precisely because the number of images in the pre-defined positioning database is limited, matching with positioning images in the pre-defined positioning database is more efficient. Therefore, this embodiment also proposes a positioning method that first matches the positioning images in the pre-defined positioning database; if a match fails, then it matches with the map feature information in the pre-defined high-precision map database.

[0172] Figure 5 This is a flowchart illustrating a positioning method according to an exemplary embodiment, using an example where the first device includes a cloud server, the second device includes a passenger device, and the third device includes a driver device. Figure 5 As shown, the method may include the following steps:

[0173] In step S501, video data sent by the passenger device and the reference positioning information of the passenger device are received.

[0174] In step S502, multiple keyframe images are extracted from the video data.

[0175] In step S503, based on the reference positioning information, multiple keyframe images are matched with multiple positioning images in a preset positioning database.

[0176] If it is determined that multiple keyframe images successfully match multiple positioning images in the preset positioning database, then step S504 is executed.

[0177] If it is determined that multiple keyframe images fail to match multiple positioning images in the preset positioning database, then step S506 is executed.

[0178] In step S504, the successfully matched keyframe image and the target location image are treated as an image pair.

[0179] In step S505, the target location information is determined based on the image pair.

[0180] In step S506, feature extraction is performed on each keyframe image to obtain the image feature information corresponding to the keyframe image.

[0181] In step S507, each image feature is matched with the plurality of target map feature information.

[0182] If it is determined that multiple image feature information is successfully matched with the multiple target map feature information, then step S508 is executed;

[0183] If it is determined that multiple image feature information fails to match the multiple target map feature information, then step S510 is executed.

[0184] In step S508, the successfully matched image feature information and the target map feature information are taken as a feature pair.

[0185] In some embodiments, the successfully matched image feature information and the target map feature information can be considered as a pair of undetermined features. The feature pair is then determined from among multiple pairs of undetermined features based on predefined road element features.

[0186] In other embodiments, the step of matching each image feature with the plurality of target map features can be performed iteratively multiple times. When the number of iterations reaches a preset number, the successfully matched image feature and the target map feature are considered as a single feature pair.

[0187] In step S509, the target location information is determined based on the feature pair.

[0188] In step S510, the reference positioning information is used as the target positioning information.

[0189] If the map feature information fails to match the preset high-precision map data, the reference positioning information can be used as the target positioning information to ensure the smooth operation of subsequent related services.

[0190] In step S511, the target location information is sent to the driver's device.

[0191] Regarding the methods in the above embodiments, the specific manner in which each step is performed has already been described in relevant documents. Figures 1 to 4 The embodiments of the method have been described in detail, and will not be elaborated upon here.

[0192] Using the above method, the passenger device sends video data and reference positioning information to the cloud server. The cloud server can then obtain multiple location feature information based on the reference positioning information, and further determine the target positioning information of the passenger device based on multiple keyframe images and location feature information in the video data. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of passenger device positioning. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, thus enhancing the passenger experience.

[0193] Figure 6 This is a block diagram illustrating a positioning device according to an exemplary embodiment, applied to a first device, which may include a cloud server or a vehicle, such as... Figure 6 As shown, the device 600 includes:

[0194] The receiving module 601 is used to receive video data sent by the second device and the reference positioning information of the second device;

[0195] Extraction module 602 is used to extract multiple keyframe images from the video data;

[0196] The first acquisition module 603 is used to acquire multiple location feature information based on the reference positioning information;

[0197] The positioning module 604 is used to determine the target positioning information corresponding to the second device based on the multiple keyframe images and the multiple location feature information.

[0198] Optionally, the location feature information includes a target location image; the first acquisition module 603 is used to determine multiple target location images from a preset location database based on the reference location information; wherein the preset location database includes multiple location images.

[0199] Optionally, the preset positioning database also includes image positioning information corresponding to each positioning image. The first acquisition module 603 is used to take the positioning image within the first positioning range of the image positioning information as the target positioning image. The first positioning range is the positioning range determined with the position corresponding to the reference positioning information as the center.

[0200] Optionally, the positioning module 604 is used to match each of the keyframe images with the plurality of target positioning images, and to take the successfully matched keyframe images and target positioning images as an image pair; and to determine the target positioning information based on the image pair.

[0201] Optionally, the positioning module 604 is used to identify one or more key feature points in each keyframe image; and for each keyframe image, to match the keyframe image with the plurality of target positioning images based on one or more key feature points in the keyframe image.

[0202] Optionally, the positioning module 604 is configured to determine that the target positioning image and the key frame image are successfully matched if the target positioning image contains key feature points in the key frame image for each key frame image.

[0203] Optionally, the positioning module 604 is configured to, for each image pair, determine the relative position information of the keyframe image in the image pair relative to the corresponding target positioning image; and determine the undetermined position corresponding to the keyframe image based on the relative position information and the image positioning information corresponding to the target positioning image; and determine the target positioning information based on the undetermined position corresponding to the keyframe image in each image pair.

[0204] Optionally, the positioning module 604 is used to map the keyframe image onto the target plane where the corresponding target positioning image is located through affine transformation; based on the target plane, determine the distance difference between the center point of the keyframe image and the center point of the target positioning image; and determine the relative position information of the keyframe image in the image pair relative to the corresponding target positioning image based on the distance difference.

[0205] Optionally, when multiple image pairs are included, the positioning module 604 is used to determine a preset weight corresponding to the target positioning image in each image pair; and to determine a weighted average of multiple undetermined positions based on the preset weight to obtain the target positioning information.

[0206] Optionally, the location feature information includes target map feature information; the first acquisition module 603 is used to determine multiple target map feature information from a preset high-precision map database based on the reference positioning information; wherein, the preset high-precision map database includes multiple map feature information.

[0207] Optionally, the preset high-precision map database also includes map positioning information corresponding to each map feature information. The first acquisition module 603 is used to take the map feature information within the second positioning range of the map positioning information as the target map feature information; wherein, the second positioning range is the positioning range determined with the location corresponding to the reference positioning information as the center.

[0208] Optionally, the positioning module 604 is used to extract features from each keyframe image to obtain image feature information corresponding to the keyframe image; match each image feature information with the multiple target map feature information, and take the successfully matched image feature information and the target map feature information as a feature pair; and determine the target positioning information based on the feature pair.

[0209] Optionally, the positioning module 604 is used to determine the feature similarity between each image feature and each target map feature for each image feature; and to determine that the image feature and the target map feature are successfully matched if the feature similarity is greater than or equal to a preset similarity threshold.

[0210] Optionally, the positioning module 604 is used to treat the successfully matched image feature information and the target map feature information as a pair of undetermined features; and to determine the feature pair from multiple pairs of undetermined features based on pre-defined road element features.

[0211] Optionally, the positioning module 604 is used to perform the step of matching each image feature information with the multiple target map feature information in multiple iterations; when the number of iterations reaches a preset number, the successfully matched image feature information and the target map feature information are taken as a feature pair.

[0212] Optionally, the positioning module 604 is used to determine the map positioning information corresponding to the target map feature information in the feature pair; and to determine the target positioning information based on the map positioning information.

[0213] Optionally, the extraction module 602 is used to perform preset processing on multiple video frame images included in the video data to obtain the multiple key frame images; wherein, the preset processing includes one or more of noise reduction processing, jitter reduction processing and deduplication processing.

[0214] Optionally, the positioning module 604 is also used to send the target positioning information to a third device.

[0215] Using the aforementioned device, the second device sends video data and reference positioning information to the first device. The first device can obtain multiple location feature information based on the reference positioning information, and then obtain the target positioning information corresponding to the second device based on multiple keyframe images and location feature information in the video data. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0216] Figure 7 This is a block diagram illustrating a positioning device according to an exemplary embodiment, applied to a second device, such as... Figure 7 As shown, the device 700 includes:

[0217] Acquisition module 701 is used to acquire video data within the target range of the environment where the second device is located;

[0218] The second acquisition module 702 is used to acquire the reference positioning information of the second device when it collects the video data;

[0219] The sending module 703 is used to send the video data and the reference positioning information to the first device, so that the first device can extract multiple key frame images from the video data, obtain multiple location feature information according to the reference positioning information, and determine the target positioning information corresponding to the second device according to the multiple key frame images and the multiple location feature information.

[0220] Using the aforementioned device, the second device sends video data and reference positioning information to the first device. The first device can obtain multiple location feature information based on the reference positioning information, and then obtain the target positioning information corresponding to the second device based on multiple keyframe images and location feature information in the video data. Through multi-stage matching and confirmation, higher-precision positioning information can be obtained, improving the accuracy of positioning the second device. In subsequent application scenarios, services can be provided to passengers based on more accurate target positioning information, improving the passenger experience.

[0221] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0222] Figure 8 This is a block diagram illustrating an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 may be provided as a cloud server or a vehicle. (Refer to...) Figure 8The electronic device 800 includes a first processor 822, which may be one or more, and a first memory 832 for storing computer programs executable by the first processor 822. The computer program stored in the first memory 832 may include one or more modules, each corresponding to a set of instructions. Furthermore, the first processor 822 may be configured to execute the computer program to perform the positioning method described above.

[0223] Additionally, the electronic device 800 may also include a power supply component 826 and a first communication component 850. The power supply component 826 can be configured to perform power management of the electronic device 800, and the first communication component 850 can be configured to enable communication of the electronic device 800, such as wired or wireless communication. Furthermore, the electronic device 800 may also include a first input / output (I / O) interface 858. The electronic device 800 can operate on an operating system stored in a first memory 832, such as Windows Server™, Mac OS X™, Unix™, Linux™, etc.

[0224] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the positioning method described above. For example, the computer-readable storage medium may be the first memory 832 including the program instructions described above, which may be executed by the first processor 822 of the electronic device 800 to complete the positioning method described above.

[0225] Figure 9 This is a block diagram illustrating an electronic device 900 according to an exemplary embodiment. For example... Figure 9 As shown, the electronic device 900 may include: a second processor 901 and a second memory 902. The electronic device 900 may also include one or more of the following: a multimedia component 903, a second input / output (I / O) interface 904, and a second communication component 905.

[0226] The second processor 901 controls the overall operation of the electronic device 900 to complete all or part of the steps in the positioning method described above. The second memory 902 stores various types of data to support the operation of the electronic device 900. This data may include, for example, instructions for any application or method operating on the electronic device 900, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The second memory 902 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 903 may include a screen and audio components. The screen may be, for example, a touchscreen, and the audio component is used for outputting and / or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the second memory 902 or transmitted via the second communication component 905. The audio component also includes at least one speaker for outputting audio signals. The second input / output interface 904 provides an interface between the second processor 901 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. The second communication component 905 is used for wired or wireless communication between the electronic device 900 and other devices. Wireless communication may include Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or one or more combinations thereof; therefore, the corresponding second communication component 905 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0227] In an exemplary embodiment, the electronic device 900 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the positioning method described above.

[0228] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the positioning method described above. For example, the computer-readable storage medium may be the second memory 902 including the program instructions described above, which may be executed by the second processor 901 of the electronic device 900 to complete the positioning method described above.

[0229] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a processor, wherein the computer program, when executed by the processor, implements the steps of the positioning method described above.

[0230] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a processor, wherein the computer program, when executed by the processor, implements the steps of the positioning method described above.

[0231] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0232] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0233] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A positioning method, characterized in that, Applied to a first device, the method includes: Receive video data sent by the second device and the reference positioning information of the second device; Extract multiple keyframe images from the video data; Based on the reference positioning information, multiple location feature information is obtained; Based on the multiple keyframe images and the multiple location feature information, the target positioning information corresponding to the second device is determined; The location feature information includes a target positioning image and target map feature information. The target positioning image is determined from a preset positioning database based on the reference positioning information, and the preset positioning database includes multiple positioning images. The target map feature information is determined from a preset high-precision map database based on the reference positioning information, and the preset high-precision map database includes multiple map feature information. Determining the target positioning information corresponding to the second device based on the multiple keyframe images and the multiple location feature information includes: Each keyframe image is matched with multiple target positioning images in a preset positioning database to determine the target positioning information corresponding to the second device; If it is determined that each keyframe image fails to match with multiple target positioning images, feature extraction is performed on each keyframe image to obtain image feature information corresponding to the keyframe image, and each image feature information is matched with multiple target map feature information in a preset high-precision map database to determine the target positioning information corresponding to the second device. The method further includes: If multiple image feature information fails to match multiple target map feature information in a preset high-precision map database, the reference positioning information is used as the target positioning information.

2. The method according to claim 1, characterized in that, The location feature information includes a target positioning image; obtaining multiple location feature information based on the reference positioning information includes: Based on the reference positioning information, multiple target positioning images are determined from a preset positioning database; wherein, the preset positioning database includes multiple positioning images.

3. The method according to claim 2, characterized in that, The preset positioning database also includes image positioning information corresponding to each positioning image, and determining multiple target positioning images from the preset positioning database based on the reference positioning information includes: The positioning image within the first positioning range of the image positioning information is taken as the target positioning image; wherein, the first positioning range is the positioning range determined with the position corresponding to the reference positioning information as the center.

4. The method according to claim 2, characterized in that, The step of determining the target positioning information corresponding to the second device based on the plurality of keyframe images and the plurality of position feature information includes: Each keyframe image is matched with the plurality of target positioning images, and the successfully matched keyframe image and target positioning image are taken as an image pair; The target location information is determined based on the image pair.

5. The method according to claim 4, characterized in that, The step of matching each of the keyframe images with the plurality of target localization images includes: Identify one or more key feature points in each of the keyframe images; For each keyframe image, the keyframe image is matched with the plurality of target localization images based on one or more key feature points in the keyframe image.

6. The method according to claim 5, characterized in that, The step of matching each keyframe image with the plurality of target localization images based on one or more key feature points in the keyframe image includes: For each keyframe image, if the target positioning image contains key feature points from the keyframe image, then the target positioning image and the keyframe image are determined to be a successful match.

7. The method according to claim 4, characterized in that, Determining the target location information based on the image pair includes: For each image pair, determine the relative position information of the keyframe image in the image pair relative to the corresponding target positioning image; and determine the undetermined position corresponding to the keyframe image based on the relative position information and the image positioning information corresponding to the target positioning image. The target positioning information is determined based on the undetermined position corresponding to the keyframe image in each image pair.

8. The method according to claim 7, characterized in that, The determination of the relative position information of the keyframe image in the image pair with respect to the corresponding target positioning image includes: The keyframe image is mapped onto the target plane where the corresponding target positioning image is located through affine transformation; Based on the target plane, determine the distance difference between the center point of the keyframe image and the center point of the target positioning image; Based on the distance difference, the relative position information of the keyframe image in the image pair with respect to the corresponding target positioning image is determined.

9. The method according to claim 7, characterized in that, In the case of multiple image pairs, determining the target localization information based on the undetermined position corresponding to the keyframe image in each image pair includes: Determine the preset weight corresponding to the target localization image in each image pair; Based on the preset weights, a weighted average of multiple undetermined locations is determined to obtain the target positioning information.

10. The method according to claim 1, characterized in that, The location feature information includes target map feature information; The step of obtaining multiple location feature information based on the reference positioning information includes: Based on the reference positioning information, multiple target map feature information is determined from a preset high-precision map database; wherein, the preset high-precision map database includes multiple map feature information.

11. The method according to claim 10, characterized in that, The preset high-precision map database also includes map positioning information corresponding to each of the map feature information, and determining multiple target map feature information from the preset high-precision map database based on the reference positioning information includes: The map feature information within the second positioning range of the map positioning information is used as the target map feature information; wherein, the second positioning range is the positioning range determined with the location corresponding to the reference positioning information as the center.

12. The method according to claim 10, characterized in that, The step of determining the target positioning information corresponding to the second device based on the plurality of keyframe images and the plurality of position feature information includes: Feature extraction is performed on each of the keyframe images to obtain the image feature information corresponding to the keyframe image; Each of the image feature information is matched with the plurality of target map feature information, and the successfully matched image feature information and target map feature information are taken as a feature pair; The target location information is determined based on the feature pairs.

13. The method according to claim 12, characterized in that, The step of matching each of the image feature information with the plurality of target map feature information includes: For each of the image feature information, determine the feature similarity between the image feature information and each of the target map feature information; If the feature similarity is greater than or equal to a preset similarity threshold, it is determined that the image feature information and the target map feature information are successfully matched.

14. The method according to claim 12, characterized in that, The step of combining the successfully matched target map feature information as a feature pair includes: The successfully matched image feature information and the target map feature information are treated as a pair of undetermined features; Based on pre-defined road element features, the feature pair is determined from a plurality of undetermined feature pairs.

15. The method according to claim 12, characterized in that, The step of combining the successfully matched image feature information and the target map feature information as a feature pair includes: The step of matching each image feature with the plurality of target map feature information is performed iteratively multiple times. If the number of iterations reaches a preset number, the successfully matched image feature information and the target map feature information are treated as a feature pair.

16. The method according to claim 12, characterized in that, Determining the target location information based on the feature pair includes: Determine the map positioning information corresponding to the target map feature information in the feature pair; The target location information is determined based on the map location information.

17. The method according to claim 1, characterized in that, The extraction of multiple keyframe images from the video data includes: The multiple video frame images included in the video data are subjected to preset processing to obtain the multiple key frame images; The preset processing includes one or more of noise reduction processing, jitter reduction processing, and deduplication processing.

18. The method according to any one of claims 1 to 17, characterized in that, The method further includes: The target location information is sent to a third device.

19. A positioning method, characterized in that, Applied to a second device, the method includes: Collect video data within the target area of ​​the environment where the second device is located; Obtain the reference positioning information of the second device when it collects the video data; The video data and the reference positioning information are sent to a first device, so that the first device can extract multiple keyframe images from the video data, obtain multiple location feature information based on the reference positioning information, and determine the target positioning information corresponding to the second device based on the multiple keyframe images and the multiple location feature information; the location feature information includes a target positioning image and target map feature information, wherein the target positioning image is determined from a preset positioning database based on the reference positioning information, and the preset positioning database includes multiple positioning images; the target map feature information is determined from a preset high-precision map database based on the reference positioning information, and the preset high-precision map database includes multiple map feature information; the multiple keyframe images and The determination of the target positioning information corresponding to the second device using the multiple location feature information includes: matching each keyframe image with multiple target positioning images in a preset positioning database to determine the target positioning information corresponding to the second device; if it is determined that each keyframe image fails to match with multiple target positioning images, performing feature extraction on each keyframe image to obtain image feature information corresponding to the keyframe image, and matching each image feature information with multiple target map feature information in a preset high-precision map database to determine the target positioning information corresponding to the second device; if it is determined that multiple image feature information fails to match with multiple target map feature information in the preset high-precision map database, using the reference positioning information as the target positioning information.

20. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 18, or the steps of the method according to claim 19.

21. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1 to 18, or the steps of the method according to claim 19.

22. A cloud server, characterized in that, include: The first memory, on which the computer program is stored; A first processor is configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 18.

23. An electronic device, characterized in that, include: The second memory stores the computer program. A second processor is configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 19.

Images